Action Research on the implementation of Writing Approaches to improve Academic Writing Skills of Namibian Foundation Programme students. By Karoline du Plessis Submitted in accordance with the requirements for the degree of MASTER OF ARTS WITH SPECIALISATION IN TESOL (TEACHING ENGLISH TO SPEAKERS OF OTHER LANGUAGES) at the UNIVERSITY OF SOUTH AFRICA SUPERVISOR: Prof. Brenda Spencer January 2012 i DECLARATION Student number: 31246524 I declare that An Action Research Study on the implementation of Writing Approaches to improve Academic Writing Skills of Namibian Foundation Programme students is my own work and that all the sources that I have used or quoted have been indicated and acknowledged by means of complete references. …………………………………………………….. ………………………………… Signature Date (Ms K. du Plessis) ii DECLARATION FROM STATISTICIAN Email: famecc@live.com or 3, Fouche Street, Windhoek West, P. O. Box 1382, Ondangwa, Namibia. info@fameconsultancy.com Tel: +264 81 27 82470. www.fameconsultancy.com Fame! Consultancy January 4th 2012. To Whom It May Concern Re: Declaration of support rendered to Mrs. Karoline du Plessis I, Anthony Muganza, hereby declare that I supported Mrs. Karoline du Plessis with her studies at the University of South Africa (UNISA) towards a Master’s degree in TESOL. I personally entered the data collected for her research project, cleaned and verified it, ran frequencies and performed relevant tests of significance using the Statistical Package for Social Sciences (SPSS Version 18) and presented the results in an analysis report. The results as presented by me in the analysis report which the student interpreted and discussed on her own are valid and a true reflection of data collected in respect of the research conducted. Signed, Anthony Muganza Research and data management consultant iii ACKNOWLEDGEMENTS My acknowledgements must firstly go to my supervisor, Prof. Brenda Spencer, for her continuous support, advice, encouragement and suggestions to guide me to completion of this dissertation. Secondly, my subjects need appreciation for their willingness to participate. I also want to thank my Foundation Programme colleagues at UNAM Oshakati Campus for their support and assistance. Mr. Anthony Muganza needs to be thanked for his help with the statistical analysis. I also want to give my appreciation to the library staff at UNAM Oshakati campus and UNISA library for their tremendous efforts in finding books and journal articles for me. Lastly, a big thank you goes to my family, who believed in me all the way. iv TABLE OF CONTENTS Action Research on the implementation of Writing Approaches to improve Academic Writing Skills of Namibian Foundation Programme students......................................................................................... i DECLARATION .................................................................................................................................... ii DECLARATION FROM STATISTICIAN ........................................................................................... iii ACKNOWLEDGEMENTS ................................................................................................................... iv TABLE OF CONTENTS ........................................................................................................................ v LIST OF TABLES ................................................................................................................................. ix LIST OF FIGURES ............................................................................................................................... xi ABSTRACT.......................................................................................................................................... xii CHAPTER 1: INTRODUCTION ........................................................................................................... 1 1.1 INTRODUCTION .................................................................................................................. 1 1.2 BACKGROUND TO STUDY ............................................................................................... 2 1.2.1 Process approach ............................................................................................................. 4 1.2.2 Modeling/Imitation Approach......................................................................................... 5 1.2.3 Combination of process and modeling approaches: Process genre approach ................. 7 1.2.4 The Process and Modeling approaches in the English curriculum in Namibia............... 9 1.2.5 Comments from five Namibian teachers....................................................................... 10 1.3 THE RESEARCH CONTEXT OF THIS STUDY ............................................................... 12 1.3.1 Student and campus background................................................................................... 14 1.3.2 Learning outcomes of the English Foundation Course ................................................. 15 1.3.3 Grading Criteria for Essays ........................................................................................... 15 1.3.4 The situation of the researcher ...................................................................................... 16 RESEARCH PROBLEM ..................................................................................................... 17 1.4. 1.4.1 The proposed interventions ........................................................................................... 19 1.5 AIM OF RESEARCH ........................................................................................................... 21 1.6 HYPOTHESES ..................................................................................................................... 22 1.7 RESEARCH QUESTIONS................................................................................................... 22 v 1.8 DEFINITION OF TERMS.................................................................................................... 24 1.9 ABBREVIATIONS .............................................................................................................. 25 1.10 CHAPTER OVERVIEWS .................................................................................................... 25 1.11 CONCLUSION ..................................................................................................................... 26 CHAPTER 2: LITERATURE REVIEW .............................................................................................. 27 2.1 INTRODUCTION ................................................................................................................ 27 2.2 THE “MYSTERY” OF ACADEMIC WRITING SKILLS .................................................. 27 2.2.1 2.3 Academic Writing ......................................................................................................... 27 PROCESS APPROACH ....................................................................................................... 30 2.3.1 What is the Process approach? ...................................................................................... 30 2.3.2 Features of the process approach ................................................................................. 31 2.3.3 Arguments against the process approach ...................................................................... 38 2.3.4 Research: process approach in Application .................................................................. 40 2.4 MODELING/IMITATION APPROACH ............................................................................. 47 2.4.1 What is the Modeling/imitation approach? ................................................................... 47 2.4.2 Features of the modeling/imitation approach ................................................................ 49 2.4.3 Arguments against the modeling/imitation approach ................................................... 55 2.4.4 Research: modeling/imitation approach in Application ................................................ 57 2.5 PROCESS GENRE APPROACH ......................................................................................... 62 2.5.1 What is the process genre approach? ............................................................................ 62 2.5.2 Features of the process genre approach ........................................................................ 66 2.5.3 Arguments against the Process genre approach ............................................................ 72 2.5.4 Research: Process genre approach in Application ........................................................ 73 2.6 CONCLUSION ..................................................................................................................... 75 CHAPTER 3: RESEARCH METHODOLOGY .................................................................................. 76 3.1 INTRODUCTION ................................................................................................................ 76 3.2 RESEARCH DESIGN .......................................................................................................... 77 vi 3.2.1 Action research: theory and justification ...................................................................... 77 3.2.2 Justification of the hybrid design of this study ............................................................. 82 3.3. METHODOLOGY ............................................................................................................... 82 3.3.1 Research Instruments: Data collection techniques in relation to Qualitative and/or Quantitative research design ......................................................................................................... 83 3.3.2 Data ............................................................................................................................... 92 3.3.3 Analysis......................................................................................................................... 97 3.4 LIMITATIONS ..................................................................................................................... 99 3.5 ETHICAL CONSIDERATIONS ........................................................................................ 100 3.6 CONCLUSION ................................................................................................................... 100 CHAPTER 4: RESULTS .................................................................................................................... 101 4.1 INTRODUCTION .............................................................................................................. 101 4.2 THE INTERVENTIONS .................................................................................................... 103 4.3 REFLECTION REVIEWS ................................................................................................. 103 4.3.1 Review 2008: process approach .................................................................................. 103 4.3.2 Review 2008: modeling/imitation approach ............................................................... 108 4.3.3 Review 2009: process genre approach ........................................................................ 111 4.4 CONTEXT OF THE STUDY ............................................................................................. 116 4.4.1 Overall Assessments .......................................................................................................... 117 4.5 ESSAYS .............................................................................................................................. 121 4.5.1 Comparison of overall pre-intervention and post-intervention essay results ..................... 121 4.5.2 Essay Class 1: process approach – 2008: .................................................................... 124 4.5.3 Essay Class 2: modeling/imitation approach – 2008: ................................................. 127 4.5.4 Essay Class 1 and 2: process genre approach – 2009: ................................................ 130 4.5.5 Comparison of the essay results of the three approaches: 2008 and 2009: ........................ 132 4.6 LABORATORY REPORT RESULTS ............................................................................. 133 4.6.1 Overview of how the approaches were used to teach the laboratory report writing unit in the English lessons: ................................................................................................................. 133 vii 4.6.2 PHYSICS ........................................................................................................................... 136 4.6.3 CHEMISTRY .................................................................................................................... 143 4.6.4 BIOLOGY.......................................................................................................................... 150 CLOSED-QUESTION QUESTIONNIARE RESULTS ................................................... 158 4.7 4.7.1 Comparison of the pre- and post-intervention closed-question questionnaire results: 159 4.8 OPEN-QUESTION QUESTIONNAIRE RESULTS ........................................................ 166 4.9 INTERVIEW RESULTS: 2009 .......................................................................................... 176 4.10. CONCLUSION ............................................................................................................... 184 CHAPTER 5: CONCLUSION............................................................................................................ 186 5.1 SUMMARY OF FINDINGS .............................................................................................. 186 5.2 CONCLUSIONS................................................................................................................. 189 5.3 DISCUSSION OF PROBLEMS ......................................................................................... 194 5.4 CONTRIBUTIONS OF THE STUDY ............................................................................... 196 5.5 SUGGESTIONS FOR FURTHER RESEARCH................................................................ 198 5.6 CONCLUDING STATEMENT ......................................................................................... 198 REFERENCES ................................................................................................................................... 199 APPENDICES .................................................................................................................................... 211 Appendix 1:..................................................................................................................................... 211 Appendix 2:..................................................................................................................................... 215 Appendix 3:..................................................................................................................................... 217 Appendix 4:..................................................................................................................................... 218 Appendix 5:..................................................................................................................................... 220 Appendix 6:..................................................................................................................................... 222 Appendix 7:..................................................................................................................................... 226 viii LIST OF TABLES Table 1: Pre- and post-intervention results of essays and laboratory reports compared ............ 118 Table 2: Independent Samples t-test of pre- and post-intervention essay and laboratory report results .......................................................................................................................................... 119 Table 3: Pre- and post-intervention overall essay mean score .................................................... 121 Table 4: Independent Samples t-test: for pre- and post-intervention overall essay results ........ 122 Table 5: Pre- and post-intervention Class 1 essay mean scores.................................................. 124 Table 6: Independent Samples t-test for significant difference in Class 1 essay mean scores ... 124 Table 7: Pre- and post-intervention Class 2 essay mean scores.................................................. 127 Table 8: Independent Samples t-test for significant difference in Class 2 essay mean scores ... 127 Table 9: Pre- and post-intervention 2009 essay mean scores ..................................................... 130 Table 10: Independent Samples t-test for significant difference in 2009 essay mean scores ..... 130 Table 11: Pre- and post-intervention Class 1 2008 laboratory report mean scores .................... 136 Table 12: Independent Samples t-test for significant difference in Class 1 2008 Physics laboratory report mean scores ..................................................................................................... 136 Table 13: Pre- and post-intervention Class 2 Physics laboratory report mean scores ................ 138 Table 14: Independent Samples t-test for significant difference in Class 2 2008 Physics laboratory report mean scores ..................................................................................................... 138 Table 15: Pre- and post-intervention 2009 Physics laboratory reports mean scores .................. 140 Table 16: Independent Samples t-test for significance difference in 2009 Physics laboratory report mean scores ...................................................................................................................... 140 Table 17: Pre- and post-intervention Class 1 2008 Chemistry laboratory report mean scores... 143 Table 18: Independent Samples t-test for significant difference in Class 1 2008 Chemistry laboratory report mean scores ..................................................................................................... 143 Table 19: Pre- and post-intervention Class 2 2008 Chemistry laboratory report mean scores... 145 Table 20: Independent Samples t-test for significant difference in Class 2 2008 Chemistry laboratory report mean scores ..................................................................................................... 145 Table 21: Pre- and post-intervention 2009 Chemistry laboratory report mean scores ............... 147 Table 22: Independent Samples t-test for significant difference in 2009 Chemistry laboratory report mean scores ...................................................................................................................... 147 ix Table 23: Pre- and post-intervention Class 1 Biology laboratory report mean scores ............... 150 Table 24: Independent Samples t-test for significant difference in Class 1 2008 Biology laboratory report mean scores ..................................................................................................... 150 Table 25: Pre- and post-intervention Class 2 2008 Biology laboratory report mean scores ...... 152 Table 26: Independent Samples t-test for significant difference in Class 2 Biology laboratory report mean scores ...................................................................................................................... 152 Table 27: Pre- and post-intervention 2009 Biology laboratory report mean scores ................... 154 Table 28: Independent Samples t-test for significant difference in 2009 Biology laboratory report mean scores ...................................................................................................................... 154 Table 29: Mean score differences of process approach results................................................... 156 Table 30: Mean score differences of modeling approach results ............................................... 157 Table 31: Mean score differences of process genre approach results ......................................... 157 Table 32: Closed-question questionnaire responses: attitude ..................................................... 160 Table 33: Closed-question questionnaire responses: writing habits ........................................... 162 Table 34: Closed-question questionnaire 2009 responses: materials ......................................... 165 Table 35: Closed-question questionnaire 2009 responses: values and purposes of writing ....... 166 Table 36: Open-question questionnaire responses: have skills improved? ................................ 174 Table 37: Open-question questionnaire responses: what skills improved? ................................ 176 x LIST OF FIGURES Figure 1: Overall context of the study ........................................................................................ 117 Figure 2: Bar graph: pre- and post-intervention results of essays and laboratory reports .......... 120 Figure 3: 2008 and 2009 pre- and post-intervention essay result comparison ........................... 123 Figure 4: Comparison of Class 1 (2008) pre- and post-intervention essay results ..................... 125 Figure 5: Pre- and post-intervention Class 2 (2008) essay results .............................................. 128 Figure 6: 2009 Pre- and post-intervention essay results ............................................................. 131 Figure 7: Pre- and post-intervention Class 1 2008 Physics laboratory report results ................. 137 Figure 8: Pre- and post-intervention Class 2 2008 Physics laboratory report results ................. 139 Figure 9: Pre- and post-intervention 2009 Physics laboratory report results.............................. 141 Figure 10: Pre-and post-intervention Class 1 2008 Chemistry laboratory report results ........... 144 Figure 11: Pre- and post-intervention Class 2 2008 Chemistry laboratory report results .......... 146 Figure 12: Pre- and post-intervention 2009 Chemistry laboratory report results ....................... 148 Figure 13: Pre- and post-intervention Class 1 2008 Biology laboratory report results .............. 151 Figure 14: Pre- and post-intervention Class 2 2008 Biology laboratory report results .............. 153 Figure 15: Pre-and post-intervention 2009 Biology laboratory report results ............................ 155 Figure 16: Types of essays written at school compared to Foundation Programme .................. 167 Figure 17: Different amounts of time spent on writing essays: pre-intervention compared to post-intervention ......................................................................................................................... 171 Figure 18: Pre- and post-intervention responses: have skills improved? ................................... 173 Figure 19: Pre- and post-intervention comparison: which aspects improved most? .................. 175 Figure 20: Challenging areas in writing...................................................................................... 180 xi ABSTRACT Foundation Programme (FP) students at the University of Namibia (UNAM) Oshakati Campus display inadequate academic writing abilities. As their aim is to gain admittance to UNAM main campus science-related courses, it is vital to have effective academic writing skills. This action research (AR) study is a comparison of three writing programmes, the process approach, the modeling approach, and the process genre approach which were implemented separately to three different class groups in 2008 and 2009 to improve the writing skills of students and the teaching practice of the researcher. The effects of the interventions were examined using a combination of the quantitative and qualitative research methods. Data were collected using questionnaires, pre- and post-intervention essays and laboratory reports and interviews. The findings indicate that all three approaches improved the academic writing skills of FP students. The process genre approach had a higher rate of effect than the other two approaches. Key terms: Foundation Programme (FP) students; academic writing; process approach; modeling approach; process genre approach; focused instruction; academic essays; laboratory reports. xii CHAPTER 1: INTRODUCTION 1.1 INTRODUCTION No single description would ever be adequate to describe writing processes, nor would any prescription be adequate to create good writing. (Liebman-Kleine 1986: 785) The focus of this study is on the English academic writing competence of Foundation Programme (FP) students and the researcher’s teaching practice at the Oshakati Campus of the University of Namibia (UNAM). The researcher (I) identified a problem with regard to students’ academic writing competence. This problem was identified in students’ essay answers in the non-standardized entry tests and students’ laboratory report writing skills. In order to try and improve the weak academic writing competence of the students, I decided to implement two different writing approaches (process approach and modeling/imitation approach) in 2008 in two different classes. These two approaches were chosen because I realized students needed guidance when writing essays, but I was not sure how to assist them effectively. In my search to find effective ways to help students write more effective essays I found information on the process approach and the modeling/imitation approach. I felt that the teaching and learning context was suitable to apply the methods and that the learners might benefit from the implementation of the approaches in their writing programme. In order to measure the outcomes of the implementation I decided to carry out action research to determine the effectiveness of the interventions. Action research is seeking information to gauge the implications of change and to reflect on practice and is therefore appropriate to improve a teaching situation (Bell 2005: 9). It is also cyclical in nature with observation, planning, action and reflection as parts of a cycle. The cycle can be repeated with a revised plan, action, and reflection (Blaxter, Hughes & Tight 2006: 71). After the 2008 study, evaluation of the outcomes of the study resulted in the application of another approach in 2009, the process genre approach, and the outcomes of this intervention were also reflected and reported on. The process genre approach was selected as it is a combination of the effective characteristics of the process and the modeling/imitation approach. I wanted to determine if the process genre approach would yield better results than the process approach and the modeling imitation approach in isolation. This study reports on the evaluation of my teaching practice designed to improve English academic writing skills of FP students and an assessment of three different writing 1 interventions, two applied in 2008 and the third in 2009. In 2008, I applied the process approach in Class 1 and in Class 2 the modeling approach was used. In 2009, I applied the process genre approach in both classes. Convenience sampling was applied as I used the existing classes of 2008 and 2009 as my population and sample since, for ethical considerations, I could not exclude any student from the implementations. However, convenience sampling might have an impact on the transferability and generalisability of this study. South African researchers Jackson, Meyer and Parkinson (2006) observed that students in South Africa who enter the university want and need to become part of a certain discourse community which has certain “literacy practices” as component of the community. However, students who want to pursue academic study in their additional language, English, are not necessarily equipped to do so effectively (Phillips 2004: 4; Heffernan 2006). At the main campus of the University of Namibia (UNAM) in Windhoek and at the Oshakati Campus, a similar situation to that of South African students exists with regard to a lack of writing proficiency with students having to use their first year of a four-year degree course to improve, among other study skills and subjects, their English competencies. Even though there is no specific empirical evidence on the weakness of Namibian students’ academic writing skills, lecturers at UNAM main campus are frustrated by students’ English writing abilities as one former lecturer mentioned, ‘I have lectured to these students for many years, the cream of the Namibian senior secondary school leavers, and can testify to the poor English writing proficiency of the majority’ (Tötemeyer 2009: 3). What are the deficiencies in students’ writing skills? 1.2 BACKGROUND TO STUDY …, the issue becomes how teachers can help students express themselves freely and fluently to be more autonomous writers, and how teachers can help students become more successful readers and writers of academic and workplace texts. (Kim & Kim 1 2005: [2]) The entry requirements at UNAM (C in English, C in Mathematics and at least 25 points in 5 subjects including English and Mathematics) lead lecturers to assume that successful 1 The page number is given in these brackets [] to refer to a page of a document without original numbers, usually html documents. I numbered the document pages, to ease the process of finding quotations in the documents. This procedure is followed throughout the dissertation. 2 applicants are able to write meaningful, coherent academic essays, as related to specific learning outcomes and assessment criteria. These essays are supposed to display students’ ability in presenting content that shows that a concept or an idea is analysed critically, there is comprehensive and accurate subject matter and relevant reading is incorporated (Murray & Johanson, 1990: 22). The structure should show that the content is well-organised and wellpresented, there is systematic and concise development of the argument and the introduction and conclusion are “well-thought out” (Murray & Johanson, 1990: 22). In terms of language and vocabulary, students are expected to have high standards of spelling and punctuation which means errors are rare; grammar issues like tenses and subject-verb agreement should show advanced skills in English writing. Students are also expected to use a style of language that is suitable for the purpose, context and audience of the piece of writing (Murray & Johanson, 1990: 22). The marking grid providing these criteria is presented in Appendix 2. Jackson, Meyer and Parkinson (1990) reported on the problem that many ESL students attending universities and colleges in South Africa do not have the necessary skills to communicate effectively using an appropriate academic register and might struggle with writing laboratory reports. Nordin, Halib and Ghazali (2010: 46) confirmed this view by citing socio-economic and political reasons, as well as overcrowded classrooms, teachers’ heavy workloads and insufficient reading practice as issues contributing to language proficiency problems. Similarly, UNAM FP students get below 60 % as related to the learning outcomes above and the marking grids in Appendix 2 and 3, in their writing activities. In fact, despite many attempts at rectifying the social and educational situation, one main impediment in the South African and Namibian students’ ambition to study for their chosen careers is still their weak writing ability (Baba 2006: iii; Reid 1984: 450; Spack 1988). Yeld (2003), and Jackson, Meyer, and Parkinson (2006) indicated that the level of academic literacy of students from various educational backgrounds is unsatisfactory as “lecturers complain about the students’ lack of expertise in writing coherently and correctly…” (Jackson, Meyer & Parkinson 2006: 266). In addition, Uys, Van der Walt, Van den Berg, and Botha (2007: 69) reported on the issue by stating: “a lack of attention to the teaching of functional language skills may be considered a raison d’être for learners’ lack of academic achievement”. This is one of the reasons I decided to conduct a study which would determine whether writing interventions would aid in improving the academic writing skills of students on the Foundation Programme (FP). My decision to find ways to improve the writing skills of the FP students compelled me to find information on different writing 3 approaches and to adapt my teaching method to incorporate the techniques and strategies of the writing approaches. Based on the strengths outlined below, the two approaches that I initially chose in 2007 were the process approach and the modeling/imitation approach. In 2008, the process approach was applied in Class 1 and the modeling/imitation approach was implemented in Class 2. 1.2.1 Process approach The value of the process approach, in teaching and learning writing, lies in its application. A writing programme based on the process approach is characterized by a range of classroom activities that are learner-centered and teacher-facilitated. These activities include elements of writing that function “simultaneously, recursively, and/or in a linear fashion” (Pianko 1979: 5) depending on the students’ preferences and the purpose for the writing activity. Kim and Kim (2005: [3]) commented on the interactive nature of these activities that can at any time or stage in the composing process be “reviewed, evaluated, and revised, even before any text has been produced at all.” Initial and final meaning is constructed by making use of one, two or more composing steps, followed and/or interrupted by review, evaluation and revision of the text. These three techniques are present in a number of steps or activities proposed by the process approach. There are variations of sets of activities and the interpretations of the steps and stages do not necessarily follow a linear step-by-step mode (Chan 1988; Geyser 1996; Kim 2005; Perl 1980; Pianko 1979; Zamel 1982). The main activities are divided into three sections: Pre-writing, writing and reviewing (Kim 2005: [3]). The pre-writing phase is often called the planning stage and includes: analyzing the topic; generating ideas, and organizing ideas (Kim 2005: [3]). Geyser (1996: 222) proposed that the writing phase involves organizing and developing ideas into paragraphs, “with the aim of developing the topic to its fullest”. At this stage, not much attention is given to grammatical issues. The content is the most important aspect and needs to be reviewed, evaluated and revised if necessary, to ensure that meaning is created. This is a stage which requires sufficient time and preferably an environment suitable for writing and thinking, which is not always the English classroom (Perl 1980: 363). The language aspect of the text written by the students can also be reviewed in three different ways or by combining two or three techniques. Students can review and edit their work independently and individually based on the criteria provided by the teacher. Peer-editing can be done, during which students help each other with the review and editing task (also using criteria provided by the teacher), or the teacher 4 can have individual consultations with the learners (Chan 1988: 85, Geyser 1996: 233). The students need to re-work the draft based on the editing before handing in a final draft. Despite the incontestable benefits resulting from the characteristics and application of the process approach as explained above, some drawbacks have to be briefly mentioned here: 1. Students do not always use the same process when writing a text; 2. There is not much information on different types of texts and reasons why certain texts are produced (genres, purpose and audience); 3. There is too little focus on rhetorical conventions and linguistic issues (style and grammar) (Kim & Kim 2005: [4]). 4. Students are not adequately prepared for the examination essay because of the time limit. (Horowitz 1986: 41-42; Caudery 1995: [1]) 1.2.2 Modeling/Imitation Approach The writing approach that was administered to Class 2 in 2008 was the modeling/imitation approach. I use the two terms together because of their compatibility but also their slight difference in my view. Modeling means two things to me: firstly, using model genre essays as examples, and secondly, facilitators actively and practically, illustrating to students how the writing steps and techniques are done. Imitation means when students try to imitate the form, the style and the language used in a genre essay that was provided as a model. In a sequential way, it can be seen that modeling happens before imitation in the writing process, they follow each other. Humans have always had a tendency to imitate other people in various sectors of their lives. This inclination to follow in somebody else’s footsteps has been ensconced in the educational area for hundreds of years. Eminent proponents in the subject of imitation include Quintilian, Aristotle, Plato, Piaget (Butler 2002: 27, 30, 31). These scholars realized that the basic educational principle underscoring this approach is that students learn better when they have an example. A number of researchers recognized that modeling and/or imitation can provide assistance in writing essays (Badger & White 2000, Cope & Kalantzis 1993, Kim & Kim 2005, Farmer & Arrington 1993). In order to avoid the ‘hit-and-miss’ attempts among writing students, the modeling/imitation approach is useful in creating an awareness that texts differ from each other in terms of content, structure, language and style based on the purpose 5 and audience of the text. This awareness is created when students receive examples of the same genre of essays they are expected to write. These examples are obtained from a variety of sources, for instance authentic materials like newspapers, journals, text books, and magazines, or from English Second Language (ESL) books like Great Essays by Folse, Muchmore-Vokoun, Solomon (1999), which have examples of academic essays that are to be dealt with in the writing lessons. This hypothesis and strategy is strengthened by research concerning imitation as a tool for learning to write. Farmer and Arrington (1993: 12-29) and Killgallon (2010: 2) commented positively on the imitation of style linked to genre-writing. In addition, the heuristic principle inherent in learner-centered teaching can be exploited by imitation and modeling when students get the opportunity to find things out for themselves without having to be expressly told. This can be done with regard to generating ideas and making meaning, using appropriate styles for different genres as well as the type of content initiated. According to Farmer and Arrington (1993: 18) there is a distinct link between stylistic and structural imitation and the invention process which determines the content of the text. One could say that the students’ texts are analogues of the model text. The following quotation captures the essence of imitation as a tool for generating ideas and, potentially, effective writing: … how imitation helps generate ideas and insights comes from Barret Mandel. Believing that the conscious mind inhibits or even precludes the kind of “fresh thinking” necessary for good writing, Mandel argues that activities such as copying, imitating, and parodying can elicit discoveries much in the same way as free-writing is thought to do – through bypassing the conscious mind. (Farmer & Arrington 1993: 19) There are a number of strategies that can be followed when adopting the modeling/imitation approach, which will be fully described in the Literature Review. Important to make clear is the fact that students should imitate the facilitators’ methods and not their words when the facilitators model their writing process in a class situation (Farmer & Arrington 1993: 20). Another justification for using the modeling/imitation approach to teach writing is explained by Falk (1979: 439) when she claims that professional writers acknowledge the “influence of other writers” on their own work. All writing is, in effect, inter-textual, and this should be acknowledged. 6 Arguments against imitation as a tool for writing are that imitation restrains the writer and that creativity, originality, and sincerity are reduced (Farmer & Arrington 1993: 23). It seems that the intellectual challenge of writing is reduced by the use of models, since it is fairly easy to follow a rule or a formula in the form of a model. Tourney ([n.d.]: 3) even called modeling and imitation “cheap, facile, bad and mere” (Farmer & Arrington 1993: 23). Doubts also arise about how much of the models that serve as input is actually taken in by writing students: “how far can study and analysis of these products strengthen the students’ understanding of how good writing is actually made, let alone help them to produce some for themselves?” (Watson 1982: 6). The modeling/imitation approach will be critiqued further in the Literature Review. 1.2.3 Combination of process and modeling approaches: Process genre approach Because of weaknesses inherent in both approaches and limitations of a single approach, in 2009, a combination of these two approaches was implemented in Class 1 and 2. The application of the combination of the process and modeling/imitation approaches is, in my view, very similar to the process genre approach and the justifications for the process genre approach coincide with those advocating the combination of the process and modeling approaches. Literature indicates that when some principles and features of the process and modeling/imitation approaches are merged, the resulting approach is called the process genre approach (Badger & White 2000, Watson 1982, Heffernan 2006, Horowitz 1986). Watson (1982: 12) made a very strong case for subsuming the modeling approach in the process approach: The model has thus in a sense been demoted by being brought within the process. But surely this is entirely desirable. If students can treat the model as a resource rather than an ideal, if they can explore it with each other as well as with the teacher, if they can comfortably compare their own products at various stages of composition with that of the professional, then the alien product is truly involving them in the original process. The process genre approach (as an outcome of the combination of the product and the process approach) will be discussed in greater detail in the Literature Review. Kim and Kim (2005: [5]) made the observation that the process genre approach is dualviewed with a focus on writing and the “development of writing”. The following elements belonging to the process approach are present in the process genre approach: 7  the “input to which learners respond” (Kim & Kim 2005: [6]), which relates to the ideas and the content of the intended text,  the techniques and strategies students can use to arrive at a meaningful, comprehensive essay in terms of content,  the continuous assistance from the teacher/facilitator. From the model/genre approach “the context in which writing happens”, a model or an example, suitable language for the specific genre, purpose (Kim & Kim 2005: [6]), and audience are utilized. Kim and Kim (2005: [7]) also suggested “scaffolding” the activities, which means writing students move from the known to the unknown in small steps and thereby learn new skills, concepts and reach “new levels of understanding”. This links in with one of the principles of the process approach as explained by Badger and White (in Yan 2005: 19): students’ background knowledge contributes to their writing skill. Badger and White (2000: 157) provided a very useful description of the synthesis of the process and genre approach. The essential idea here is that the writing class recognizes that: Writing involves knowledge about language (as in product and genre approaches), knowledge of the context in which writing happens and especially the purpose for the writing (as in genre approaches), and skills in using language (as in process approaches) Writing development happens by drawing out the learners’ potential (as in process approaches) and by providing input to which the learners respond (as in product and genre approaches). The value of combining the process and modeling approaches in the process genre approach as an instrument in the writing class lies in the fact that the two approaches are different but complement each other. Newfields (1999: 47) stated: “not only are both approaches compatible in some respects, they may actually be complementary”. Furthermore, the focus on the process and the emphasis on form in the process genre approach make it a potentially useful approach for beginning writers. If the relevant features of the approaches are employed effectively based on the level of the writing skills of the students, it could “help 8 students use their individual writing processes to construct a text in a familiar genre” (Yan 2005: 22). 1.2.4 The Process and Modeling approaches in the English curriculum in Namibia Apart from a multitude of reasons that may account for weak writing skills of students (which fall outside the scope of this study), it is my view that the writing curriculum in Namibian schools contributes to the inadequacy of students’ writing skills. The weak English results as well as comments on the lack of writing skills indicated by a number of Namibian researchers like Benjamin (2004: 7), Nyathi (2001: 9), Willemse (2005: 1, 9) and Wolfaardt (2005: 2360) attest to students having inadequate writing skills. The starting point to determine why students are unable to live up to the writing expectations, is to look at what academic writing conventions are dealt with in senior secondary level syllabi. While this is not in the scope of this dissertation, a brief look at the syllabus explains the learning outcomes students are supposed to have mastered before entering tertiary education. In addition, five secondary school English teachers agreed to provide information on the application of the two approaches or elements of the two approaches in the actual teaching setting, since no information is available in the literature about writing skills in Namibian schools. The current Namibia Senior Secondary Certificate (NSSC) syllabus for English as a Second Language for Grades 11 and 12 was first implemented in 2006 and the first examinations written in 2007. The document was developed by the National Institute for Educational Development (NIED) in collaboration with the University of Cambridge International Examinations. The curriculum content for writing is divided into three categories: Skills, Objectives and Competencies. A few competencies related to academic writing outlined in the syllabus are worth mentioning here: writing paragraphs, using introductory, developmental and concluding paragraphs, linking ideas, planning, structuring, drafting and editing, using appropriate style, format, layout, vocabulary, grammatical structures and focusing on interpretation of the topic, giving factual information, defending ideas and opinions (NIED 2005: 10-11). On inspection of the competencies listed in the syllabus, it seems that the writing curriculum has been extended to comply with the features of the process approach. One would, therefore, assume that students should have better writing abilities than reflected in the actual grades of the final examination (refer to Appendix 1: the spread of symbols for English as a Second Language) and that they would have more effective writing skills at university level. In the syllabus there is no mention of using any texts as examples to guide students in any of their writing activities, which leads me to 9 assume that the modeling/imitation writing approach is either not well-known or not deemed acceptable as a learning tool. The International General Certificate of Secondary Education (IGCSE) English as a Second Language syllabus used before the NSSC syllabus (from 1991 to 2006) was vague in terms of curriculum content. The writing component prescribes learners to: - Carry out simple writing tasks, such as completing forms, writing postcards or short letters in an appropriate and accurate form of English in response to a written stimulus; - Demonstrate the ability to describe, report, give personal information; identify, organize and present given material in particular form; - Carry out longer writing tasks on a wider range of topics (only for the Extended option) (University of Cambridge International Examinations 2004: 15). These curriculum objectives are linked to six (6) assessment objectives which relate to clear communication, conveying information and expressing opinion; employing and controlling grammatical structures; understanding and using a range of vocabulary; observing conventions of paragraphing, punctuation and spelling and employing appropriate register (University of Cambridge International Examinations 2004: 2). Again, nothing is said anywhere in the syllabus about the process approach or the imitation approach or any other approach in relation to any of the skills that students are expected to acquire. The current NSSC syllabus is much more descriptive in terms of academic writing learning outcomes and specific items to be covered by students during Grade 11 and 12, which theoretically results in students being able to meet the academic writing learning outcomes at university level. But, the final IGCSE and NSSC results (as indicated in the statistics) and the testimonies of UNAM lecturers display another picture. This inconsistency directed my study in terms of finding a method to improve the academic writing skills of students at UNAM, Oshakati Campus. 1.2.5 2 Comments from five Namibian teachers As part of background information to determine whether the process and/or modeling/imitation approach were formerly or are currently applied in secondary schools I obtained information from five English teachers regarding the application of these 2 The identities of the teachers are concealed. Teachers granted permission to use the information. 10 approaches. No print information is available on the extent of the application of the process and modeling/imitation approach in Namibian schools. Therefore, five teachers were contacted for first-hand information on the application of the process and imitation approaches in Namibian schools. The teachers have been teaching at secondary school level for more than ten years. I believe their approach to teaching writing has changed over the years depending on the level and the need of the students, but I have not asked whether they had always had followed the same procedure of teaching writing since they started teaching. These comments should by no means be seen as representative of the whole country. In terms of the process approach the following information was gleaned from teachers’ responses: three teachers announce the topic of the essay and analyze or explain the topic. Learners generate their own ideas without assistance from teachers. Two of the three teachers allow time for planning in the form of writing down notes of the ideas, after the teachers have provided a simple example. One teacher believes that pre-writing activities are a waste of time since learners are not committed enough to do that activity. The following activities or features are not explicitly dealt with at all: peer-editing, writing multiple drafts, revision and re-writing. Two of the three teachers give time for language editing. There is no mention of the recursive nature of the process approach. In fact, when learners are required to write an essay, they have only one English lesson to write their essays. Peerediting is avoided as teachers reason that learners’ abilities are too weak, and thus they will not be able to help each other and there is no time to do peer-editing. These three teachers felt that the process-genre approach could be useful in improving the writing skills of learners, but at primary school level, since there was not enough time in Grade 11 and 12 to apply this approach to writing. The imitation approach seems to be the least-applied of the two approaches: only one teacher mentioned that she tells a story which is similar to the essay that learners are expected to write. The teacher believes that learners will merely copy the example, therefore no modeling of the actual writing process is done either. Teachers do not explicitly teach or model any of the following aspects: structure and paragraphing, topic and supporting sentences, writing introductions and conclusions. Teachers commented that it is too time consuming to use the modeling approach at secondary school level. According to the two teachers in Grootfontein and one in Windhoek modeling/imitation is something that learners should have done at primary school level and there is no time to catch up on skills that were 11 neglected at primary school level. One teacher provides a model essay to students after the graded essays have been handed back to learners. Two teachers at two local schools, however, make use of the process approach in their Grade 11 and 12 classes to a certain extent. One teacher mentioned it to be “a very effective approach especially when introducing/teaching essay writing for the very first time”. She also mentioned that some learners are still struggling with writing despite the extensive guidance she offers. The other teacher allows her learners to do the pre-writing activities in groups of three and the teacher marks the group work. After checking the group work (editing by learners themselves and the teacher), learners write their individual essays on a similar topic. She uses the group essay as a model that students can use when they prepare for their individual essays. Learners are allowed to change or add ideas at any time during the process at both schools. One teacher only uses the modeling approach when she does revision with the learners towards the end of the year before examinations, when she provides model answers from previous examination papers, Examination Busters and Namibian Open College of Learning (NAMCOL) modules. These findings influence the research context of the study in that they present information which shows that students have not had much (if any) exposure to the process approach, or the imitation approach at school level. In the next section the context of the current study will be explained. 1.3 THE RESEARCH CONTEXT OF THIS STUDY With the language policy that emphasizes English as a sole official language, Namibia, like other countries in Africa such as Zambia, will limit the participation and involvement of the majority of its people in economic, political and social development. (Mutumba 1999: 4) Namibia, like many other African countries, is characterized by linguistic complexities. According to Tötemeyer (2009: 1) there are 14 spoken and written languages in Namibia. Since language is the tool for communication “and the medium through which thoughts, values and attributes are transmitted” (Mutumba 1999: [1]), multilingualism is an objective incorporated in the national language policy. 12 Before Namibia obtained independence from South Africa in 1990, the language policy stated that Afrikaans was the medium of instruction (MOI) in schools and Afrikaans and English were the official languages (Tötemeyer 2009: 1). After Independence in Namibia, learning in students’ mother tongue is limited to the first three years of primary education (Ministry of Education and Culture (MEC) 1993: 4). English is introduced in the Namibian primary school curriculum as the MOI in Grade 4 and English as a subject is offered from Grade 1 (Government: Presidential Commission Report 1999: 109). Students at school and at tertiary institutions are challenged with understanding the subject content in the MOI through listening, reading, writing and speaking, “‘no matter how imperfectly [English is] used’” (Tötemeyer in Harlech- Jones 1990: 87). Yet, only 0, 8% of the Namibian population use English as a First Language (Wolfaardt 2005: 2360). English is a second, and sometimes even a third language, to almost all the learners in the north-central region of Namibia, since the mother tongue is Oshiwambo, which has 7 dialects (Nyathi 2001: 3). Despite the number of years learners had been in contact with English, the failure rate in Grade 12 is a serious problem as the statistics3 indicate. However, some students do manage to obtain a C-symbol on the International General Certificate of Secondary Education (IGCSE) or National Senior Secondary Certificate (NSSC) for English as a Second Language and are therefore able to apply at UNAM main campus or Oshakati Campus. Once students are at the university, lecturers find that despite students having a Csymbol or better in English they lack academic writing proficiency skills (Willemse 2005; Nyathi 1999; Tötemeyer 2009: 3). Little formal writing practice is done in lesson time at UNAM in English Communication and Study Skills and English for Academic Purposes. The learning outcomes for the unit focusing on academic writing conventions are: role and purpose of writing; paragraph writing mechanics (thesis statement, main and supporting sentences); determine and categorise arguments for and against a topic; use cohesive devices in writing; distinguish between academic writing and other forms of writing; generate an academic text following academic writing conventions; apply the American Psychological Associations (APA) documentation style; plagiarism in academic writing; identify effective introductions and conclusions in academic writing; employ the mechanics of drafting and editing essays; design essay titles, plans and outlines; compose memos, reports and business letters; and prepare minutes and assemble business reports (Language Centre UNAM 2009: 3 Statistics and graphs are included as Appendix 1. These statistics are referred to again later in this section. 13 vii). These outcomes are measured with essays that students have to write during the semester tests and examinations. Marking grids are provided to lecturers for grading essays (Appendix 3) and students to use as guidelines during the writing process. 1.3.1 Student and campus background UNAM Oshakati Campus is a satellite campus of the main campus in Windhoek and has tuition and research facilities, which include classrooms, a library, computer centres, video conferencing rooms and offices and laboratories. The Oshakati Campus offers a full-time nursing degree and also focuses on part-time studies in education. Another programme currently running at UNAM Oshakati Campus is the Foundation Programme (FP). This programme is a full-time 9 month pilot course (March to November). Classes started in February 2005 with an intake of 60 students from the north-central region of Namibia. Currently, the intake is 150 students per year. The FP offers Physics, Chemistry and Biology in addition to Mathematics and English. Materials were written by lecturers based on a needs analysis that was done at the conception of the course. Each class gets one hour more for English and Mathematics as these are the two subjects which are deemed to require more input. Students also receive Career Guidance and Computer Literacy lessons. Students on the FP come from weak educational backgrounds as many schools in this area are resource poor. The majority of students come from the rural areas of the north-central region of Namibia where parents are subsistence farmers with very little or no financial income. The ages of students range from 17 to 24. Most students have as their mother tongue Oshindonga or Oshikwanyama, but there are students from other regions who have different mother tongues like Otjiherero, Damara, Nama or Rukwangali. The main aim of the English syllabus is to develop students to be conversant with English for academic purposes. Overall, the teaching philosophy intends to enable FP students to acquire the necessary skills for university studies in general and science courses in particular. The English Communication and Study component of the curriculum focuses on the full range of communication skills needed for academic studies at university level. There is no formal textbook that is used by the lecturer and students. The lecturer selects materials that can be used in an interactive way in the English classroom to enhance students’ English reading, writing, listening and oral skills. The content of the materials is often changed depending on the interests, needs and the level of the students. In addition, cross-curricular teaching is 14 encouraged and the English lecturer(s) therefore often use science-based texts in the English lessons. 1.3.2 Learning outcomes of the English Foundation Course (Extract from the Course Outlines of the Foundation Programme UNAM Oshakati Campus) Here, only the learning outcomes and skills relevant to the research project are provided. These learning outcomes were determined before I started with the research. Writing skills: Write clear, accurate summaries and notes; write clear, accurate scientific reports and assignments; convey information effectively using a range of appropriate vocabulary; employ and control a variety of grammatical structures; observe conventions of paragraphing, punctuation and spelling. Write logically and cohesively within a structured argument; use definitions, examples; plan and edit work; and clarify observations and descriptions. Critical reading and thinking skills: Understand and respond to information presented in various academic genres; select and organize material relevant to specific purposes; infer, interpret information and draw conclusions from scientific texts. Identify key points and analyse and synthesize information from texts. Study skills: Research and reference texts; time management and organizing work; rules concerning plagiarism; library skills; revision and memory techniques. 1.3.3 Grading Criteria for Essays The grading criteria used in the Foundation Programme to grade essays are extracted from Murray and Johanson (1990: 22) (Appendix 2). These criteria were selected as suitable for the writing purposes of the FP English course as they address the issues raised in the curriculum of the course. The grading criteria of content relate to the writing skills outcomes in terms of accuracy of facts, meaningfulness of information, critical analysis and synthesis of information and key points. Regarding form or structure of essays, the outcomes and criteria show a relationship: paragraphs, coherence and 15 cohesion, introductions and conclusions, topic and supporting sentences and logic are factors addressed by both the learning outcomes and the marking criteria. The grammatical learning outcomes are reflected in the criteria by the focus on punctuation, general grammar conventions and sentence structure. Style in the FP English learning outcomes is connected to the marking criteria by the following points: purpose, sense of audience, and useful rhetoric. 1.3.4 The situation of the researcher I am the English lecturer on the FP at UNAM Oshakati campus. After completing my degree in languages in South Africa, I have been teaching English as a Second language for 18 years in Namibia. I started at secondary school level, then taught for three years at a Teacher Training College. I then moved on to teach English International General Certificate of Secondary Education (IGCSE) to students on the Access Course (UNAM), a former programme offered at the Oshakati Campus, before teaching the English component on the current course, the Foundation Programme. In 2006, I obtained my Honours degree through UNISA in Teaching English to Speakers of other Languages. Two observations are central for understanding the choice of my research: The value and importance of writing is recognized by many ESL practitioners, especially for students at tertiary level. A quotation from a well-known South African linguist explains the essence of the relationship between coherent writing and academic achievement very effectively: How important is it that students ‘write well’? Although it is very difficult to quantify the relationship between students’ ability to write coherently and their levels of academic achievement, these two variables could be expected to correlate positively – an expectation strongly supported by one of the findings of a recent empirical study (Hubbard 1989). Given this correlation, and the present widespread concern about the first-year failure rate at South African Universities, the need to improve students’ writing skills becomes a matter of urgency. (Hubbard 1989: 1) I believe that focused instruction in the writing process and extensive practice with models as examples will improve the academic writing skills of Foundation Programme 16 students. The process approach will help with showing and practicing the actual process of writing as advocated by Ho (2006); Liebman-Kleine (1986); and Geyser (1996) and the imitation approach gives examples of genre essays which help students to understand what is expected of them as explained, amongst others, by Geyser (1996); Flowerdew (1993); Horowitz (1986); and Jones & Freeman (2003). This statement immediately brings to mind the tension between process and product, where the process approach has a strong focus on the actual composing process whereas the modelling/imitation approach to writing emphasises the product. Yet, writing is both, and thus a combined approach should be more effective than one that focuses on only one aspect. Geyser (1996: 222) quoted Richards (Richards et al 1992: 229-230) and explained the difference between process and product very clearly, but she also argues for the combination of the two approaches: The process approach to writing differs from the product approach, in that the whole process of writing, from the pre-writing to the final editing and proofreading, should be done under the supervision of the teacher. This process results in a final writing product, namely the letter, proposal, essay, et cetera, that the student had to write. Language teaching and learning are concerned with both process and product. Geyser (1996) and Richards (1992) believe that the values of process and product as portrayed in the modeling/imitation approach should not be pitched against each other as opposing strategies to teach writing effectively. Instead, process and product both form an important part of language learning and could be effectively used to teach writing, hence my aim to determine whether a combination of both approaches results in materials that are more effective in improving students’ writing skills and ultimately their English results, than just using one of the two approaches. This controversy between process and product will be discussed in more detail in the Literature Review section. In the next section the research problem and the proposed intervention are discussed. 1.4. RESEARCH PROBLEM English Additional Language (EAL) academic writing proficiency among secondary and tertiary students has been identified as a problem in previous studies (Benjamin 2004, Nyathi 1999; Nyathi 2001; Pflaum 1996; Swarts 1995, Yeld 2003) conducted in Namibia and 17 elsewhere. Reasons for the challenges in the Namibian educational system were provided. However, after visiting the UNAM main campus library, the Oshakati Campus library and the National Institute for Education Development and conducting a desktop search, no Namibian research could be found on the application of different writing approaches to determine the level of efficiency of improving students’ writing skills. Three studies (Nyathi 1999; Nyathi 2001; Willemse 2005) have reported on the impact of weak academic writing skills and offered recommendations on improving the status quo of EAL teaching and learning in Namibia but these are broad in nature and should have a positive effect in the long term. Nyathi (2001: 9) contends that a “lack of academic writing skills in the ESL curriculum definitely affects academic writing required in other subject areas like geography and history”. However, I believe that a more pragmatic, hands-on approach is necessary to help teachers and learners in their quest for effective academic writing skills. Observations of weak and/or ineffective academic writing skills were made among the Foundation Programme (FP) students at UNAM Oshakati Campus. These students displayed inadequate essay writing skills in terms of content, structure and language usage (grammar). This is obvious in the essays written by students who attended the course in 2005, the year of inception of the FP. The English lecturers found that writing skills of students were not up to standard based on the marking grid (Appendix 2). Prospective students’ marks for the entry test essays ranged between 17% (the lowest) to 76% (the highest score), with the majority between 30 and 49%. The FP Biology, Physics and Chemistry lecturers also confirmed this finding by stating that former (2005 – 2007) students’ laboratory report writing skills were below the level of accepted academic writing conventions in the science fields. Interviews held in 2006 with 3 UNAM Main campus science lecturers and 2 lecturers from the main campus Language Centre confirmed the researchers’ assumptions about the weak academic writing skills. The current study intends to address an issue raised by Nyathi (2001: 178) when he recommended that the Ministry of Education introduces writing materials that better prepare students for tertiary education in terms of academic writing. His recommendation entails the following: These activities should entail among others, helping learners to use their cognitive skills to generate ideas and organize them in their written compositions. Activities should involve practices like, planning, organizing information in accordance with 18 academic requirements e.g., how and what to include in the introductions, the body and conclusions. For instance, all introductions would need to have thesis statements, all paragraphs would need to have topic sentences that develop the thesis sentence of the introduction and the conclusion that sums up what the entire composed text carries (Aaron, 1997). Similarly, purpose and audience should be stressed in this teaching of writing through composing as these carry the core of ESL writing under the communicative syllabus (Krashen & Terrell, 1983; Rivers, 1968; Bialystok, 1990; Lightbrown & Spada, 1993). This study, therefore, aims to report on the levels of improvement recorded in the implementation of three approaches: process approach, imitation approach and the process genre approach. In each case I developed special writing programmes to improve the general academic writing ability and laboratory report writing skills of FP students. The Research Problem was approached in the following manner: 2008: Action research on the effectiveness of: a.) The process approach to improve academic writing skills of FP students (Class 1); b.) Modeling approach to improve the academic writing skills of FP students (Class 2); 2009: Action research (AR) on the effectiveness of the process genre approach, a combination and adaptation of materials used for the process and the modeling approach in 2008 in Class 1 and 2. This means in 2009 both classes had the same materials which were based on the process genre approach. The history, principles and applications of these three writing approaches as well as information on academic writing will be explained in the Literature Review. In the next section, however, brief mention will be made to the proposed intervention, which entails using the modeling and process approach and the process genre approach to teach academic writing skills. 1.4.1 The proposed interventions An intervention is deemed necessary when the current situation yields unsatisfactory results. An intervention is inextricably part of an action research, in the form of an action that is intended to bring about a beneficial change (Bell 2005: 8). In the case of the current project 19 an intervention was called for to improve the academic writing skills of FP students and to improve the teaching practice of the lecturer/researcher. This forced me to think critically and systematically about the content of the curriculum and syllabus and what approaches are to be followed. It also linked with the next planning and decision activity: what type of texts will be used? According to Shih (1986: 618), “to prepare students for university courses, it is important to have information about the types of writing tasks actually required across academic disciplines and about instructors’ purposes in assigning these tasks.” The intervention is based on the belief that students need explicit tuition and instruction in the value and practice of academic writing skills in order to enable them to improve their abilities in ESL writing and consequently in writing in their other subject areas, in this case sciencebased subjects. The intervention is grounded in theory relating to writing. These components form part of the conceptual framework of the study and will be mentioned here but discussed as part of the theory base in the Literature Review. The conceptual framework of the intervention:  Academic Writing: This is a formal type of writing that requires students to adhere to certain conventions or characteristics which identify the specific type of writing as academic, for instance hedging, using third person singular, passive voice, having topic and supporting sentences and many more to write different kinds of reports and essays (descriptive, process, argumentative, cause and effect, compare and contrast). Thaiss and Zawacki (2006: 4) define academic writing as “any writing that fulfills a purpose of education in a college or university”.  Process approach: Proponents of this approach maintain “that writers create and change their ideas as they write” (Caulk 1994: 181) and that writing is a recursive activity (Geyser 1996: 222), which involves a number of not necessarily linear steps like prewriting, composing, revising, evaluation and the final publishing of the product.  Modeling approach: The modeling approach to writing involves providing students with examples of effective pieces of academic writing or laboratory reports and it helps students to emulate the stages that are involved in the writing process. 20  Process genre approach: This is an eclectic approach with features from the process approach, the modeling approach and the genre approach (in the case of this study). According to Badger and White (2000: 159) the following components from part of this approach: situation, purpose, mode, field or tenor, planning, drafting, publishing. In this study (and in Falk 1979: 438; Spencer: 1982-1983: 42; Yan 2005: 21) the importance of example texts for analysis and imitation is stressed as a useful tool to improve academic writing. The focus on form and structure of different genres is given equal weighting as the focus on content. In addition, revising and re-writing are also encouraged. The proposed intervention is anticipated to address the research problem. 1.5 AIM OF RESEARCH This mixed method action research study focuses on the experiences and perspectives of the researcher, the students and four science lecturers on the FP in relation to the specially designed materials for the three different approaches. This links with the purpose of qualitative research, which is to understand situations from various angles and to offer a rich and deep interpretation of data. However, as this study also has a quantitative element, the aim can be extended to explaining whether there is a relationship between the intervention materials used to improve academic writing skills of FP students and their actual performance after the intervention. Therefore, the specific aim of this action research is to determine the effectiveness of the writing intervention programme that is developed to improve academic writing skills. For the purposes of this study, action research is defined as a collaborative, pragmatic and critical activity (Herr & Anderson 2005: 4), which is carried out by the researcher, in this case with the students of the Foundation Programme. Specific aims:  To determine whether the process approach is effective in improving the academic writing skills of FP students;  To determine whether the modeling/imitation approach is effective in improving the academic writing skills of FP students; 21  To determine whether the process genre approach is effective in improving the academic writing skills of FP students.  To determine which of the three above-mentioned approaches is the most effective in improving FP students’ academic writing skills. 1.6 HYPOTHESES Hypothesis is defined by Nunan (1992: 230) as “A formal statement about an expected relationship between two or more variables which can be tested through an experiment.” Leedy and Ormrod (2005: 7) propose that a hypothesis is a “tentative solution to the problem(s)”. The following hypotheses of this research study are constructed based on the four research aims:  There is a positive relationship between the application of the process approach and improvement in academic writing skills of FP students.  The application of the modeling/imitation approach results in an improvement in academic writing skills of FP students.  The process genre approach leads to improved academic writing skills of FP students.  The three writing interventions are effective in improving academic writing skills with regard to English essays and laboratory reports in Physics, Chemistry and Biology. 1.7 RESEARCH QUESTIONS Research questions are very specific questions related to the topic and the aims of the research project. The purpose of having research questions is to ensure that every part of the research problem is addressed and the problem is broken down into smaller, more manageable chunks. Questions (a) and (b) aim to determine whether students had any background knowledge or experience of academic writing. The remaining four questions deal with the focus of the research: the effectiveness of the three writing approaches and show a clear relationship to the hypotheses. 22 This study aims to find answers to the following questions4: a) What background knowledge do FP students have of ESL academic writing and laboratory report writing? A questionnaire was used to determine the students’ experience and knowledge about academic and laboratory report writing. (Appendix 6) b) To which extent were they exposed to academic writing skills at secondary school? Did they do any activities that would introduce them to academic writing skills? Did they do any pertinent academic writing activities? The same questionnaire used in (a) was used to find out whether students had any specific exposure to academic writing skills. (Appendix 6) A pre-test essay (before the teaching actually starts) was also used to determine any practical application of what students learnt in terms of academic writing before coming to the FP. This essay is evaluated by 2 UNAM lecturers who do not teach the 2 classes which are involved in the intervention. The marking grid with clear criteria was used (Appendix 2). Pre- and post-intervention laboratory reports were written by students to determine their level of academic writing skills. c) How can the interventions be described? A brief explanation of the interventions is provided in Chapter 4. I wrote the materials and made use of authentic materials based on the content of the other subjects on the course. I also used books on academic writing and general writing skills to provide theoretical background and model essays as well as activities. d) How have students’ experiences of and attitudes towards academic writing changed during and as a result of the writing intervention? 4 These questions were set up based on the research of Creswell (1998: 105). 23 This was determined by supplying a questionnaire (the same questionnaire as in (a)) to students which asked questions about their experiences and attitudes. e) To what extent do the students believe they have benefited from the writing intervention programme? The questionnaire proposed in (a) provided feedback on their individual and collective benefits (if any). Also, the interviews with science lecturers and a comparison analysis of the pre- and post-intervention essays and laboratory reports provided an answer to this question. f) How effective have the interventions been to improve students’ writing results in academic essays and laboratory reports? The marks of the pre-intervention essays and laboratory reports were used as an indicator of the students’ initial academic writing ability. The marks of the postintervention essays and laboratory reports provided information about the extent of their improvement (if any). 1.8 DEFINITION OF TERMS Approach: An approach is a method which is used in teaching language. Each approach is based on theoretical principles with regard to language learning and teaching. Brown (1990: 158) defines ‘approach’ as follows: “a set of assumptions dealing with the nature of language, learning, and teaching.” Laboratory report: This is a report that is written after students have conducted experiments in the Biology or Physics laboratory. The reports have to show the topic, the aim, materials, procedure, results, and conclusion of the experiment. Modeling approach: This approach involves students reading and analyzing effective examples of writing and using the features of the examples to write their own pieces of writing effectively. The focus is on the product. It also includes the emulation or imitation of activities used to follow the different stages of the writing process. Process approach: This approach is based on the principle that writing is a process that involves a number of steps and skills that are applied in a recursive manner to produce a clear, understandable, meaningful piece of writing. The focus is on the process of writing. 24 Process genre approach: The process genre approach is an eclectic approach that combines the development of writing/composing skills (process) with the situation, audience, function and form of a text (genre). In the case of this study the imitation aspect of the modeling approach was fore grounded with multiple examples of a specific genre used for analysis and emulation. 1.9 ABBREVIATIONS BSc. Bachelor of Science ESL: English Second Language FP: Foundation Programme MOI: Medium of Instruction UNAM: University of Namibia NSSC: National Senior Secondary Certificate IGCSE: International General Certificate of Secondary Education VSO: Volunteer Service Overseas IT: Information Technology 1.10 CHAPTER OVERVIEWS Chapter 2 gives information on the three writing approaches: process, modeling/imitation and process genre approach. This Chapter provides the features of each approach, information on research of the approaches in application and arguments against the approaches. In Chapter 3, the focus is on the qualitative and quantitative action research design and the research instruments. The Chapter explains how the interviews, questionnaires and essay and laboratory report marks were used to collect data. Attention is also given to the population and convenience sampling and how the data were analysed. Chapter 4 deals with the results of the data. Extracts and a meta-analysis of the interventions are provided in the Addendum. Personal reviews about the effectiveness of the three approaches explain how I evaluated the actual application of the approaches in the FP English lessons. The results of the pre- and post-intervention essay and laboratory report marks are presented in mean score and Independent Samples t-test tables as well as Bar graphs. The closed-question questionnaire results are provided in frequency tables to compare the pre25 and post-intervention academic writing experiences, habits and attitudes. Open-question questionnaire results are dealt with in a qualitative, descriptive manner with a few tables and graphs to illustrate the results. The results of the 2009 pre- and post-intervention interviews with science lecturers are explained and discussed lastly. In Chapter 5, I present the findings and conclusions of this study. Observations regarding the contributions of this study and suggestions for further research are provided in this last Chapter. 1.11 CONCLUSION In the first chapter, the scene has been set for the proposed research study. This chapter provides information on the research problem, and gives an overview of the complexities leading to the problem. The implications of the research problem are broken down into research questions. The next chapter will be a synthesis of Literature reviewed on the theory and principles of the three writing approaches and of similar studies conducted in other situations. 26 CHAPTER 2: LITERATURE REVIEW 2.1 INTRODUCTION In this chapter I aim to inform the reader about the theoretical perspectives underlying the current study. I also provide a critical analysis of previous research on the process approach, the modeling/imitation approach and the process genre approach. In the first section of the chapter, I will look at academic writing, a broader issue related to the research problem. The next section deals with the domain of the approaches used in the intervention: the process and modeling approaches and lastly the process genre approach. 2.2 THE “MYSTERY” OF ACADEMIC WRITING SKILLS 2.2.1 Academic Writing ESL students experience great difficulties when they make the transition to the Englishmedium academic mainstream. (Shih 1992: 290) Academic writing is like trying to hold a slippery fish: you know it is there and it has actual, real consistency, but it is very hard to pin down and actually describe it in clear, uncontested terms. Lillis (1999: 127) explained that “academic writing is ‘mysterious’”, and that it is commonly misunderstood by students and lecturers. The history of academic writing does not make it easier to find a tangible definition (Spack 1988; Horowitz 1986a, 1986b, 1986c; Liebman-Kleine 1986). Thaiss and Zawacki (2006: 4) reasoned that the concept academic writing is “used imprecisely yet almost always for what the user regards as a precise purpose; e.g., commonly by teachers in explaining what they want from students.” Furthermore, they contended that abstract definitions of academic writing are abundant and there are “differences in standards and expectations among disciplines and among teachers” (Thaiss & Zawacki 2006: 4). Students need to be made aware of and able to use the basic rhetoric, linguistic aspects, form and the cognitive processes involved in academic writing at their specific level of education. Hofstee (2006: 187) proposed that academic writing has the following characteristics: clarity, accuracy, brevity, simplicity, and focus, whereas Thaiss and Zawacki (2006: 4-6) listed the characteristics as follows: attention to the topic of study and reflective thought about it, that reason dominates emotion, and that an academic writer should display analytic ability. There 27 are also differences in the meaning and teaching of academic writing depending on the institution where students are studying and their level of education. The main aim of writing about academic writing is to illustrate the competencies and skills that students are expected to master at tertiary level. Academic writing at UNAM is required in the first year English courses as well as in other subjects. Even though there are differences in the lecturers’ understanding of what constitutes academic writing, in my view there is a basic definition: academic writing shows students’ understanding of an expository or argumentative topic and of writing conventions by providing a clear, meaningful thesis statement that is discussed in an organized, logical, fluent and accurate manner. Academic writers have to use semi-formal or formal voice and mainly third person point-of-view. I would like to further clarify the issue by providing some information that “demystifies” academic writing. Weideman (2003: 55-65) outlined the “abilities and components ... students are required to have at tertiary level”. According to Weideman (2003: 61) the following important components constitute academic literacy. Students need to  understand a range of academic vocabulary in context;  interpret and use metaphor and idiom, and perceive connotation, word play and ambiguity;  understand relations between different parts of a text, be aware of the logical development of (an academic) text, via introductions to conclusions, and know how to use language that serves to make the different parts of a text hang together;  interpret different kinds of text type (genre), and show sensitivity for the meaning that they convey, and the audience that they are aimed at;  interpret, use and produce information presented in graphic or visual format;  make distinctions between essential and non-essential information, fact and opinion, propositions and arguments; distinguish between cause and effect, classify, categorise and handle data that make comparisons;  see sequence and order, do simple numerical estimations and computations that are relevant to academic information, that allow comparisons to be made, and can be applied for the purposes of an argument; 28  know what counts as evidence for an argument, extrapolate from information by making inferences, and apply the information or its implications to other cases than the one at hand;  understand the communicative function of various ways of expression in academic language (such as defining, providing examples, arguing); and  make meaning (e.g. of an academic text) beyond the level of the sentence. These components were identified in the South African field of education as relevant to improve academic literacy of all students. Since Namibia has a similar situation with regard to underprepared students wishing to do degree courses at tertiary institutions, these components are relevant in the Namibian context as well (and a number of abovementioned aspects are present in the current FP English curriculum). The aim of academic literacy programmes according to Blanton (1994: 230) cited in Weideman (2003: 58) is to get students to “speak and write with something we call authority”. This notion is in line with the aim of my intervention materials of the current study at UNAM Oshakati Campus. While knowing “that academic writing conventions are neither universal, nor independent” (Peretz 2005: 55) and not static (since they change as a result of linguistic, educational, and technological development), it is critical that students have guidelines for their initial academic writing activities across the curriculum (Leki & Carson 1994: 82). The responsibility of the writing teacher is to expose students to various writing strategies which “include combinations of activities such as outlining, drafting, or free writing” (Lavelle & Bushrow 2007: 808; Spack 1988: 34) based on their level of general and academic writing experience. Given the limited time of many academic writing courses, the materials and teaching and learning situation are often a “short-cut method of raising students’ proficiency” (Dudley-Evans: 2002 [2]) to reach the required level before starting their undergraduate studies. The “short-cut method” is supposed to help students achieve at least the following: “an expository writing model which includes a thesis statement in the introduction, followed by paragraphs that start with topic sentences, and examples that support the thesis, which are then followed by a logical conclusion” (Heffernan 2006: 4-5). This should help students write more effective essays especially in tests and examinations, since they would have been exposed to a range of genres identified as core writing tasks in English as well as other subjects, like Biology or Physics. However, the composing process 29 vis a vis the process approach “emphasizes linguistic skills, such as planning and drafting, and there is much less emphasis on linguistic knowledge, such as knowledge about grammar and text structure” (Badger & White 2000: 154). The next section provides a detailed account of the principles and applications of the process and the modeling approaches and evidence for the functional features of the process genre model. This approach will be dealt with lastly. 2.3 PROCESS APPROACH To think of writing as a process instead of a product is simply one perspective, a way of looking at writing, an orientation that has led to hundreds of different approaches for researching and teaching and theorizing about writing. Process is not a dogma, but a concept that enables people to see writing in a new way and thereby ask questions that were not asked as long as people saw writing simply as finished products. (Liebman-Kleine 1986: 785) 2.3.1 What is the Process approach? The process approach5 emerged in the mid-1970s as a counter reaction to the product approach (Yan 2005: 19). According to Nordin and Norhisham (2006: 76) the criticism leveled against the product approach dealt with formalistic rhetoric which was said to devalue the actual, individual creative writing process and an over-emphasis on the product. Other issues, like expecting students to “create a perfect first draft” (Yan: 2005: 19), and the constant focus on error correction had writing teachers and researchers concerned. The process approach movement began with studies about the composing process of writers (Emig 1971; Perl 1980; Pianko 1979) and resulted in informing students how to approach a writing task. The process approach was developed initially for first language classrooms (Caudery 1995: [1]) in English-speaking countries (Gao 2007: [5]). It has later been adapted for additional language teaching (Caudery 1995: [1]). According to Brown (1994: 320) at that time the product approach was used in composition classes where the focus was on grammar, vocabulary, spelling, punctuation, content and organization (Yan 2005: 19) and the writing activities were mostly decontextualized (Gao 2007: [4]). Learners were instructed to write essays on given topics. These were collected, critical comments were provided and the 5 I indicate the key concepts of the principles and features of the process approach in bold for easier reading 30 essays were assessed and evaluated (Caudery 1995: [1]). The essays were returned and learners were supposed to do corrections. The actual process of how people write was negated by the use of the product approach. Also, the continuous error correction reduced learners’ motivation and self-esteem with regard to writing. Consequently, learners seldom learnt that an effective piece of writing can hardly be produced in only one draft. In addition, Zamel (1983: 165) pointed out that written products do not show teachers much about learners’ “ instructional needs”, which means then that teachers do not know how to help their learners if only written products are expected and if teachers are not part of the learners’ writing process at all. Even though an effective piece of writing, or the product, is the ultimate aim for any writer, there are different methods or strategies available to reach the product stage. This was recognized by some composition teachers and researchers (Emig 1971; Zamel 1982, 1983; Hairston 1982) and the traditional thinking about writing was questioned. 2.3.2 Features of the process approach In the 1970s writing teachers of native English learners started to explore “the processes that went on in the creation of written text” (Caudery 1995: [1]). It was found that during the writing activity a variety of sub-processes occurred in “cyclical and in varying patterns” (Caudery [1]). Another observation was that experienced writers used different processes as opposed to inexperienced writers (Caudery [1]; Zamel 1983: 165). “Prescriptive, formulaic” (Zamel 1983: 165) writing approaches with the focus on “correctness” were challenged by the realization that the composing process of learners needs to be examined and learners should be exposed to an instructional design that gives “students direct experiences with the composing process, that establishes a dynamic teaching/learning relationship between writers and their readers, and that enhances further linguistic development in the context of making and communicating meaning” (Zamel 1983: 165). In order to find answers about the dynamic composing process teachers and researchers (Flower 1979; Pianko 1979; Zamel 1976; Zamel 1982; Zamel 1983) asked the following regarding the composing process: How do writers write? Where do the ideas come from? How are the ideas formulated? How are the ideas developed? What stages do writers use? On account of these questions studies (Emig 1971; Perl 1980; Pianko 1979; Zamel 1976; Zamel 1982; Zamel 1983) were aimed at finding information about how writers write and how writers discover ideas and make meaning. In addition, these studies which involved observing writers during the composing process also showed that writers write according to basic patterns and that they make 31 decisions and choices in the writing process. Individual differences in making decisions, choices and in the use of patterns during the writing process were also reported. Revision and several drafts on the same topic were highlighted as common characteristics among all studies mentioned above. Consequently, various strategies and techniques were implemented or experimented with by writing teachers who addressed writing as a holistic process comprising not a single method but a culmination of writing considerations, which were then called the process approach. Furthermore, a new perspective was adopted regarding the role of the writing teacher. In contrast to the limited input from the teacher using the product approach, process approach teachers supported learners throughout the whole writing process and showed the learners how to overcome difficulties when composing effective essays (Caudery 1995: [1]). According to Zamel (1982: 195) a writing teacher should adopt the role of a guide who intervenes when necessary. The responsibility of a writing teacher depends on the level of the learners’ writing experience. Skilled writers need less guidance than unskilled writers. Most importantly, at the beginning of a writing programme for inexperienced writers, writing teachers are advised to explain the composing process and that writing is not a matter of getting everything right with the first attempt. All the misconceptions about the so-called importance of form before content and knowing exactly what to write, should be cleared (Zamel 1982: 205). Writing students need information on the characteristics and the benefits of the writing process. As an introduction writing teachers can show their own examples of the composing process including especially the behaviour during the actual pre-writing, drafting and other stages, which are hardly ever clear right from the start of receiving a writing task. Shaugnessy (1977: 222) agreed with this perspective: one of the most important facts about the composing process that is not made clear to the students is that “the process that creates precision is itself messy” (Zamel 1982: 205). Writing teachers are advised to illustrate to students that ideas get deleted and re-written, whole paragraphs become transformed after adding new ideas, revision of ideas and organization is a constant process and editing of form should take a less prominent position in the composing process (Zamel 1982: 205). While the level of involvement differs for skilled and unskilled writers, Zamel (1982: 203) advised that writing teachers of less skilled writers should actually teach learners pre-writing techniques and how to generate ideas to get them started. Moreover, research (Brostoff & Beyer 1980) suggests that further involvement from the writing teacher in terms of developing the ideas and finding relationships between the ideas results in 32 learning to use different types of “invention techniques” (Zamel 1982: 204). The value of the writing teacher’s focused facilitation and intervention is reflected in the learners’ ability to become effective writers when exposed to the writing process approach (Flower 1981; Graves 1983; Ho 2006; Zamel 1982). Teachers are required to be flexible, and interact in the learning environment. They have to offer a variety of methods and stimuli (Pianko 1979: 21). These qualities are constructive factors in the complex writing process which consists of hierarchical stages as explained in the next section. The process approach involves several stages which appear to follow each other but which do not necessarily result in clear sequential steps. The process approach is not linear, but rather recursive, which means the stages can appear anywhere in the process depending on the writers’ choices. Most proponents of the process approach (Yan 2005; Shih 1986; Geyser 1996; Tessema 2005; Zamel 1983) agree that the number of stages range from three (3) to five (5). Pianko (1979: 7) calls the stages composing behaviours and contributes to information of the stages by categorizing them as follows: Prewriting – all the activities that occur from the moment writers receive the essay topic until they write their first words. Planning – this stage involves the setting of parameters for the essay to be written. Planning activities can be mental, written, or both. Composing – this is what happens between the writing of the first word for the essay and the final part of writing. When writers compose they display three major types of behaviors exhibited - writing, pausing, and rescanning. Writing is the actual writing of the text. Pausing is when writers take a break in the actual writing for the purpose of thinking or for diversion (when the writer takes a break from the actual writing and does something totally different like, playing soccer or drinking tea). Rescanning – this involves a rereading of words, or sentences, or a paragraph. It is not a rereading of the entire text. During rescanning, revisions are usually made, most of which are single word, multiple words, or punctuation changes. At this time writers might also contemplate what they are writing. Rereading - when this occurs, writers reread the entire script for the purpose of seeing what has been accomplished, revising and proofreading, and, in some cases, for deciding on a conclusion. In many instances, rereading is done also 33 for the counting of words. Stopping occurs when students think they have written all they wish to about that topic and for that particular time. Contemplating the finished product – Writers often contemplate the finished product. The contemplation is most often of quite brief duration. Handing in of the product – writers hand the finished product to the teacher. (adapted from Pianko 1979: 7-8) The stages and explanations above as proposed by Pianko can be adapted to suit the writing teachers’ and the students’ needs and objectives. These stages involve a variety of different techniques or strategies which can be selected by learners depending on their preferences, skills and abilities, also called the “unique composing process” (Emig 1971). The stages that are the norm according to researchers (Zamel 1983; Shih 1986; Geyser 1996; Tessema 2005; Yan 2005) are (1) the pre-writing stage, (2) the composing/drafting phase, (3) the editing and revising stage (4) the rewriting or final draft stage. A crucial issue about these stages is the fact that they are non-linear and recursive (Emig 1971; Flower & Hayes 1981; Zamel 1983). This flexibility and uniqueness of the composing process is supported by Pianko’s study (1979: 20) which shows that “the text of a written composition unfolds” and that writing students should never be forced to write outlines if they do some of the composing processes mentally. Pianko (1979: 20) is of the opinion that writing students can consider the first writing of a text as the first draft, which can be added to, taken from or totally re-written without necessarily always having a clear outline of ideas. The fluidity of the process approach allows students to think outside the box in terms of generating ideas, writing, editing, revising, and rewriting. If writing tasks have been set for evaluation purposes, in my opinion, stage four (4) will be the last stage since a final draft - the product – is frequently handed in to the teacher for assessment and evaluation. Students can go back and forth between the stages depending on their needs, which are not the same with each essay. This leads to learners having multiple drafts based on one topic, and a continuous thinking about the topic of the intended product (Liebman-Kleine 1986: 786). Liebman-Kleine further emphasized the complexity of the writing process which encompasses among others, “cognition, emotion, sense of self, sense of others, situational background, experience, development” which all depend on the fact that this is a “pedagogy of difference” (1986: 785). Each writer has a different set of parameters regarding abovementioned factors (cognition, emotion, sense of self, etc), which indicates that one set of rules 34 or guidelines regarding the writing process is not suitable for every single writing learner. Each writer is unique, and does not use the same process as the next to compose writing. The dissimilarity in the composing process between skilled and unskilled writers (Perl 1980; Sommers 1980; Zamel 1983) also creates a case for acknowledging the complex nature of the writing process. For example, Perl (1980) found that even though skilled and unskilled writers showed similar behaviour in discovering ideas during the composing process, unskilled writers became pre-occupied with language factors such as sentence structure, spelling, and vocabulary earlier on in the composing process than skilled writers. Consequently, the exploration of ideas of unskilled writers was done at a more superficial level than that of skilled writers (Zamel 1983: 166). The individual variables among writing learners add to the complexity of the writing process as explained by Zamel and Spack (2006: 127): some learners had limited access to academic experiences, resulting in difficulties to adapt to unfamiliar linguistic practices and new classroom customs; others can speak more fluently than write; or there might be opposing attitudes towards the practices of the writing process approach. In acknowledging the multiplicity of the writing experiences of the learners and the complexity of the writing process itself, writing teachers should introduce and/or allow an assortment of strategies and techniques to equip learners to become effective writers. Apart from the variety and sequence of techniques and strategies that can be employed in the writing process, a deeper level of thinking about writing was established when the process approach emerged. The process approach to writing takes into account that a multi-faceted thinking process takes place when composing a text. During the composing process writers do not consistently think only about the writing assignment expected to be completed. The ‘composing thoughts’ get derailed by distractive thoughts about our lives outside the realm of the composing process. Perl (1980: 363) included a quotation of a teacher which shows the reality of writing behaviour, for instance the teacher who was part of the study explained that while she was supposed to be composing, her thoughts shifted from what vegetables she needed for dinner to the rain outside and back again to comments made by her colleagues in the writing class. According to Liebman-Kleine, the similarity between the process approach and academic writing is useful in the sense that in both the writer has to think about “ideas, self, audience, situation and purpose” (1986: 786). The underlying factor present in both concepts is the ability to “invent, organize, and revise” (Liebman-Kleine 1986: 786) which indicates a 35 dynamic, generative process. Whereas the process approach assumes that writing is learning, academic writing is done to illustrate students’ knowledge and understanding, in other words: their learning. But academic writing also assumes that the writing tasks in the content subjects like Biology, Physics or Chemistry develop learning. Therefore, in academic writing, writing to learn is also reinforced. In brief, this means writing in the process approach assumes the following: writing as a tool for thinking, and thinking how to write. These are vital skills to be adopted by especially tertiary education students who need to display their knowledge and understanding through academic writing. The deep level of thinking during the writing process allows learners to explore the generative and inventive nature of composing (Zamel 1983: 166). This characteristic of the process approach is present throughout all the stages of the writing process, which emphasizes that the final text - the product - is not pre-determined. At the outset of the task of having to write an essay, the learners do not know what to write, as writing is not the result of “the development of some preconceived and well-formed idea” (Zamel 1982: 197). On the contrary, Shaughnessy is credited with illustrating that writing is an instrument that is used to record the ideas that are developing (Zamel 1982: 197). Shaughnessy states that writing in this perspective is seen as “an act of discovery” (Zamel 197). The learners initially generate ideas based on the topic and the purpose of the writing task, the audience, and background knowledge (Zamel 166). In the stages following the pre-writing stage, more ideas may be generated, because writers read and re-read their texts to reflect on what was written and how that contributes to the meaning of the text and the direction of their thoughts (Zamel 1982: 197). Meaning is constantly discovered by the interactive process of thinking, writing, and revising in a non-linear fashion. The actual act of writing assists writers in the generating and invention procedure of composing a text, which guides us to another crucial facet of the process approach to writing: revising and re-writing. Goldstein and Carr (1996: [1]) view process writing as highly recursive as a result of the actions and writing behaviour that include planning, translating ideas into words and sentences, and then especially reviewing what has been written – not necessarily in that order. A number of studies concur with the key notion of recursion (Badger & White 2000; Emig 1971; Perl 1980b; Zamel 1982): “the recursive nature of the writing process … is that writers go back in order to move forward” (Zamel 1982: 197). Murray (1980: 4-5) observed, that the recursive quality creates interaction between the different stages which contributes to the discovery of meaning in a more comprehensive way than a linear model (Zamel 1982: 197). 36 Flower and Hayes (1981: 374) commented on the recursive nature of the writing process by stating: “The sub-processes of revising and evaluating, along with generating, share the special distinction of being able to interrupt any other process and occur at any time in the act of writing.” In addition, the role of the teacher as intervener is emphasized in the recursive nature of the writing process, as the teacher offers guidance throughout the composition process instead “of reacting only to the final product” (Shih 1986: 623). The different strategies or techniques used in the writing process can be viewed as tools to help the writers achieve their goals. The tools can be used at any time and in any order. In reality and based on its recursive nature, the writing process seems like a chaotic, unstructured way of composing, yet, it is believed to lead to writing that is coherent, purposeful and meaningful (Flower & Hayes 1981: 377). A main aspect in the process approach and in fact, the recursion process is the act of revising. Revising or review occurs when the writer re-reads the text for different reasons, either to check the relevance and progress of the content, or the appropriateness of the form and style or language issues. Shih (1986: 630) offered a distinction between two revising strategies: “internal revision” or “revising to fit intentions”. Internal revision refers to the content and structure of the written text. Students re-read their texts to determine whether what they have written matches with their intentions. The second type of revision, “external revision” deals with “conventions of grammar, diction, style, and mechanics” (Shih 1986: 631). Review is usually followed by re-writing, which is preceded by generating additional ideas. The reflection activity and consequent action(s) are triggered by seeing the actual ideas on paper. This corresponds with the idea that writers only know what they write if they can see what they write. Writers can only see what is missing in their writing if they read what they have written already. This concept is called “retrospective structuring” by Perl (1980b: 368). Drafting, reading, revising and re-writing take time. Time is a crucial pre-requisite to allow for purposeful, goal-oriented writing that is characterized by the use of recursive writing behavior as well as meaningful review. As part of the theory of the writing process, sufficient time to compose is a keystone. Geyser (1996: 222) referred to time in the following manner: “A process approach requires enough time, which in turn means that the teacher has to plan time and resources very carefully. Spending enough time on the pre-writing phase, but neglecting the rewriting phase because of a lack of time, is unacceptable.” She also stated that although teachers and writing learners perceive planning and pre-writing activities as “time-wasting” the activities usually 37 yield an improved “final product” (Geyser 1996: 231). The characteristics of the process approach to writing (recursive, stages, complex, generative and inventive) indicate that the writing behaviour, which involves meta-cognitive thinking about writing, is a time consuming act. According to Shih (1986: 623) it is crucial that students have “sufficient time to write and rewrite, to discover what they want to say and to consider intervening feedback from instructor and peers as they attempt to bring expression closer and closer to intention in successive drafts”. 2.3.3 Arguments against the process approach In the mid-1980s the process approach was heralded as an approach to writing that yields many benefits. But in the late 1980s Horowitz (1986: 141) argued that discussions about its drawbacks “are almost nowhere to be found” due to the zealous advocacy of the success of the process approach in improving writing skills of learners. Horowitz stated that the process approach has been accepted without looking at it critically, which has resulted in a number of issues which could affect students’ writing negatively. The first issue Horowitz (1986: 141-142) addressed as being overlooked by process approach writing teachers is the fact that the process approach does not prepare the students adequately for writing in the examinations. The reason for the underpreparedness is the issue of time. In-class writing activities allow the learner to write multiple drafts. But Horowitz (1986: 141-142) stated that it has not been determined yet whether this approach leads learners to writing essays faster in an examination setting. In addition, the teacher’s role in the process writing approach as guide and facilitator and ‘being there’ for the students who need help might also result in students’ higher level of dependence on the teacher. Is that useful in the examinations, where the teacher may not help the student? Furthermore, Horowitz (1986: 142) commented on the variety of writing processes and the fact that “there are as many writing processes as there are academic tasks and that anyone who claims to understand the former had better have a specific taxonomy of tasks in mind.” It is not proven beyond doubt that the writing process approach has a beneficial impact on all writing tasks, especially examination essays. Moreover, Badger and White (2000: 157) asked the important question about the validity of the process approach. Horowitz concurred by asking: “is there only one process, only one product?” (1986: 142). It appears that the same sort of composing processes can be used to produce any kind of writing. Very little attention is given to the different types of texts that 38 need to be produced and even less focus on the purposes of having different texts (Badger & White 157). In addition, very often the social context in which the text is supposed to be written is not regarded at all (Yan 2005: 19). The result of these inadequacies of the process approach might lead to questions among writing learners about “linguistic rhetoric choices” (Gao 2007: [6]) and a possible failure when having to produce texts that are context relevant. Another issue that was observed by a number of researchers (Rose 1983; Rose 1984; Shih 1986: 627) is that the process approach is particularly useful in the genre of personal narration or descriptions. While many of the prewriting techniques and strategies are useful for different types of personal writing, brainstorming and free writing cannot be as successfully applied to all academic writing activities, (Shih 1986: 628). According to Rose (1984: 91) the inappropriate use of free writing and brainstorming might cause “more disorder than order, more confusing divergence than clarifying focus”. Shih further contended that if writing students are advised to primarily use their own background knowledge and personal experience in the composing process then the skills of collecting ideas from external sources, interpreting and synthesizing ideas are neglected (Shih 1986: 628). This is true for especially students in the academic field, who seldom receive assignments which require personal experience or opinion. In most tertiary academic courses students are required to complete assignments which involve using lecture notes, text books and other academic materials (Shih 1986: 628). This requires a more abstract type of thinking, something that is far removed from the generic first year writing course as well as the bridging course (FP) in which much of the process-related research was conducted. Horowritz’s critical stance against the absolute usefulness of the process approach claimed by advocates continues when he questioned the activities and behaviours that characterize the beginning of the composing process, the pre-writing activities. He mentioned that some writers “do better by producing an outline first” (1986: 142), implying that other writers do not necessarily always need an outline. He also noted that the writing requirements for some academic courses are very prescriptive and often include a rigid structure of points to be included in the essay answer, which leave the writers no choice about different techniques and strategies to include the relevant content. His conclusion about this point is that it is not clear whether the writing process approach prepares students adequately for these tasks (Horowitz 1986: 142). According to Rodrigues, (1985: 26) the pure application of the process approach is “artificial”, because students need structure, models, instruction with regard to mechanical skills, time to generate ideas, write and revise and consider purpose and 39 audience. These issues need to be explicitly taught and not left for students to figure out, especially when students enter a new academic environment with requirements regarding a “rhetorical community” (Purves 1986: 39) in a specific discipline. Swales (1986) feels that the process approach does not prepare the students adequately for the expectations regarding writing in reality where the audience is conversant in the genre of his/her respective discipline (Horowitz 1986: 446). In the same vein, Horowitz also claims that the process approach does not take into account the external factors that cannot be manipulated by the writer “which define, shape and ultimately judge a piece of writing” (Horowitz 1986: 446). Horowitz (1986: 143) also believes that a pure application of the process approach is “too soft” in the sense that students get false expectations when their work is peer-edited. Although he believes in the value of peer-editing as students can learn from each other, he also states that the reality of academia does not work that way. The humanistic approach does not do justice to the rigor of examination marking of essays, where markers focus on the product and not the student. Badger and White (2000: 157) remarked that teachers using the process approach to teach writing, in trying to be humanistic and student-centered, fail to give enough input regarding linguistic aspects, different types of texts (genres) and purposes for writing. Horowitz (1986: 143) argued against a pure use of the process approach because it colours the application negatively by suggesting that any other approach is wrong (Caudery 1995: [2]). Horowitz also commented that the essence of the process approach has been reduced by its confrontational slogan: “Process, not product” (Caudery [2]). 2.3.4 Research: process approach in Application There are three research methodologies relating to research of the process approach to writing. One focuses on the actual process that writing learners use when composing. This means researchers gave learners topics to write about and then they observed how the students went about composing. Learners also kept journals to record their composing process. And their ‘think aloud protocols’ during the process of composing were recorded on a tape recorder. Another examines whether the process approach has affected students’ writing abilities in a positive way. These studies determine whether students would have improved writing skills and consequently have better marks when using the process approach to writing. A third, more unfamiliar methodology, is that of determining “the ease or difficulty of implementing this non-traditional approach in the situations in which individual 40 teachers find themselves” (Pennington & Cheung 1995: 22). The focus here is on how teachers cope in different circumstances when actually applying the process approach to writing. The teaching contexts of the studies reported on here differ in terms of geographical location and language background of research subjects as well as teachers and/or researchers. All the studies reported on here refer to teaching contexts outside of Africa. Most studies were conducted in the United States of America (USA), a mainly English first language speaking country. One study reported on here was done in Hong Kong, where English is presumably an additional language, like in Namibia. The subjects of the studies described in this section are mainly from cities, whereas my study is in the rural areas. Most students in Namibian schools and tertiary institutions use English as their additional language even though it is the medium of instruction. Unfortunately, no Namibian research could be found involving the application of the process approach. 2.3.4.1 Sample studies based on the actual composing process Zamel (1983), Flower and Hayes (1981) and Perl (1980) did research on how students actually compose using some or all process approach techniques proposed as beneficial for essay writing activities. Brief summaries of two other research studies into the composing process with regard to the process approach are presented here, followed by a more comprehensive description of Pianko’s research study on the effect of process approach instruction on students’ writing strategies and techniques. Zamel (1982: 199-201) did a study at the University of Massachusetts, Boston with eight ESL students. He described a case-study that investigated the extent of attention spent on correct language and form, how students generated ideas, what happened after students had recorded the ideas and what their writing experiences were. He observed eight proficient ESL students. The experiences and writing behaviours of these students were recorded. He found that students benefited from classroom discussions about the writing topic, reading and note taking strategies to generate ideas, notes and outlines, thinking and re-thinking of ideas, using a great deal of time to write and re-reading and changing if dissatisfaction is felt with the text. They felt that writing was ‘painful’ but had a great sense of accomplishment when they finished and felt satisfied with their efforts. Urzua (1987) did a study in the southwest of the United States of America to determine what effect certain process approach instructions had on the children’s composing strategies and 41 secondly “the ways in which the children used the processes to write more effectively” (281). She used four children from two different schools and tested how children were prepared to help each other, and what effect having the knowledge of an audience will have on their writing. She found that the children helped each other in the revising process when they read their pieces to each other, gave peer responses and feedback, and drafted and re-drafted until they got tired of the project and started a new one. What culminated from that was a sense of audience, a sense of voice and a sense of power in language. The main conclusion of this study is that revision done in different ways and also by different people has benefits for the composing student and results in improvement in certain areas like manipulating language to suit the audience and the purpose of the text. While an in-depth analysis of the research about the actual composing process is outside the scope of this study, the study reported on by Pianko (1979) will serve to represent this research. The information above indicates that research into the composing process shows that the process is non-linear, recursive and complex, and requiring metacognitive skills and behaviour, but a study by Pianko (1979) revealed that college freshman writers seemed less concerned with some of the issues regarding the process approach. Pianko (1979: 6-7) selected 24 students from a freshman composition course at a community college in rural New Jersey. She selected the students based on 3 categories: class status (traditional versus remedial), age (under 21 versus above 21), and gender (male versus female). She managed to get 12 students above 21 and 12 below 21, 12 traditional college freshmen and 12 remedial, and 12 males and 12 females. Seven students dropped out during the course and the remaining subject population was from an economically mixed group. The research lasted for 5 weeks and students were given one writing assignment of 400 words each every week. They received the assignment in the morning and it had to be completed by the same afternoon. They were allowed to take as much time as necessary within the parameters of the same day of receiving the assignment. The assignments ranged from narratives to description, argumentation and exposition. The studies showed that even if students got expository or argumentative topics, most students (55%) wrote in the narrative. The researcher observed and video-taped each student at least once and the students were interviewed on their composing behavior after each writing episode. The questions covered topics based on the students’ writing behavior, and their feelings and attitudes during the writing session to determine their level of self-awareness and reasons and causes for the individual writing behaviours. The first surprising finding was that students did not make use 42 of the maximum time-limit they were given to compose the essay. Despite the time that students had to write the essay, they seemed to be unmotivated to commit. Students spent less than two minutes on pre-writing activities (Pianko 1979: 7), their composing time was 41.61 minutes for males and 35.75 minutes for females. The re-reading activity took them on average 3 minutes as did the revision activity. Pauses lasted on average 16 minutes (Pianko 1979: 7). The reasons could relate to the fact that students had to remain in the class while composing; they were not allowed to finish the essay at home. This was done for better control measures by the instructor. According to the students they did not have enough time “to regroup their energies and thoughts, therefore they merely attempted to complete the assignment in some expedient fashion and ‘give the teachers what they want’” (Pianko 1979: 9-10). This dilemma had two sides: if instructors allowed students to write at home, friends or family members might have been contracted for help and instructors would not really know whether the final product reflects the students’ real ability. On the other hand, it would be beneficial if students were allowed to complete the majority of the writing at home in their own environment with a more relaxed atmosphere and less pressure. A second contradiction relates to the fact that not all students need to write a detailed outline before attempting the first draft. In Pianko’s study (1979: 9), students started writing without having an outline and without a clear idea of what they would write about. On the contrary, it seemed that the most planning occurred while students were writing and it happened mentally. Thirdly, Pianko found that poor writers’ experience of the writing process seems to be incomplete in the sense that they get instruction on grammar and usage, but not much on the actual composing process. They basically do not know how to compose, they know very little about re-reading and revising in terms of content and stylistic issues. They lack the ability to reflect on what was written before. If poor writers do attempt any of the strategies that are used in the re-reading and revising process they do it for a much shorter time than stronger ESL writing students (Pianko 1979: 11). In addition, in Pianko’s study (1979: 10), students chose to write only one draft. This stands in contrast to the one feature of the process approach: writing multiple drafts. Students cited having to write the essay in class resulted in writing only one draft, and that they would sometimes write more drafts if allowed to write at home. The word number limit was also a contentious matter and students would often stop to count words to measure how much more 43 they would have to write. If they were close to the allocated word number they would then write one or two more sentences to conclude the essay. To summarize these observations by Pianko, it seems that when students were requested to write the essay in class time, they did not go about the process as critically and reflectively as expected. They claimed that they would have been more critical and reflective if they were able to write the essay at home. Students indicated that at home they would have been able to seek out diversion after a certain time of composing and then return to the essay again at a later stage to continue. It seems therefore that having to write the essay at school was an inhibiting factor relating to the composition process. On the other hand, Zamel (1983: 181) commented on the effectiveness of the process approach by linking the effects to certain beneficial outcomes. Even though the study did not gauge the improvement of students’ writing skills per se, he states that through his study of and research into the composing process he found the process approach allows students to explore ideas based on a topic, to solve problems as they proceed in the process of writing, to evaluate and value the content which gives meaning to their own writing, to incorporate their own voice in their text, and to have a dynamic relationship between writer text and reader. These features aim at making writing students more autonomous, confident writers, an aim of most writing teachers. 2.3.4.2 Studies based on the effectiveness in improving the writing skills based on the process approach Ho (2006: 4) reported on a number of studies focusing on the process approach to writing. She mentioned that Goldstein and Carr (1996) conducted a study in America based on the 1992 NAEP writing assessment which involved about 7000 grade 4 learners, 11 000 grade 8 learners and 11 500 grade 12 learners. The findings of the study indicated a strong relationship between the process approach to writing and writing proficiency. Another study reported on by Ho (2006: 4) refers to the findings of Jacob and Talshir (1998). The study was conducted in the grade 4 and grade 6 classes of the Pisgat Ze’ev Bet School in Israel. The process writing approach was adopted and the findings showed that the learners “developed into active, independent writers” (Ho 2006: 4). Mahon and Yau (1992) also implemented the process approach to writing in a primary school class of thirty five learners, who have improved their writing ability as a result. Pianko’s 44 study (1979: 21) reported on the composing process and behaviour of college freshmen writers and although she did not mention any improvement in students’ writing grades, she conveyed the message that a “definite, prescriptive writing programme [can] enrich students’ composing processes”. A 6-month observational study was conducted by Urzua in 1987 in the northwest of the USA to determine the effect of the process approach to ESL writing. It was a small scale study involving two Grade 4 learners and two Grade 6 learners. Urzua intended to find out whether “aspects of process writing instruction which had been demonstrated by Graves (1983) to be efficacious with children and by Flower (1981) to be effective with adults” (Urzua 1987: 281) would be effective with the 4 ESL writers. In addition, they also set out to find how the learners used the process approach to write more effectively. Urzua was particularly interested in determining how the learners might help each other and “what effect having an audience for their written work might have on their reading and writing” (Urzua 281-282). Their writing was not restricted to class time only and learners did revising, peer response and feedback. Learners also did dialogue journals in which they could write anything that interested them, and Urzua’s co-researcher read the journals and sent them back in time for the following week’s entry. The data indicated a development in the learners’ writing in terms of the following: sense of audience which was linked to the peer response and to the co-researcher’s response of the dialogues; the revision process was developed also as a result of the peer response activity; and meaningful revision during which learners added information and changed the sentence structure (Urzua 1987: 283-288). A second aspect that was influenced by the process approach to writing was the learners’ sense of voice, which relates to how an idea or a concept should be said “to an audience of another culture” (Urzua 289). Voice indicates the uniqueness of each writer, and in the study this became more developed as the learners progressed with their writing and showed more confidence in their own style. The study thirdly showed that the learners’ sense of power in language increased. Through peer discussion and revision of their written texts the learners managed to manipulate the language and change the information as they started feeling more confident to use new language forms and learn new vocabulary to express their ideas more accurately. Tessema (2005) also commented on the usefulness of the process approach when he gave 45 Ethiopian advanced English students problem-based writing tasks. Tessema introduced a writing project to his students which proved to be successful in terms of motivation to write, adding confidence about the students’ own writing abilities, a higher level of independence in 45 writing activities, authentic audience and purpose for writing, helping each other and generating ideas in different ways. Tessema’s study at the Alernaya University in Ethiopia indicated that students’ writing abilities did improve and had further positive implications on their next writing projects. 2.3.4.3 Studies based on the ease or difficulty of the implementation of the process approach Pennington and Cheung (1995) addressed the issue of how eight teachers of secondary schools in Hong Kong managed to adopt the process approach in their writing classes. According to Pennington and Cheung (1995: 16) any educational change or innovation affects contextual issues like culture, politics, administration, education, institution and finally the class room, since change does not occur in a vacuum. In addition, personal teaching philosophies and style also have a bearing on how readily change is accepted and implemented. If an intended change is perceived as challenging the existent issues mentioned above, then the agents to adopt and implement the change might be resistant. On the other hand, should the agents perceive the change as beneficial, the adoption and application of the innovation would be accepted more willingly. The study indicated various levels of knowledge and understanding of the process approach, a disparity between teachers’ ideal teaching situation and outcomes and the practical reality in the class room, and a difference in levels of enthusiasm to introduce the process approach. Another observation was the matter of compatibility which indicated a level of difficulty in implementing the process approach due to learners’ “low language proficiency, previous product writing experience, inability or unwillingness to follow new approaches, and concern about public examinations” (Pennington & Cheung 1995: 28). The authors indicated that although these issues were valid the problems were not insurmountable, but that they needed addressing in terms of teacher training in the process approach. Teachers should also be involved in the problem solving process since they know and understand the teaching and learning situation and the constraints the best. The final outcome of the study revealed that the Hong Kong teachers felt the move towards the implementation of the process approach would be beneficial but not without obstacles in terms of syllabus design, available materials, classroom situations and teacher and learner attitudes. 46 2.4 MODELING/IMITATION APPROACH The instinct of imitation is implanted in man from childhood, one difference between him and other animals being that he is the most imitative of living creatures, and through imitation learns his earliest lessons, and no less universal is the pleasure felt in things imitated. (Aristotle 1895: 5) 2.4.1 What is the Modeling/imitation approach? Imitation and modeling were recommended as effective tools in ESL writing instruction (Stolarek 1994: 154). Over the years these two concepts have been named differently by various ESL rhetoric and writing theorists, for instance the Sample Approach (Harwood 2005: [1]), Prose Modeling (Stolarek 1994: 154), traditional formal approach (Coe 1987: 14), model-based tradition (Watson 1982: 6) and then the more common names: imitation approach (Butler 2002; Farmer and Arrington 1993) or the modeling approach. There are subtle differences among the theoretical characteristics and parameters and the application of the modeling/imitation approach based on the perspective(s) of the authors, but in general the understanding is that model texts are used as input to acquire and/or improve ESL writing skills. Farmer and Arrington (1993: 13) stated that imitation in ESL writing is an intangible concept since there are so many varying meanings attached to imitation in the area of writing as well as other fields of inquiry, like Psychology. Their definition however, serves the purpose of explaining how imitation is viewed in the context of this current study: “Imitation is the approximation, whether conscious or unconscious, of exemplary models, whether textual, behavioural or human, for the expressed goal of improved student writing” (Farmer & Arrington 1993: 13). Bender (1993) reported extensively on the history of imitation as a tool to learn effective writing. The principle of emulation in education and other sectors of human behaviour has been recognized and applied since ancient Rome (Bender 1993: 120; Watson 1982: 5; Jones & Freeman 2003: 169). Flanigan (1980: 211) explained that rhetoricians have used modeling and imitation “as far back as 3,000 B.C. when the Sumerians had students keep ‘copybooks’ to imitate works central to their culture”. He mentioned Cicero and Quintilian as strong supporters of imitation and modeling to consummate content, form and style. Farmer and 47 Arrington (1993: 12) describe the influence of well-known authorities in the field of imitation on different educational areas, for instance Plato, Erasmus and Blair. In addition, Irmscher (1976) and Winterowd (1975) have been reported to perceive models as an integral part of writing instruction of especially novice writers (Flanigan 1980: 211). The pivotal position of imitation as tool in education and specifically writing instruction has undergone various changes as a result of the inevitable evolutionary character in philosophies, theories, applications and evaluations “In the last few decades, however, we have witnessed dramatic changes in how we look upon imitation – changes largely influenced, we think, by the ‘process movement’, with its various emphases on invention and revision, expression and discovery, cognition and collaboration” (Farmer & Arrington 1993: 12). Stolarek (1994: 155) and Flanigan (1980: 211) reported that the use of imitation in writing classes has never ceased: “The use of models has not decreased by any means today.” Flanigan (1980: 211) further explained that curricula and books are written in such a way that novice writers have models which they can use to improve their writing skills. Watson (1982: 5) confirmed this perspective even though she does it with a sense of skepticism: “it is still assumed that the study and imitation of a model, a sample of writing that is by definition successful, is a valid means of helping students to learn to write in their first or second language”. Other scholars (Butler 2002; Bender 1993) believe that the imitation and modeling approach, or certain aspects of it, has to be re-introduced. The word “re-introduced” implies that the imitation approach has been abandoned for another writing instruction approach, the process approach. Nevertheless, it is interesting to see that the debate is still going on and history is still in the making. Knowing that debates create more questions encourages us to look at whether and why a technique or an approach might be useful in our own ESL writing instruction situations. This has indeed happened in the ESL writing instruction realm and many questions were asked about the imitation/modeling approach. Questions are usually linked to analysis, examination, assessment and evaluation which results in either approval or objections. Scholars analyzed and examined the imitation/modeling approach to teaching ESL writing and a number of these scholars found objections against this approach, whereas others still believe in the beneficial nature of the modeling/imitation approach. 48 Most of the scholars quoted here (Bender 1993; Farmer & Arrington 1993; Flanigan 1980; Harwood 2005; Kim & Kim 2005) seem to agree on the beneficial value of the imitation approach. They, nevertheless, reported on other scholars who had refuted the positive claims of the imitation/modeling approach as a useful ESL writing instruction strategy (see 2.4.3). 2.4.2 Features of the modeling/imitation approach Jones and Freeman (2003: 169) attest to the useful position of imitation and modeling in the general learning process in informal or educational contexts. Bender’s key point regarding imitation supports the view of the usefulness of imitation in writing practices: “I wish to argue that certain ‘ancient’ writing practices can enhance a widely recognized need to make writing instruction truly democratic, aware of and responsive to culturally diverse audiences within our national life” (Bender 1993: 108). Bender argued further that writing as part of language training using imitation as tool can be real discovery writing when discussions in classrooms about contemporary issues like marriage, fame, the “generation gap”, or other issues have taken place (Bender 1993: 110). This view strengthens the notion of flexibility and that imitation and modeling as tools do not restrict the teacher and learners to write about a teacher-initiated topic only. Speaking and reading activities before, during and after the actual writing are encouraged by Bender (1993: 110) and Spencer (1982-1983: 43). Bender’s (1993) argument focuses on oral discussions about content, whereas Spencer explained how reading the same narrative in four different styles exemplified the meaning of style. In addition, Harwood (2005) commented that, what he calls ‘The Sample Approach’, is economic and efficient especially in situations where a number of different essay genres need to be written over a short period of instruction time by inexperienced writers. Watson (1982: 6) offered a qualified case for modeling/imitation in ESL writing. She does not deny the value of models as strong “input”, but she questions the extent to which models can become “intake”. Her justification of models also refers to the choice of an effective model, using model-based tasks and exercises and using the product within the process approach. This position counters the fears of the critics who perceive that there is very little flexibility (Kim & Kim 2005: [3]) in the modeling/imitation approach. In fact, writing instructors can decide when and how they want to introduce the model and for what purposes. Models do not necessarily have to be placed at the beginning of the teaching and learning sequence. They can be used if and when the need arises and for the purposes that students need the models. Butler (2002: [2]) and Gorrell (1987: 53) commented on the usefulness of the imitation and modeling techniques: both feel that imitation and modeling is particularly 49 useful for beginning writers who do not have enough experience or skill regarding the use of words, sentence structure, the structure of paragraphs and entire essays in general. Therefore the reading of examples and models is widely encouraged among imitation/model advocates. Reading various examples and models often precedes discussions in the classroom, as Butler explained: “I found that these sample ‘themes’ helped me give form to my ideas, to construct essays that were organized, clear, and coherent” (Butler 2002: [1]). According to Ferris and Hedgcock (1998: 34) and Abe (2008: 14) reading models and examples enhances the writing ability of students since it is difficult for additional language learners to acquire and/or improve the writing skills by writing only. This perception is challenged by Watson’s remark that we do not know how much “input” (reading) becomes “intake” (effective application), since we have inadequate knowledge of how the transfer between input and intake works. On the one hand, Watson (1982: 6) expressed her doubts about models: “How far can study and analysis of these products strengthen students’ understanding of how good writing is actually made, let alone help them to produce some for themselves?” On the other hand, she recognized the value of models as examples by stating that writing teachers and students can use models as a “resource and support, stimulus and guidance, experience and reassurance” (Watson 1982: 6). But, Watson answered her own questions by explaining that if models and examples are used creatively and analytically at various stages throughout the writing process, they can prove useful to improve students’ writing (Watson 1982: 7). In her view, the positive impact that reading has on writing cannot merely be dismissed. The positive link between reading and writing in the writing instruction class is emphasized by Ferris and Hedgcock (2005); Gay (1921); Watson (1982); Jones and Freeman (2003) and Farmer and Arrington (1993). Bender (1993: 111) agreed that extensive reading of a variety of models allows writing students to “activate the writer’s own personal interests and emotions”, and Butler believes that form and content are inseparable (2002: [2]), which allows the writer to acknowledge that time, audience, and place are reflected in writing. Bender emphasized the importance of reading with regard to writing and that imitation can enhance “social awareness and formal technique” (1993: 123). Shih (1986: 622) agreed that especially nonnative writers benefit from using models that show rhetorical conventions, display principles of organizing ideas in academic discourse and illustrate patterns in certain genres of academic essays. 50 Flanigan (1980: 212) implied that when students read models and examples they move backward in terms of determining how the writer arrived at the product, identifying effective features and techniques and considering how the writer manipulated content to form a meaningful, effective essay. This leads to the choice of model texts as an illustrative tool in the modeling/imitation approach. According to Butler (2002: [3]) “composition teachers [should] teach students what we expect them to write”. Writing instructors at any level of education should avoid assuming that students know what to write and how to write, but they should also bear in mind that models might lead students to merely “ape” the model and become “stuck with nothing but the ability to ape” as warned by Flanigan (1980: 214). Writing instructors should always keep in mind that the model is a “bridge to uncovering the individual writing talents in each person” (Flanigan 1980: 214), and therefore an over-reliance on models is not recommended. Illustrating the writing expectations by means of examples is a valuable strategy to encourage students to write effective essays. It is therefore crucial to select a model that is directed at achieving the aim of the writing activity. The choice of model essays depends on factors such as the objectives of the course, the interest, culture and educational level of the students and the accessibility to resources. It is preferable that instructors choose authentic models rather than write texts to use as examples, but in some cases the latter has its advantages as well (Watson 1982: 7). The criteria for the model texts (whether authentic or specially written) should be of such a nature that the texts illustrate the thematic content as input for ideas, rhetorical features, structure, and linguistic aspects (lexical and syntactic considerations). It is often difficult to find a text which explicates all the requirements to the same extent, and then a text that fits the objectives of the genre closest should be chosen. Students can then identify that gap in the illustrated requirements of the text. The purpose of showing the model is that students become familiar with how a certain text is written in a specific genre, for a specific purpose, preferably a text that is realistic in nature, something which they will be required to do in their further studies (in the case of this study) or in the working world. It is important to choose models which exemplify the “communicative functions” (Watson 1982: 7) prescribed by the objectives of the course. It is said to be even more desirable to have more than one model on one theme or genre to expose students to multiple choices which might lead to a deeper understanding of the concept in question or the rhetoric, structure and linguistic quality and features (Bender 1993: 119). Hudson (1991: 87) proposed criteria when choosing model texts: the materials must be current and 51 contemporary; the materials should be relevant to students’ background knowledge; the length of the text should be realistic; the language usage and grammar should be suited for the purpose of the objectives; students should be able to follow arguments; and subtitles, divisions, graphs, tables and illustrations are beneficial but not obligatory. Analyzing examples and model texts: Reading of models and examples should be an active activity and therefore analyzing the characteristics and features of texts is encouraged (Bender 1993; Butler 2002; Connors 2000; Delpit 1995; Farmer & Arrington 1993). If students read and analyze models, they are exposed to content, genre, style, lexical, syntactic and structural features which all contribute to the cohesion and coherence of a specific text (Watson 1982; Jones & Freeman 2003; Swain 2002). Students can see why a specific text is effective in its communication of meaning and use the features as objectives to write parallel essays. Analyzing models encourages the student writers to become analytical and critical when reading texts. In addition, if these activities encourage students to become analytical and critical of their own writing, they will become more independent learners. The success of the analysis activity partly lies in how effective the model is which the instructor has chosen and to which extent the model relates to students’ background knowledge about the concept, their environment, culture and socio-economic experiences. The analysis of a text should be instructor facilitated at the onset of the writing course, because it is very possible that students do not know how to analyze a text. Students need to be prompted and can discuss the prompts in pairs or in groups as a collaborative activity. After a few instructor facilitated analyses, the students should know what is expected of them and should be able to analyze the text independently. The focus on form, style and rhetoric is frequently referred to in literature about imitation and modeling. Flanigan (1980: 215) offered constructive advice about this issue. He explained that models should be used as problems arise among student writing. The model should be directly related to the students’ social or intellectual problems which need solving. Then the model would be used as a dynamic, actual tool to solve writing problems. This would then mean that student motivation and personal input are not necessarily negated by the modeling/imitation approach. In fact the interaction between the model and the student becomes purposeful and intentional. Rodrigues (1985: 26-27) challenged the arguments against focus on form: “[Students] need structure, they need models to practice, they need to improve even mechanical skills,…”. 52 Flanigan concurred by saying that models in the pre-writing stage help writing students to discover ideas, and the active interaction between writer and reader (writing student) “unlocks the creative process” (Butler 2002: [6]) when students manipulate the text. According to Coe (1987: 20) form has generative power only if it encourages an information search, but it limits the exploration of more information that does not fit the form. The focus on form provides a type of rhetorical apprenticeship to inexperienced writers and it is believed to help students to find effective ways to express themselves in a structured way. Coe agrees with this standpoint: … the traditional formal approach essentially taught good form. It answered, formally, the question, ‘What is good writing?’ Because it radically dichotomized form from ‘content’, its answer emphasized structure, even the proper structures for term papers, business letters, resumes – all that Winterowd calls ‘dismal stuff’. (Coe 1987: 14). Even though Winterowd (1975: 163) showed his aversion to form he has, however, also acknowledged the importance of form,”…the concept of form in discourse is utterly fascinating, for it concerns the way in which the mind perceives infinitely complex relationships. The way, indeed, in which the mind constructs discourse” (Coe 1987: 13). Purves (1986) asserted that the process of learning and teaching is “acculturation”, when the students are lead to the “interpretive community” of any discipline, which in turn has its own “rhetorical community” with certain distinctive conventions and requirements (Horowitz 1986: 446). In essence this means that each discipline, including academic writing in ESL classrooms, has a unique rhetorical community. The features and characteristics of these unique rhetorical communities are meant to be adopted and “absorbed” by students of that community by exposure to what to write and how to write. The focus on form, style and rhetoric may result in improvement in their writing. Coe (1987: 16) urged that “form” should not be seen as the evil opponent of “content”. Form stands in relation to content as form gives meaning to content in terms of context and without form content becomes “unknowable” (Coe 1987: 16). Based on Richard’s explanation in Coe, a lack of form in ESL writing might lead to a breakdown in understanding meaning and intention. We have a shared “schema” of how texts look (poems, songs, dialogues, abstracts, narrative essays, newspaper reports, etc.), and these schema contribute to the effectiveness of meaningtransfer. Coe (1987: 17- 19) made a very strong case for the relevance of attention to form by stating that knowledge of formal patterns enables students to communicate more effectively, 53 whether they read (models: receptive) or write (their own essays: productive). He contended that when writers and readers use the same form, understanding is a given. The generative characteristic of form allows student writers to look for information, to be creative and communicative and to fulfill the social purpose of the specific essay that is dealt with at a certain stage. Coe argued for two implications of the emphasis on form in the imitation/modeling approach: form should be explained in relation to function, and as the world develops and new writing needs arise, new forms should be invented (Coe 1987: 21). Form, rhetoric and style and the new discourse family cannot be divorced: students have to be conversant in the writing genre that is expected of them to succeed academically in a specific discourse family. Therefore focus on form should not be regarded as unworthy. Imitation also acknowledges the impact of our environment, our culture, our parents and our childhood experiences on our writing efforts (Bender 1993: 113). Bender called these issues ‘life’ models that enter into the classroom subconsciously and states that they have a crucial role in our writing attempts (Bender 113). In practice this would mean that we have internalized models about many issues and concepts in life and we measure each new experience with what we have known before. If the experience was positive or if the knowledge is beneficial, then we model it in order to get a positive outcome in a new experience. If the experience was negative we seem to want to avoid it in future, but we still have a model which we use to measure new experiences against. Learning effective writing skills may be based on the same principles. If we show students a variety of models and discuss the content and identify why the models are effective, students should have access to “rhetorical and moral multiplicities” as well as “linguistic and cultural diversity” (Bender 1993: 115). Quintilian commented on the advantageous nature of imitating a successful invention (in Bender 1993: 119): “And it is a universal rule of life that we should wish to copy what we approve in others”. According to Bender an imitator is actually a synthesizer who constructs new discourse based on information and structures from examples and models, who adapts and/or adopts features of the original text (1993: 119; Watson 1982: 7). He explained that Quintilian compared the imitative writing process with the digestive process, where eating is compared to reading different types of texts, chewing is related to thinking about all the aspects of the text and getting familiar with those aspects. An important part in the next stage is what Quintilian calls “dialysis: rhetorical nutriments are split from unwanted detritus, and as this separation continues, the model disappears, becoming a set of stylistic and intellectual 54 materials from which the new writer’s commitments, interests and stylistic preferences will be written” (Bender 1993: 121). Pincas (1982a: 22) agreed with this notion by stating that when students reach the stage of independent writing their initiative is used by them to generate a text that is eventually their own. This ultimate aim of the imitation writer, to become an independent and individual writer with the help of models and examples, is underscored by Butler: “Using the writing of others to teach writing can yield effective and long-lasting benefits that give students the confidence to write convincingly on their own” (Butler 2002: 3). 2.4.3 Arguments against the modeling/imitation approach According to a number of scholars imitation and modeling have been relegated to teaching techniques that belong to the past (Bender 1993; Butler 2002; Farmer & Arrington 1993). The reasons for this occurrence range from the perception that imitation equals plagiarism (Jones & Freeman 2003: 168) to the view that imitation stifles the inventory nature of the writing process or, that students produce “artificial” (Watson 1982: 6) essays. Butler (2002: [2]) and Bender (1993: 109) alluded to the fact that imitation and modeling are product approaches, which was viewed as an undesirable approach in writing instruction. Gruber (1977: 496) commented on the effect that the imitation approach to teaching writing might have on the students’ opinion of themselves. He mentioned that imitation might result in a feeling of initial suspicion and inferiority among students, as imitation might be a totally unfamiliar approach to them. He went as far as to say that to the inexperienced and less confident writer, the imitation approach might seem self-destructive if students believe that they are not good enough to produce an original piece of writing without the help of models. Therefore the use of imitation and modeling was ceded in favour of other techniques that were deemed more useful and in the process approach which followed the product approach. Critics comment on the fact that imitation and modeling is static and unthoughtful, artificial, it lacks originality, it is mindless, repetitive, and anti-intellectual (Flanigan 1980; Harwood 2005; Kim & Kim 2005). Models are also viewed as a mental aid used by those writers who have a mind that is “careless, unexacting, and unable to manage complex systems of abstract representation” (Pemberton 1993: 40). Using models to teach writing is often perceived as an inferior modus operandi and does not result in true scientific thinking (Pemberton 1993: 40). The implication is that the approach is said not to address the cognitive and meta-cognitive processes that occur during the writing process, and according to the critics it stands in opposition to the process approach which encourages originality, individuality, and invention. 55 According to Bender (1993: 109) “Imitation is stigmatized by the sign of its origin. An ‘ancient’ kind of composition instruction, imitation subordinates individual talent to the unsavory goal of standardized, hence derivative, fluency.” Imitation as part of the product approach is further criticized for its focus on one draft only, constant error correction and very little student motivation (Yan 2005: 19). In addition, student involvement is relegated to attention to superficial form and syntax and therefore rhetorical drills (Flanigan 1980: 214; Nordin & Mohammed 2006: 76). Nordin and Mohammend view this approach as teacher centered “as the teacher becomes the arbiter of the models used” (2006: 76). The teacher initiates the writing, so writing is not viewed as a “need” by students to say something that relates to their problems, ideas or concerns in a form and structure that they have found to suit the content of their writing. Rather students are given forms, structures and linguistic rhetoric which have to be used to convey their ideas. This is viewed negatively by Flanigan (1980: 214) as he believes that the actual purpose for writing and the process of problem solving are negated by the product approach. He mentions that the product approach and therefore imitation is like “playing with a puzzle or performing an exercise” (Flanigan 214). The focus on form, style and rhetoric is frequently referred to in literature about imitation and modeling. Critics disapprove of the focus on form, stating that the emphasis on grammatical accuracy, and structure negates the importance of content and the ability to invent, to create something new and unique (Flanigan 1980; Harwood 2005; Kim & Kim 2005; Pincas 1962). Shih (1986: 622) contended that requiring students to start the writing process from given patterns that need to get topics and content is to reverse the normal writing process. Then the importance of purpose and audience and the inventive and generative nature of writing are disregarded. Similarly, Pincas and Flanigan questioned the place for student motivation and the search for truth and real experience in the modeling/imitation approach if the focus is on form. Flanigan (1980: 214) also reported that the focus on form overlooks the fact that students write better when they write about a ‘felt need’, their concerns or interests. When students have to ‘solve a real problem’, writing becomes purposeful and has a true, realistic intent. But with the imitation/modeling approach students are merely “figuring out its parts or shapes” (Flanigan 1980: 214) to solve the problem, like building a puzzle. Bender (1993: 109) also reported an objection to the focus on form by stating: “…classical writing pedagogy is seen as formalism, and organizing schemes, rules and formulas. Imitation is seen to commemorate ideal structures and desired consummations, which, frankly, do not guide students in their inner exploration of possible meanings.” In the same vein, Spack reported 56 that Horowitz’s limited survey of real university academic writing assignments that students needed to complete, indicated that students were not asked to “start form patterns and produce essays to fit them” (1988: 31). Boyd (2006: [2]) provided a totally different perspective as argument against imitation. He comments that university lecturers assume that students want to be part of a discourse family that is based on the language that is used in academia. And, if students through imitation successfully manage to enter the privileged discourse family, they can write from a position of power and knowledge. Boyd questioned that assumption: do students really want to be part of a distinguished discourse family or do they have other motives for writing academically well? He believes that students come to university to get a degree which would result in an economically profitable job for them. Students perceive their writing instructor to be the person who has to provide them with skills to reach their goal (a degree which leads to a job). When students do not get the results they had hoped for in their writing assignments, they do not complain that they have not been assimilated into a new discourse community. Instead, they complain because the writing instructor has failed to supply them with the skills to reach their pragmatic goal. In addition, the role of the teacher as model to be imitated was also drawn into question by Boyd. He criticized the claim that students want to be like their instructor, even take on their identity, when imitation is used as teaching strategy. The problem with that is that the instructor assumes the role of power and authority and that the students’ identity is subsumed and surrendered to be replaced by that of the desired model (Boyd [3-4], Sullivan 1989: 18). Sullivan also commented negatively on the instructors’ use of a text as model to be imitated. He mentioned that if writing instructors recommend models as examples they actually demote their ability to teach students how to write and with that admit that they do not know every aspect of the writing process (Sullivan 1989: 18). To conclude the arguments against modeling/imitation, Sullivan’s conviction must be mentioned: “They, or perhaps I should say “we”, came to think that our primary goal was to help students find their own voices, and so it seemed ill-advised to recommend that they imitate others, a practice that obviously produces conformity” (Sullivan 1989: 17). 2.4.4 Research: modeling/imitation approach in Application Stolarek (1994: 155), Bender (1993:118), and Flanigan (1980: 213) reported that there exists a distinct lack of accurate recent research on the use of modeling and imitation as a technique to improve writing skills. In the Namibian context there is no existing research of which I am 57 aware. According to Flanigan and Stolarek, referring to studies conducted by Mills (1967) and Wheeler (1978), the research that was indeed done on the usefulness of models as examples, seemed to be of a confusing, inconclusive or questionable nature. Very few post2000 texts on the application of the modeling/imitation approach are available. This occurrence is probably as a result of the emergence of more eclectic approaches in teaching writing. Stolarek conducted her own study at Ferris State University, Michigan, USA, to determine whether prose modeling would have a positive effect on the writing skills of expert and inexperienced writers. Her findings illustrate that the students who more consistently used the model examples also showed most improvement in their meta-cognitive skills and were “most successful in completing the study” (Stolarek 1994: 157). She requested 143 college freshmen (all volunteers) and 21 university composition instructors to produce a text based on an unfamiliar prose form, the “modified chosisme”. The students were categorized into five groups. These groups received five different instructions for producing their own writing, as illustrated below:  Group 1: Description Only  Group 2: Model only  Group 3: Description/Model  Group 4: Model/Explication  Group 5: All Items (Description, Model, Explication). The student essays were marked by two English department graduate students and various tests were done. The outcome was that students who followed the Description Only approach scored more or less the same marks than the students in the Model Only group, whereas the faculty members had a higher score for Description Only than in the Model Only Group. But the students in the Group 3 and 4 scored higher than both Group 1 and 2, while the faculty members of Group 3 scored higher than Group 4. Finally though, students and faculty who received all three items showed the best improvement. An important conclusion was made after the study: conscious imitation of the modeled form by expert as well as inexperienced writers leads to greater effectiveness in writing skills. Another relevant observation, rather than a full-scale study, was made by Spencer (19821983: 42-45) with regard to the relationship between imitation and style. She found that students in her class did not understand the literary concept of style and decided to use the imitation strategy to teach style. She chose the fairytale of “Goldilocks and the Three Bears” 58 and presented it in four different styles: business-like, the dramatic style of Shakespeare, the King James Bible style and a South American dialect style). Spencer reported that after their reading the texts and the analysis of style, students succeeded in understanding and applying the concept style to their own writing. She also used copying sentences from various sources as technique to improve students’ (and her own) writing skills, and found that she and the students benefited from this imitation technique to improve their writing with regard to spelling, vocabulary, sentence structure, fluency and style. Gorrell (1987: 53-59) agrees: “Another form of imitation that is often effective with inexperienced writers is exact sentence imitation”. Gorrell has applied this strategy in her class successfully to teach students how to use sentences structures and punctuation (Gorrell 53-59). Focusing on another aspect of imitation, Brooke (1988: 26) described how a teacher he calls Janet Rich, dealt with the relationship between a writer’s identity and the writing students’ ability to approximate this identity in a quest to improve writing skills. Janet taught an introductory English class at the University of Nebraska at Lincoln in 1986 and provided the students with a model of a writer’s identity that the students could choose to emulate or not. The writer that was modeled was Margaret Laurence, author of A Bird in the House. Janet established a relationship between the identity of the author and the students’ own writing by providing in-class reading and discussion sessions and by giving scaffolded writing activities that were also peer-edited. Three patterns emerged from the study: acceptance, transformation and rejection. The majority of students managed to accept the writer’s identity and model it in their own writing, some students struggled with feeling excited and/or threatened by the author’s identity and had to focus on what excited them. One student could not accept Laurence’s identity at all to help her improve her own writing. The overwhelming outcome however, was that modeling a writer’s identity seemed to be useful in improving students’ writing abilities. This unusual technique of the imitation approach can, in my view, only be used in certain writing situations, for instance when students are aiming to write certain types of narratives. Academic and/or scientific writing is a genre where high value is placed on objectivity and therefore this technique would probably not be suitable. There might be ethical issues as well, as accepting another author’s identity might lead to plagiarism. Charney and Carlson (1995) gave an account of a study that they conducted at The Pennsylvania State University with ninety-five Psychology majors. The aim of the study was to determine how useful models would be when students had to learn to write in a new genre 59 of their discipline (Psychology). The researchers also had as a goal to determine whether graded models have an influence on the students’ writing ability. The students were required to write the Method section of two simple experiments using different types of materials provided by the lecturers. The students were divided into two groups and each group received different information about two topics. One group received models to help with the construction of their own writing, whereas the other group received no models. The group that received models was further divided into two categories: one group received graded models including good, average and poor models, whereas the other group received only good models. The researchers came to the following conclusion: “Taken as a whole, the results indicate that models do not have automatic benefits for the writing process. Likewise, they begin to reveal why models have produced mixed results in previous studies and in classroom practice (Charney & Carlson 1995: 111). However, the researchers succeeded in finding that models do have an impact on certain aspects of students’ writing skills. It was found that reading models helped students remember and write about some concepts that they would otherwise have omitted. The results also showed that models helped with the genre-specific organization of ideas. In addition, the researchers found that labeling and grading the models did not influence the quality of students’ writing. The final conclusion was that “active analysis of a model (even before taking on a specific writing task) may help students construct new textual patterns or enrich the patterns they know” (Charney & Carlson 1995: 114). Secondly, when students were allowed to consult models during the writing process they could determine whether their ideas were relevant in the text they aimed to produce. The following study is very valuable in terms of similarity to my study. Even though the geographical context differs, there are similarities with regard to the purpose of the study (improve laboratory report writing), the subjects, who had English as their second language, and the outcome of the study. This study was conducted by Jones and Freeman in late 2000 and early 2001 at Macquarie University, Sydney, Australia (Jones & Freeman 2003). The study involved 240 students who enrolled for a Computing Science degree and had to take an introductory Physics course for which they had to write laboratory reports. The students were a mixed group, the majority coming from East-Asia, and others from Australia, most had English as a second language, but did not have Physics as a school subject. Students displayed a distinct lack in laboratory report writing skills as observed by their Physics lecturers. Two needs were identified: students needed guidelines about the structure of a 60 laboratory report as well as the language that is appropriate when writing laboratory reports. The lecturers (researchers) provided students with a host of materials including a course guide, course notes, laboratory notes, report writing worksheets and Physics textbooks. Lecturers also modeled ‘drafts of example laboratory reports’ on Overhead Projectors to illustrate the writing process. Students were required to write laboratory reports using the sources they received. The researchers found that students often copied verbatim from the laboratory manuals despite instruction from lecturers not to do so. This was probably done because students lacked the “understanding of the grammar and lexis of the specific register, an appropriate level of complexity, and familiarity with the procedural expressions common to scientific and academic writing” (Jones & Freeman 2003: 174). Students managed to include all the relevant sub-headings in their reports, but nearly half of the group wrote PROCEDURE instead of EQUIPMENT and METHOD in the sections where they had to report how the experiment was conducted. Some students incorrectly copied from the wrong models or left out “important functions from the appropriate sections” (Jones & Freeman 2003: 175). Even though 18% of the students used the imperative mood in the methods section, the rest managed to paraphrase. The researchers found these results to be an eye opener on how to further streamline the materials and instructions to the benefit of students’ writing abilities. Some suggestions on how to improve the writing of laboratory reports were given by the students themselves: they would have liked instruction on laboratory report writing earlier in the course, and they would have appreciated more lecturer input and feedback. The lecturers decided that a more critical analysis of the implementation of models and instruction would be needed. Twomey (2003) completed a study for her Master’s degree with first year students at Virginia Tech in 2002. Her observation was that students entered the university without enough background knowledge on how to write essays and without sufficient strategies to use models as writing aides (Twomey 2003: 38). Her study involved 49 students from diverse educational backgrounds. Twomey’s study aimed to determine whether students’ writing skills would improve if the writing process was modeled to students, and if written products were used as examples. She also taught students adaptive skills that are needed when writing in a new writing process and new written forms. The nature of her study focused on the following: would students improve if they were given opportunities and study models and imitate the models to produce their own texts? 61 Twomey showed students examples of different types of writing appropriate to different situations and discussed these examples. Students did activities based on audience, purpose, and conventions using the examples. Students had to draft, edit and re-write their own texts based on the discussed examples. She exposed them to different types of academic texts and focused on issues like thesis development and genres. These texts were analysed and the writing behaviour was imitated by students when they produced their own essays. She also made use of other imitation techniques like copying sentences and paragraphs and discussing these before attempting to write their own sentences or paragraphs. The results of the study show that the students benefited from the modeling and imitation strategies used in the lessons: six students received an A symbol, twenty seven got B’s, 12 got C’s and only 3 scored D’s (Twomey 2003: 55). Twenty seven students commented positively on their writing experience. The researcher’s observation was that teaching writing with the help of models is better than without models. She felt that “learning to write with models is more productive and less frustrating than trying to meet the writing specific expectations of teachers or employers with only a description or a set of assumptions about the desired product” (Twomey 2003: 68). Unfortunately no mention was made of a pre-test result of the students which would have made the study more reliable. 2.5 PROCESS GENRE APPROACH Such an eclectic approach offers advantages such as a more focused use of texts as models without sacrificing the flexibility to acknowledge elements of other approaches. (Nordin 2006: 75) 2.5.1 What is the process genre approach? The evolutionary character of ESL writing approaches is well-known by ESL researchers and teachers. The rigid adherence to only one writing approach has been questioned by several writing teachers and researchers like Horowitz (1986) and Yan (2005) who looked with new lenses at the benefits and drawbacks of the process, the imitation and the genre approach. As a result, a combination of approaches has been proposed by writing teachers and researchers such as Badger and White (2000), Harwood (2005), Henry and Roseberry (1998), Kim and Kim (2005) as beneficial in improving students’ writing skills at various levels of their education. In this section I explain the process genre approach. Since I have already given elaborate information on the process and the imitation approaches, it is not necessary to repeat their 62 features, benefits and disadvantages here. A brief recount of the genre approach will be presented, but the major focus will be on the process genre approach. In the 1980s the genre approach was hailed as constructive in the ESL writing mainstream based on the perception that the different purposes, social contexts, structures and linguistic features of specific texts should be taken into consideration when teaching writing (Bhatia 1993; Henry & Roseberry 1998; Halliday 1985; Swales 1981; Yan 2005: 19). The context, structure, purpose and linguistic features are reflected in typical texts according to their genre (Nunan 1999: 280) and students can attempt to “approximate” these when writing their own texts (Harwood 2005: 3). According to Kim (2007) emphasis on the reader and the purpose for writing are paramount in the genre approach. As the reader is usually an experienced member of a specific community, albeit academic, technical or in the business field, he/she expects the writing discourse to comply with known, acceptable schemata and writing conventions based on the identifiable genre (Silva 1990: 16-17). Kim (2005) reported that, the genre approach acts as a support mechanism in ESL writing instruction, where examples of a particular genre are provided to students which can be used systematically to meet the expectations of the reader, the linguistic and structural requirements and the communicative purpose of the text. Students’ knowledge of linguistic features and structural conventions of a variety of genres based on their communicative purposes is often very limited (Kaunda & Ball 1998: 130; Swanepoel 1999). Therefore, the writing teacher can play an active role in guiding, assisting and supporting students to advance to the point where they can employ their skills to be conversant in a variety of genres. In comparison to the product approaches, clear similarities can be detected in terms of input, as examples or models play just as an important role as in the imitation approach (Badger & White 2000: 155). The genre approach (just like the imitation approach) assumes that students learn more effectively when exposed to multiple examples of texts (in the genre approach the importance of genre is emphasized more than in the imitation approach). In fact, the theory of learning of the genre approach seems to consist of three parts: imitation, understanding and application (Badger & White 2000: 156). The exposure involves reading and analyzing the texts and determining audiences, communicative purposes, common patterns and features that, when combined, make up a text which can be classified as belonging to a specific genre (Kim 2007). 63 In the genre approach students know exactly what is expected of them since they have received explicit instruction in and examples of the specific genre (Kim 2007). The awareness of the association between content, purpose, audience, style, structure, and language usage will stand students in good stead when encountering a similar writing situation later in their academic career or even the working world. Dudley-Evans (2002: [2]) suggested that “genre knowledge involves increasing awareness of the conventions of writing, and teaching students to produce texts that, by following the conventions, appear well-formed and suitably structured to native-speaker readers. Indeed, it has been argued that knowledge of organization, arrangement, form and genre can systematically lead to knowledge of subject matter.” They can then tap into their rhetorical conventions background knowledge to write a text that is acceptable and effective for its purpose. The genre approach is believed to release the stress and anxiety about writing experienced by especially ESL writers (Kim 2007). The process genre approach is the product of a carefully selected set of features that would address issues like instruction in different genres, example essays, and a focus on linguistic skills to compose more effective essays. In other words, the beneficial parts of the different approaches are preserved, but in essence a new approach is formed, in this case the process genre approach (Kim & Kim 2005: [7]; Nordin 2006: 79; Yan 2005: 20). Badger and White (2000: 158) stated: “An effective methodology for writing needs to incorporate the insights of product, process and genre approaches.” For the sake of practical application it means that the modeling and imitation activities and the analysis and manipulating activities are added to the recursive phases and the linguistic skills and strategies of the process approach (Yan 2005: 20). The genre approach is used to provide the input and strengthen students’ knowledge of genres and the process approach is mainly employed to allow students to use their potential creativity in an extended, recursive procedure to produce effective essays. The justification for using the process genre approach is based on the fact that both approaches (process, genre) have very useful benefits on the one hand, but some limitations on the other hand (Gao 2007: [7]). For instance, Lindemann (1995: 295) portrayed process writers as “isolated individuals … divorced … from the social contexts in which language always operates.” She continued by saying that writers should not always confine the audience mainly to the self, since “expressivist pedagogy had stripped rhetoric of its important cultural, often political, force” (Lindemann 1995: 295). This is a ‘gap’ that can 64 be comfortably filled by the genre approach which acknowledges that language is a tool for and a form of social interaction. On the other hand, the process approach allows students to tap into their own cognitive abilities and use linguistic strategies suitable to their unique writing style to solve problems and write meaningfully (Lindemann 1995: 293). Therefore, it can be derived that the process approach and the genre approach complement each other since the features of both approaches are compatible. If the features of the two approaches are combined then the writing lessons will include the model of the genre, emphasis on purpose and audience of the genre, an awareness of the context to make the activities more realistic as well as linguistic writing skills and the various phases that students might follow over an extended period of time (not just one lesson) to produce an essay. Gao concurred by stating that “…, teachers should raise learners’ awareness of a variety of genres…” (2007: [11]). Since process approach writing emphasizes mainly narrative and self-initiated writing, where students determine their own rhetorical problem and collect their own material, it is not always useful in the academic writing realm (Shih 1986: 627). On the other hand, the genre approach helps students to gain entry to discourse communities, like university classrooms. In academic writing at university level, students are often presented with certain rhetorical problems for which they have to gather information and write an essay that is appropriate for a specific genre. For academic writing purposes and in order to avoid confusion and divergence from the task, it would benefit the students if they receive direct instruction in pre-writing activities which centers their attention on techniques and strategies that would enable students to gather, analyze, synthesize, and interpret information (Shih 1986: 627; Gao 2007: [7]). These pre-writing activities can effectively be applied using the process genre approach. In addition, at university not much personal writing is expected of students. Horowitz (1986: 447) stated that an analysis of writing tasks at university level will “create a new writing syllabus by telling us which tasks our students can expect to encounter during their academic careers”. Students get the opportunity to transfer skills from one subject to another, which makes the English courses more meaningful in the context of their studies. The application of the process genre approach could bring students into contact with the relevant content schemata (genre approach) required of them, and also with linguistic skills to apply the phases of the process approach (Gao 2007; Kim & Kim 2005). Shih (1986: 628) concluded her argument by saying that a practically-oriented process genre approach might strengthen and expand 65 students’ comprehension of the topics in their content subjects, which are often used in writing classes as basis for writing activities. In addition, the content of the writing at university level is at a higher cognitive level than at primary and secondary schools. If the value of the activities in the process genre approach is examined we can find that the process genre approach facilitates cognitive skills of students in the sense that they have to identify, determine, analyse, evaluate and apply, all elements of Bloom’s taxonomy (Atherton: 2011) for effective learning. These skills will in turn assist in problem solving and decision making skills. Furthermore, student motivation may be increased when their writing is linked to the tasks that they are required to do in their content subjects (Shih 1986: 628). The process genre approach fosters active learning and learner-centredness as students are actively involved in the teaching and learning process and the teachers act as facilitators, guides and assistants. 2.5.2 Features of the process genre approach The features of the process genre approach are exhibited in the application or activities and phases involved in the actual process before, during and after composing a text. Writing is constrained by specific social situations: Writing is a tool for communicative discourse, which, in the real world, is hardly ever done without attentiveness to the audience, the purpose, the subject matter or topic, style, appropriate vocabulary and related aspects of the social context of the text to be produced. Ferris and Hedgcock (2005: 48) stated the necessity for L1 and L2 students to have an awareness of “genre knowledge, text comprehension, and production skills”. They also emphasised the role of the writing teacher to expose students to the social discourses and to “meet the demands and challenges of academic institutions” (Ferris & Hedgcock 2005: 48). For instance, an application letter and Curriculum Vitae have a unique set of conventions that will be suitable for the purpose of writing the letter and the CV. Similarly, a laboratory report has a distinctive style, structure, vocabulary and grammatical conventions based on its purpose and audience. In order to get the piece of writing to achieve the purpose, certain writing schemata have to be acknowledged and applied. The form and function nature of the genres is often unfamiliar to inexperienced writers. The writing teacher can use the process genre approach effectively to introduce the students to the writing constraints determined by the contextual nature of the genres. Badger and White (2000: 158) suggested that, “teachers need to replicate the situation as closely as possible and then provide sufficient support for learners to identify the 66 purpose and other aspects of the social context”. The teacher creates a situation suitable to the genre that students are supposed to learn about. In fact, according to Ferris and Hedgcock (2005: 51), ESL composition teachers can help their students become part of “institutional and discipline-specific literacy clubs, discourse, and communities of practice by acquainting learners with the enterprises and literacy practices of people who are already part of these communities”. Models or examples as input: Badger and White (2000: 159) recommended that writing teachers should provide students with “sets of corpora of the kinds of texts their learners want to write”, in other words, they need model or example texts. Kim and Kim (2005: [9]) explained that genre specific texts are “functional planning tools” which assist in the teacher’s role of setting up syllabus goals, materials and the techniques and strategies in the actual classroom to help students to write effectively. The specific conventions based on the chosen genre are usually found in “expert” texts of the specific genre. Therefore, the teacher selects texts which exemplify the conventions in all respects regarding subject matter, purpose, audience and other associated aspects (Gao 2007: [11]). The example texts and types of activities should be pitched at levels which show a scaffold from the known to the unknown, from the simple to the more complex, from understanding to synthesis and evaluation. The example texts should adhere to certain criteria, which were discussed in the Section on Imitation. Kim and Kim (2005: [9]) suggested that the reading material should be selected from a wide range of sources to show that different genres have a useful purpose in various sources. According to Kim and Kim (2005: [9]), this will help students to broaden their general knowledge, expand their vocabulary, and they encounter the “linguistic and semantic features of a language”. The principles and features of the process genre approach are of such a nature that some diversion from the prescribed syllabus can be allowed. Analysis and/or manipulation of models: The students should be actively involved in the appropriation process by completing language awareness activities or activities relating to the structure, grammar, vocabulary or activities which show the relationships between the different parts of a text and the subject matter that is presented in these parts. Kim and Kim (2005: [7]) emphasized the importance of language and structure activities to illustrate to students “the roles they play in conveying appropriate meaning”. The activities should incorporate all the aspects relating to the genre, so that at the end of the activities students would be able to write their own criteria based on their observations, analyses and active involvement. Types of activities would vary depending on the genre that students are busy 67 with, since “different genres require different kinds of knowledge and different sets of skills” (Badger & White 2000: 158). In that way students determine their own objectives and aims, which makes the whole exercise more learner-centred and motivational. Students are trying to reach goals which they have set themselves. Student independence and autonomy is created in a true sense and students manage to enter the desired discourse community feeling that they have accomplished it mostly by themselves, with the teacher as guide and facilitator. Another important point in the presentation and execution of the activities is scaffolding. Scaffolding is the assistance provided by the teachers. It is a temporary strategy (Kim & Kim 2005: [7]) that helps students to reach higher levels of understanding, to learn new skills and to understand unfamiliar concepts. In the process genre approach, teachers assist in various ways from the beginning of the process until the end. The extent of involvement depends on the students’ level of dependence and ability. Some students need more assistance than others, and with close monitoring and control the teacher will be able to recognize how much support is needed. There should be clear parameters in the measure of involvement from the teacher, with minimum and maximum levels. This ties in with Vygotsky’s (1978) Zone of Proximal Development, where the students move from the known and what they can manage independently without the help of the writing teacher to the unknown and challenging position which they can reach with the assistance of the teacher. Consolidation: Students and writing teachers do “joint construction”. A topic based on the specific genre is chosen and students and the teacher write a text together based on all the conventions that were regarded as useful and suitable for the genre. The teacher can model functional techniques to gather ideas, organize the ideas and write a first draft: Learners may also require input about the skills needed for writing. A rich source here comes from observing other students and the teacher. Teachers may find direct instruction on skills effective - but an alternative is a demonstration by the teacher or other skilled writer, possibly accompanied by a commentary attempting to explain the mental processes that underlie the exercise of the skill. (Badger & White 2000:159) Students can follow suit individually, in pairs or in groups until enough key points were gathered to write a meaningful essay. The teacher may be able to determine the students’ level of mastery of the skills and knowledge needed to produce an effective essay. If group work was opted for, students can help each other with the subject 68 content and construction, and the teacher monitors the students’ level of understanding and success in applying the aspects that were explained and exemplified. The writing teacher and students should attempt to complete the whole essay using the features of the process genre approach and emphasizing the recursive nature of the process approach. For instance, after doing an experiment in the English lessons, the students and the writing teacher can write a laboratory report together, with the teacher eliciting writing information from students but also modeling techniques to illustrate how information is found, analyzed, synthesized, and used in the text. Reflection and feedback are also stressed, to teach students to be observant and critical. In this way the academic writing process is ‘demystified’ (Kim & Kim 2005: [8]) for the students and they will become more motivated, secure and confident to complete the writing task effectively. Additionally, the jointly constructed text can serve as another example for students. Stages or phases of the actual writing process: Just like in the process approach, students must firstly acknowledge the recursive nature of the writing process. The categories mentioned below are presented in a linear fashion, but do not necessarily have to be followed in that order. Secondly, some phases overlap, meaning while students are busy with one phase they can concurrently employ skills or activities from another phase. For instance, while students compose their first draft, they can already do rereading and revising. Or, as declared by Zamel, “…planning [for example] is not a unitary stage but a distinctive thinking process which writers use over and over again during composition (1982: 206)”. An additional point is that the process genre approach does not only involve the skill of writing. Speaking, reading, listening are also incorporated in the lessons and according to Yan (2005: 20) the “four language skills promote the expansion of the students’ overall language competence”. Pre-writing phase: After students have been familiarized effectively with the genre and the relating conventions by direct instruction, models and the manipulation of the models, they are ready to create their own text. Students then use the background knowledge about the possible subjects or topics, the linguistic features and linguistic skills to write their own text in the specified genre. Students would now be at a higher level of independence, but close monitoring and assistance is still crucial. 69 What do students actually do in the pre-writing phase? Students would receive or decide on their topic for the essay with the guidance of the teacher. They would discuss issues pertaining to the subject content, the audience, the purpose, the style (formal/informal), and grammar issues like tense, in other words the constraints of the genre. The main part of the pre-writing stage is gathering information from various sources and recording the information in a useful form, like a table for compare and contrast or key arguments in note from for a discussion essay. This stage would take up to several hours, depending on the accessibility of sources and the students’ skills on identifying sources, recording the key points and arranging the points in logical order. The teacher can still provide several techniques to read about a topic, gather ideas, analyze, synthesize and organize these ideas. The teacher’s involvement should be constructive and sensitive. Some students are capable of finding information themselves, whereas others need some help. A certain amount of autonomy and confidence should be demonstrated by students at this stage, and their individual creativity should be encouraged and acknowledged (Gao 2007: [8]). Composing: Composing means to structure the ideas in meaningful sentences based on the conventions of the specific genre. When the students have the topic and the gathered ideas, they do not necessarily know exactly how they will use the information in their independent essays. Shih (1986: 628) illustrated the action of writing the first draft as follows: “…writers take the material previously gathered and organized and structure it into a linear piece of discourse”. In other words, they construct sentences and paragraphs. But, since the students’ ideas are hardly ever completely formulated before they write their first draft, multiple drafts on various levels are to be expected. There are differences in the composing process of the writing students and the process genre approach allows writers to go about the composing task in their unique way. Re-reading and revising: A very powerful observation about revision and re-writing was made by Maimon et al (1982: 61): “successful papers are not written; they are re-written” (Clenton 2005: [3]). Once the first draft is completed or while students are still busy composing (depending on the length of the text and the preference of the students), they are encouraged to re-read their text firstly to 70 determine whether their subject content matches the topic and what they intended to say (Shih 1986: 630). They look at their ideas critically and evaluate the meaning and message; changes or alterations can be made. Students can even add or delete ideas. The second focus is on structure where students evaluate the organization of their text to make it more readerorientated (Shih 1986: 631). Students should check whether their paragraphs have a logical order with clear topic and supporting sentences. After that, another revision technique is to check the grammar used in the text. If the students used sources, then the referencing of these sources should be checked and edited as well, if necessary. Peer-editing: Peer-editing is a text production skill that is characteristically applied in the process approach. Peer-editing means that students read each other’s work, and then offer feedback on content, structure and grammar concerns. This skill can successfully be used in the process genre approach to writing if administered effectively. Peer-editing is also a form of input, as discussion on content with other students might lead to addition of ideas. Students should get guidelines on how to peer-edit. It is always useful to give students the criteria as a guideline in the form of a structured feedback form or checklist to be filled in or ticked off (Gao 2007: [9]). Objectivity must be stressed and the teacher could model a peer-editing session before students embark on editing each other’s work in pairs. After the peer-editing session, students should be allowed more time to re-write the text if necessary. Teacher feedback: Once the first draft is written, self-edited and peer-edited, and revised, possibly re-written, the teacher is responsible for editing and evaluation. After having read the essay, the teacher can use one or more methods of feedback. A very useful but time-consuming method is to go through the essay together with each student, asking questions and making suggestions in a positive, motivating way. This is called teacher-student conferencing (Kim & Kim 2005: [9]; Nordin 2006: 81; Gao 2007: [13]). This technique can give the teacher insight into the students’ level of competence and it helps the students to recognize their strengths and weaknesses in terms of academic writing ability. Another technique favoured by many ESL writers is written feedback from the teachers (Kim & Kim [9]). This means the teacher evaluates the essay and does error correction on a grammatical level, but also makes 71 suggestions about structure and content (Gao [13]; Nordin 81). The teacher might want the student to re-write the essay again based on his/her recommendations and suggestions. The final draft is handed in to the teacher, who can then evaluate the essay and give written feedback and allocate marks based on the specific course and genre criteria and marking grid. 2.5.3 Arguments against the Process genre approach The potential advantages of the process genre approach are well recorded (Badger & White 2000, Gao 2007, Kim & Kim 2005). However, it is challenging to find concrete criticism against the use of the process genre approach in the literature about writing and among the prominent writing researchers or practitioners. Perhaps that might be one argument against using the process genre approach: its value and effectiveness have not been determined fully yet, the approach is still ‘young’, ‘new’ and innovative. It might currently still be viewed as a tool of deliverance from the process -, imitation - or the genre approach with their welldocumented drawbacks. It will take some time before obvious weaknesses are determined and recorded by researchers and practitioners. On the other hand, some points of criticism against the three approaches that were synthesized to form the process genre approach are still prevalent. They did not disappear miraculously. Since these disadvantages are thoroughly discussed in their respective sections, I will only mention them here to avoid repetition. The first concern raised by Horowitz (1986: 141-142) is time. In fact, the synthesis of the three mentioned approaches compounds the time problem. If time was a problem when using the process approach, it is an even greater issue in the process genre approach. The reason for that is that more activities and strategies, like reading, manipulating language features and analyzing model examples are added to help the students to write more effectively. This leads on to the matter of examination-writing. It has not been determined yet, whether the process genre approach helps students to write better and/or faster in examination settings. The assumption underlying the process genre approach is that if students are instructed based on genres and have had the opportunity to analyse and manipulate model examples, then they should be able to compose more effectively in an examination setting. I failed to find any empirical studies that could corroborate that assumption. The level of dependence on the teacher is also not reduced by the process genre approach. Indeed, students might demonstrate a higher level of dependency, because they are expecting to get input in the form of model essays, genre analysis and feedback. This might compound 72 the difficulty for students when dealing with writing under time pressure and without any resources or guidance to help. Another factor is the peer-editing feature of the process approach. As a feature of the process – and the process genre approach, its disadvantage was stated by Horowitz (1986: 143) as being “too soft”. Even though students can learn from each other, false impressions might still be created if students edit each other’s work whether it is the process approach or the process genre approach. It remains to be seen how useful, practical and functional the process genre approach really is in real teaching and learning situations. A few studies summarized below give information on the effect of the process genre approach on students’ writing skills. Note that, of the three studies summarized only one study compared the effectiveness of the process genre approach with another approach, the genre approach. 2.5.4 Research: Process genre approach in Application The mêlée regarding the effectiveness of major ESL writing approaches has resulted in a number of studies to determine which approach is most suitable to improve students’ writing abilities. Three studies regarding the effectiveness of the application of the process genre approach will be briefly accounted here. Nordin, Halib, and Ghazali, (2010: 46) conducted a study at the University Teknologi Petronas, Malaysia to determine the effect of the process genre approach on the writing skills of engineering students. Researchers selected 138 students through stratified random sampling; these students were divided into an experimental and a control group. The researchers did a pre-test to determine the mean scores of the students’ writing. The experimental group received writing instruction based on the process genre approach, whereas the control group used the genre approach. In the ninth week of treatment a post-test was administered. The researchers provided students with a technical writing test and 3 independent raters used holistic scoring to assess the papers, where a score of 6 is ‘clearly excellent’ and a 2 is ‘rudimentary’. The raters used the following elements to award a score: purpose, organization, development, style, correctness and visual aids (Nordin, Halib & Ghazali 2010: 50). The findings of the study indicate that the writing ability of students participating in the experimental group is better than those in the control group. An Independent Samples t-test was used to determine whether there was a statistically significant difference between the mean scores of the experimental group and the control group. The 73 test indicated that the experimental group benefited more from the application of the process genre approach than the control group who did the genre approach. The majority of students (79.6%) who received process genre approach instruction scored between 5.00 and 5.63, which is translated into ‘still impressive’, just below ‘clearly excellent’. No student scored below 4.13 (adequate). However, only 23.1% of the students in the control group who received genre approach instruction, scored between 5.00 and 5.25 and 8.6% of the students received scores between 3.00 – 3.63 (developing). The study highlights the advantageous nature of the process genre approach to teach technical writing. The implications of this study are that besides the practical benefits of the genre approach, some features of the process approach seem to facilitate improved writing skills among students. These features are: peer- and teacher feedback throughout the composing process, drafting, revision and individual pace. In 2009, Nihayah did a study at SMP6 Negeri 8, Malang in Indonesia, to determine how the implementation of the process genre approach can improve students’ writing ability. Before the study the students’ writing ability exhibited weaknesses and it is reported that “the teacher used to ask students directly to write without teaching them the ways of doing it” (Nihayah 2009: [1]). The researcher recorded the students’ mean scores before the study and again after the study. The researcher used Collaborative Classroom Action Research to conduct the study with the assistance of the class teacher. The findings showed that there was a notable improvement in students’ writing ability after the implementation of the process genre approach. A comparison of the mean scores (with a maximum score of 4) taken before and after the study indicate the improvement: 1.29 to 3.15 on content, 1.62 to 3.01 on organization, and 1.55 to 2.98 on language use. The researcher failed to report whether a control group was used to compare the results of the experimental group. The study would have been more convincing, had the researcher included a control group as measure. Foo (2007) did a PhD study on the effectiveness of the process-genre approach in a Malaysian secondary school. The study focused on how successful the process genre approach was in contributing to students’ ability to develop strategies to write more effective essays. The researcher applied the process genre instruction to the experimental group (30 students) for sixteen eighty-minute sessions. The control group (30 students) received the same number of sessions but with product-centred writing instruction. The researcher did a 6 The researcher could not determine the meaning for SMP. 74 pre-test and two post-tests: one immediately after the treatment session and another three months later. The study reports that there was a notable improvement in the experimental groups’ ability to communicate their ideas, and to develop more relevant ideas related to the purpose of the task as compared to the control group. However, no improvement was found in the organization of ideas or in the control of language. The students of the experimental group commented that they had a better understanding of “conceptual writing strategies” (Foo 2007: 16) and that they would be willing to apply practical strategies when writing essays. The consequence of the process genre writing instruction was that students sustained their improved writing status quo “over a period of time without any further instruction” from the researcher (Foo 2007: 16). No evidence of any Namibian studies on the application of the process genre approach could be found. The studies described above, which were mainly conducted in Malaysia, suggest that the process genre approach is beneficial. It is not only the writing instructors’ confusion about which approach “is best” that has led to the application of the process genre approaches in various ESL and ESP situations. It is also the needs of the students that have compelled writing instructors to evaluate critically the effectiveness of their approaches in the writing class. There are numerous other factors that have led to an innovative shift in the ESL and EFL writing instruction approach. True to the evolutionary nature of ESL writing, a ‘post process-genre approach’ (my own coin) is probably being contemplated as this dissertation is being written. 2.6 CONCLUSION Chapter 2 provides an overview of literature on the three writing approaches which were applied in this study: process approach, modeling/imitation approach and the process genre approach. In my explanation I described the origin and history of the writing approaches, features and principles. Another point discussed in each of the three approaches was criticism brought against the approaches. I furthermore included a summary of previous studies on the application of each of the three approaches in various settings. The next chapter will focus on the method and techniques I used to conduct my study with regard to the implementation and evaluation of the three writing approaches. 75 CHAPTER 3: RESEARCH METHODOLOGY For successful completion, a building requires plans that are clearly conceived and accurately drawn. A research project should be no less completely visualized and precisely detailed. (Leedy & Ormrod 2005) 3.1 INTRODUCTION I used action research as my research design as the study consisted of two cycles and several steps in the cycles. The intention of the study was to determine the effectiveness of the three writing interventions in improving the academic writing skills of FP students. In 2008, I implemented the first action research cycle, the process and modeling writing approaches, in two different classes: Class 1 did the process approach, whereas Class 2 did the modeling approach. Cycle 2 was implemented in 2009 as a result of the reflective stage in the action research which considered the effectiveness of these two approaches and led to a change in direction. Consequently, the process genre approach was implemented in the 2009 FP English course to determine whether this approach served to improve the academic writing skills of students in the FP to a greater extent than the two writing approaches in 2008. This means that in 2009 both classes did exactly the same writing syllabus based on the process genre approach. I did not use the experiment and control group format on ethical grounds. I could not ethically justify giving a writing intervention to one group, and exclude control group students from the potential benefits of the interventions. Convenience sampling was thus used. This Chapter will firstly look at the theoretical and practical issues relating to research design, and the research instruments against the background of the researcher’s philosophy regarding classroom practice and effective materials. The next section in this Chapter will provide information about the methodology and data which includes a description of the population and sample. The methodology explanation provides information on the techniques used to analyse the data and a justification of the techniques. Also included in this Chapter are the limitations of the research project and ethical considerations. The structure followed in this Chapter was extracted from Hofstee (2006: 112). 76 RESEARCH In this dissertation I adhered to the belief that “research is a systematic process of collecting, analyzing, and interpreting information (data) in order to increase our understanding of the phenomenon about which we are interested or concerned” (Leedy & Ormrod 2005: 2). Key issues are: problem, goal, plan, sub-problems, research question/hypothesis, critical assumptions, collection and interpretation of data, and the cyclical nature of the research process in action research. In addition, my study illustrates that research is a process and a product as stated by Nunan (1992: 2). The process of this study involves the journey from identifying the problem to the interpretation of the findings. Research is a product when knowledge is acquired as a result of the process. Furthermore, research normally relates to a design that is used to map out the steps used to come to a valid conclusion about the research problem. 3.2 RESEARCH DESIGN Research design refers to the plan of scientific inquiry (Babbie 1991: 89). I applied a research design because I wanted to find something out and reach a conclusion. The impetus for the study came from my concern about the inadequate academic writing skills of FP students. Therefore, I planned a strategy to observe and interpret the observation based on a research design that dealt with the procedures that were followed, the data that was collected and the way the data was analysed (Leedy & Ormrod 2005: 85). The research design for this study to determine the effectiveness of the intervention materials is an action research using qualitative and quantitative data collection and analysis methods. The next section explains these three concepts and gives a justification of the mixed method. 3.2.1 Action research: theory and justification Action research: I took a critical position regarding my teaching practice and the writing materials used to improve students’ academic writing skills. I selected action research as research design with qualitative and quantitative data collection methods. Action research is a reflective problem solving process “in which participants examine their own educational practice systematically and carefully, using the techniques of research” (Ferrance 2000: 1). The goals of this research project are to firstly, improve the academic writing skills of the FP students and secondly to improve my teaching practice. These goals are related to action 77 research which has as its aim professional self-development based on an investigation and evaluation of what is happening in real-life, practical situations (Varasarin 2007: 86). The features of action research relate to the emancipatory goal and critical research methods that allow for changes to the status quo (Myers 1997: Philosophical Perspectives, par 9). In addition, Welman, Kruger and Mitchell (2005: 205) illustrated the versatility of action research by stating that the research design is not formally finalized in advance and then followed strictly until the end of the research. Rather, the tentativeness of the action research process is emphasised by explaining that action research is cyclical and the next cycle depends on the outcome of the previous cycle. Action research is a holistic process during which measurement instruments may change in the process of data gathering. In my study the cyclical feature is reflected in the fact that the first cycle took place in 2008, followed by critical reflection. This led to the second cycle in 2009, with the addition of interviews with the science lecturers as another data collection method. My study reflects Nunan’s observation (1992: 17) that action research is practitionerorientated, “collaborative” and with the purpose of changing or rather improving an unsatisfactory real-life situation. I am involved in the research (practitioner) as the person who initiated the study and whose practice is critically examined and I am also leading the research. I have an “insider perspective” (Nunan 1992: 3). My study is not so much ‘collaborative’, except that students and lecturers are requested to cooperate and provide information that would yield relevant results. The findings are reported in a literary narrative, with reference to the participants’ perceptions and contributions (Leedy & Ormrod 2005: 95). The participant features prominently in action research. Ferrance (2000: 1) stated that in the educational field “action research specifically refers to a disciplined inquiry done by a teacher with the intent that the research will inform and change his or her practices in future”. The connection between this statement and the goals of the research project undertaken are that the research would lead to a change in my teaching practice and improvement in students’ academic writing abilities. The functionality of an action research as strategy for the current study lies in the emphasis on change through research (Davis 2003: 5; Blaxter, Hughes & Tight 2007: 68). The present study not only investigates and describes a problem experienced in practice (like a case study would do), but it focuses on the action that can be researched and “changed and reresearched” (Davis 2003: 6). Davis also emphasised another feature of action research, 78 which is the constant need for critical reflection throughout the process of research. Critical reflection can be obtained by regular feedback during the process of the action research. Feedback might lead to either positive confirmation of the applied change or a change of direction, should the feedback illustrate that the intended change does not show the assumed results. The systematic checking of progress links in with the cyclical pattern of action research (Blaxter, Hughes & Tight 2007: 69). A practical way of conducting an action research is to have a cyclical plan which involves posing a question, gathering data, reflecting on the data, and applying an action to change the initial problem situation, getting feedback about the effectiveness of the action and possibly having a second cycle, if necessary. This cycle and, in fact the term action research, was coined by Kurt Lewin in the 1940s when he found that investigation into work and the action necessary to solve the identified problem in a natural setting can develop into a valid research strategy (Ferrance 2000: 7). Action research is not limited to one cycle only. In the case of the current project two cycles are described. Two different interventions were applied in 2008 and a third intervention in 2009. The implementations of the interventions of both years yielded meaningful data and therefore the data of both years were included in this study. Students’ lack of effective academic writing skills was identified as a problem. This came about as a result of a needs analysis conducted in 2007, but for the sake of brevity the needs analysis and a detailed description of the students’ learning background is not provided in this study. The first cycle of the action was planned and implemented in 2008, and through critical reflection the researcher determined the effect and effectiveness of the action to bring about beneficial change. This change was analysed and further changes were made to improve the effectiveness of the action, which was evaluated again in 2009. This indicates the importance of flexibility of the researcher, as the researcher knows where the problem comes from, and has an idea where the research could lead to, but also realizes the fickle environment of working with humans who could react differently to change than anticipated. This means that transformation is possible but cannot conclusively be predicted (Wadsworth 1998: 5). In addition, “change does not happen at ‘the end’ – it happens throughout” (Wadsworth 1998: 6). This notion is also confirmed by Davis (2003: 7) where she cited Winter (1998: 63-64): “The progress of one’s inquiry over time – noting what happens as different things occur, as the situation develops: all this is essential to the learning process ... .” 79 The value of having used an action research design lies in the fact that action research is suitable for longitudal, cyclical research like mine. In addition, a researcher can be flexible when using action research, which was useful when I added the interviews as data collection tools and two sections to the questionnaires in 2009. Action research is a holistic process during which measurement instruments may change in the process of data gathering. Action research is also useful in measuring qualitatively and quantitatively whether the goals and objectives of the study have been achieved, i.e. have the students’ writing skills improved, has my teaching practice improved, has a change been observed? Elements of Qualitative data gathering and reporting: According to de Vos et al (1998: 267, 269) a qualitative design is determined by the researchers’ actions and choices regarding the data collecting and reporting method. Leedy and Ormrod (2005: 133) explain qualitative research as a design that focuses on issues in their natural setting. Qualitative data collection was used in this study to ensure that all the “complexities” and contextual “idiosyncrasies” (James 2006: 3) were taken into account. The qualitative nature allows for a rich description extracted from the interviews and openquestion questionnaires used in this study. Qualitative enquiry is said to be subjective and from an “insider-perspective”, that is the perspectives of the students, the science lecturers and my own (Nunan 1992: 3). In addition, human behaviour is fore grounded and therefore detailed descriptions and verbatim responses extracted from interviews and open-ended questionnaires in the current study show the “multiple … realities of phenomena” and as a result provide findings that are fluid and dynamic in nature. (Neill 2006: 1; Leedy & Ormrod 2005: 95). The participant features prominently in qualitative research which makes the connection between a qualitative design and action research particularly effective. Elements of Quantitative data gathering and reporting: Phenomena can also be measured objectively, “that is the feelings and the opinions of individuals” (Welman, Kruger & Mitchell 2005: 6) are not taken into account and the researcher is the ‘outsider’ (Nunan: 1992: 4), who is not concerned with any subjective matters going on during the research process. In this study pre- and post-intervention essays 80 and laboratory reports were marked by two independent markers. The marks were analysed objectively and statistically, thus treated quantitatively. In quantitative research, also called the positivist approach, causal relationships are attempted to be uncovered (Welman, Kruger & Mitchell 2005: 7) by using particularistic measurements and data collection techniques. Positivists support research that is started with a specific hypothesis to be tested (Leedy & Ormrod 2005: 94) about something that can be measured objectively. This is reflected in the current study by the research instruments used to gather statistical data in an objective manner: pre- and post-intervention essay and laboratory report marks and pre- and post-intervention closed-question questionnaires. In addition, the causal relationship is reflected in the hypotheses about the effect of the three interventions on academic writing skills of students. The statistical data “exists independently of the feelings and opinions of individuals” (Welman, Kruger & Mitchell 2005: 6) and the relationship between the two sets of data is explained objectively, which links it to logical positivism. Quantitative, statistical information yielded by the questionnaires and essay and laboratory report scores is analysed, interpreted and reported on objectively, which then either refutes or confirms the hypothesis that the writing interventions improve the academic writing skills of FP students (Leedy & Ormrod 2005: 94). Validity in action research: The concept of validity of action research is inextricably linked to validity of qualitative research. According to Golafshani the terms ‘validity’ and ‘reliability’ cannot be used in qualitative research the way they are used in quantitative research (Golafshani 2003: 600). He is of the opinion that in qualitative research these two terms cannot be seen as two separate entities, and should rather be substituted by concepts like “credibility, transferability, and trustworthiness” (Golafshani 2003: 600). Hofstee (2006: 127) specifically warns the action researcher to be careful of subjectivity; generalisability of results and replicability. This is because of the contextual nature of the research problem. Subjectivity was reduced in this study by contracting external markers (who did not know the students) for marking the essays and the science lecturers marked the laboratory reports. 81 3.2.2 Justification of the hybrid design of this study Research activity is traditionally categorized as being either qualitative or quantitative. However, as Nunan reports: “a binary distinction between qualitative and quantitative research … is simplistic and naïve” (1992: 3). In the case of my study, I accepted and adopted “methods and values of the alternative paradigms” to suit the purposes of the research aim. Leedy and Ormrod (2005: 95) confirm this perspective by stating that occasionally both research designs can be used to answer different types of questions which will yield more comprehensive knowledge about the world. Creswell (1998), Glesne and Peshkin (1992) and Moss (1996) agree that the qualitative and quantitative approaches can complement each other in the research process (Leedy & Ormrod 2005: 95). This means that even though I agree with Nunan that “all knowledge is relative, that there is a subjective element to all knowledge and research, and that holistic, ungeneralisable studies are justifiable” (1992: 3), I also believe that my study can generate data which is generalisable to the Namibian education community and that quantitative data that will be collected can be interpreted to enhance the descriptive interpretation of the qualitative data presented in this study. This is sanctioned by a critical discussion of research designs by Grotjahn (1987), in which he explains that hybrid forms of research paradigms are possible (Nunan 1992: 5). In addition, Nunan (1992: 20) remarks that “the distinction [between quantitative and qualitative research] is a philosophical one which is not always reflected in the actual conduct of empirical investigation”. Leedy and Ormrod (2005: 97) mention that the two approaches “are not mutually exclusive” and “it is not unusual for researchers to count (and therefore quantify) certain kinds of data, in what is, for all intents and purposes, a qualitative research investigation”. In practical terms this means, I have used data collection, analysis, interpretation and reporting techniques that fall into the qualitative camp (open-question questionnaires and interviews) and others that are more quantitative in nature (closed-question questionnaires and essay and laboratory report results). The techniques will be discussed in detail in a later section of this Chapter. 3.3. METHODOLOGY Here I explain the research instruments used and how I used the research instruments. I also illustrate how I applied the action research cycle to gather the data. 82 3.3.1 Research Instruments: Data collection techniques in relation to Qualitative and/or Quantitative research design 3.3.1.1 Interviews: theory, uses, strengths and limitations An interview is defined by Nunan (1992: 231) as “the elicitation of data by one person from another through person-to-person encounters”. I chose interviews because these research tools are widely used to collect data about facts, people’s feelings, motives, beliefs and perspectives and people’s reasons for actions or behaviours (Leedy & Ormrod 2005: 146). Interviews can be placed into three categories: structured, semi-structured and unstructured, depending on the type of information sought after, the degree of formality and “the degree of control the interviewer wishes to exert” (Nunan 1992: 149). The interviewer chooses a structured interview when a “collection of questions from a previously compiled questionnaire” (Welman, Kruger & Mitchell 2005: 165) is used. The interviewer does not deviate from the questions at all, and records the responses of the interviewee on a specific record-sheet. A semi-structured interview is conducted if the interviewer has a list of questions, topics and themes, which is called an interview guide (Welman, Kruger & Mitchell 2005: 166). It is important that all themes or topics are covered during the interview, but the interviewer has the freedom to change the wording or the order of the questions depending on the responses of the interviewee. The interview is normally recorded with a tape recorder and the interviewer also writes the responses down. In an unstructured or informal interview the interviewee is looking for in-depth information about a specific theme. No pre-determined questions are set up, but the interviewer needs to have a thorough idea about what needs to be explored. The interviewee is allowed to talk freely about the theme or topic. The interview is recorded with a tape recorder and the interviewer takes notes. For the purpose of this study semi-structured interviews were used and are discussed here. I used semi-structured interviews because I had certain topics and issues in mind, rather than pre-determined sets of questions. This means that although the same questions were asked, I had a certain degree of flexibility to find rich and informative data that allow for meaningful interpretation (Nunan 1992: 149). Questions were mostly open-ended and I was able to ask follow-up or elaboration questions (Welman, Kruger & Mitchell 2005: 167). Advantages of semi-structured interviews are, firstly, that the interviewer has power and control over the interview (Nunan 1992: 150). Secondly, the semi-structured interview allows for flexibility in elicitation techniques and the interviewer gets a picture of the interviewee’s life related to the theme of the study. The semi-structured interview was the 83 best type of interview in this study because I had access to the interviewees. The interviewees are known to me and therefore the setting was not formal and threatening. On the other hand, the information yielded might not be comparable because different interviewees might give different information (Leedy & Ormrod 2005: 146). Another disadvantage relates to the role of the interviewee, which might cause bias (Nunan 1992: 150), as the interviewee does not have the same rights as the interviewer. Also, information provided by the interviewee mostly relates to the past which the interviewee might not remember so well and that might yield inaccurate information, or the interviewee is intimidated by the interviewer and gives information that he thinks the interviewer would like to hear. The apparatus or equipment used during an interview (tape recorder) might be intimidating for the interviewee. The possibility of disadvantages was reduced because I conducted the interviews in an unthreatening manner in a familiar room. Interviewees were informed well in advance about the interview and how it would be conducted. The interviewees are my colleagues, and understand the necessity of recording the interview. Ethical considerations, such as privacy and anonymity were considered. How did I use interviews as data collection technique? The data collection method of the first cycle of the AR in 2008 was intended to be quantitative, using only the pre- and post-test marks and questionnaires of students as data. However, after reflecting on the proposed tools, I realized that the data would be too narrow and I decided to add interviews as research tools in 2009 to complement and add depth to the quantitative data. Information from and perspectives of FP Science lecturers on the quality of laboratory reports would be useful to indicate the extent of the writing problem with regard to laboratory reports and to include their suggestions on how to improve the quality of the students’ reports. The post-intervention interviews would also provide qualitative, interpretive data about the effectiveness of the three writing approaches in improving laboratory report writing skills of FP students. I scheduled individual interviews with four Science lecturers. The interviews were tape recorded. The pre- and post-intervention interviews had questions based on the quality of students’ report writing abilities. See Appendix 5. No interviews with the students were held because I believed that questionnaires would be more suitable to target the whole group of students. It would prove challenging to have interviews with 66 students, and analyse and interpret the data. 84 3.3.1.2 Questionnaires: theory, uses, strengths and limitations Questionnaires are defined by Nunan (1992: 231) as instruments “for collection of data, usually in written form, consisting of open and/or closed questions and other probes requiring a response from subjects”. Questionnaires used in this study provide quantitative data for and relate to questions of the research problem (Welman, Kruger & Mitchell 2005: 174). It is also advisable to use questionnaires from previous studies that have a distinct bearing on the current study. Therefore I used and adapted the questionnaire from a study called Effectiveness of using the process approach to teach writing in six Hong Kong primary classrooms (Ho 2006). As the title of the study indicates, the two studies (Ho and the study described here) are comparable in aim and content. The questionnaire designed by Ho focuses on key issues in my research. The constructs are relevant and applicable to determine FP students’ attitudes and writing habits. However, whereas Ho used the questionnaires to determine the effectiveness of only the process approach, I used it for all three approaches. In addition, I adopted only two sections in 2008: attitudes and writing habits and in 2009 I added the section about the evaluation of materials following the critical evaluation cycle of my action research project. The closed-question questionnaire was set up with pre-coded items. In the case of the current study students had to respond to statements by indicating their degree of agreement: strongly agree, agree, uncertain, disagree and strongly disagree – a five-point Likert scale. The advantages of this scale are that it is easy to compile, “multi-dimensional attitudes” (Welman, Kruger & Mitchell 2005: 157) can be measured using a collection of statements and that numerical data can be obtained. Another advantage relates to the time it takes to complete a questionnaire, as closed-question questionnaires are usually not as time consuming as openended questions. The disadvantage is that the respondents do not get a chance to explain why they have ticked a certain block. In addition, a respondent who does not want to fill in the questionnaire can tick any answer, just to finish quickly, and then the reliability and validity of the analysis and interpretation is compromised. Interests, attitudes, and beliefs or experiences are difficult to measure accurately as a result of “faking, and social desirability and the response style of acquiescence” (Welman, Kruger & Mitchell 2005: 143). Respondents can tick boxes that they think would “create a desired impression” (Welman, Kruger & Mitchell 2005: 143) and this is referred to as faking. Social desirability means respondents give answers that they believe are socially acceptable. Acquiescence refers to respondents answering for instance “yes” consistently without even considering the content 85 of the question. To counter these potential problems posed by the questionnaire and Likert scale, I decided to also employ an open-ended questionnaire considering the same constructs. Anonymity can have advantages as well as disadvantages. The advantage is that no matter what the respondents’ attitudes, beliefs, experiences or perspectives, they will be protected and their answers might be more truthful. Anonymity could also be a disadvantage in that any of the above-mentioned response problems might occur (faking, acquiescence, and social desirability). The danger of a low response rate was averted as the questionnaires were presented to the participants and they were answered in class time. However, misinterpretation of questions might be another drawback (Leedy & Ormrod 2005: 185). I selected closed-question questionnaires for quantitative data collection and open-question questionnaires were used for qualitative data collection. I wanted to give students the opportunity to fully explain their views and opinions by answering the open-question questionnaires to complement the results of the closed-question questionnaires and to add depth to the data. How did I use questionnaires as data collection technique? Pre-intervention: Closed-questions questionnaire (Appendix 6): This questionnaire was given to students before the interventions in 2008 and 2009 respectively. The questionnaire firstly elicited biographical and educational background information. The statements in the following parts of the questionnaire were based on the following themes: attitude towards writing; writing habits; materials; importance of writing. Participants had to respond on a Likert scale: strongly agree, agree, uncertain, disagree, strongly disagree. Students received the questionnaires in lesson time and had to complete the questionnaires in the lesson. Open-question questionnaire (Appendix 7): Students received these questionnaires before the intervention in lesson time. This questionnaire elicited information based on the same themes as the first questionnaire, but students were required to write longer responses, to allow them to give more detailed information regarding the themes of the questionnaire. 86 Post-intervention: The same questionnaires were used after the intervention with the only difference that the statements on the closed-questions questionnaire were written in such a way that students had to give information based on the writing materials of the FP. The questions on the openquestion questionnaire also focused on the writing experience of students and materials used on the FP. 3.3.1.3 Results of the (pre-intervention and post-intervention) essays and laboratory reports How did I use essay and laboratory report marks as data collection technique? Scores/Marks: Results of the pre-intervention and post-intervention essays and laboratory reports 2008: Students had to write an essay before the intervention started. The essay topic was: Explain how the current floods (January 2008 – March 2008) affect the living conditions of people in the north-central region of Namibia. The essays were marked and graded by two English lecturers on campus. I selected these lecturers as they are familiar with the educational backgrounds of the students and they know what is expected of the students in terms of academic writing ability as they teach first year English. These essays were marked and graded according to the marking scheme provided in Appendix 2. After the intervention, students wrote another essay on exactly the same topic as the pre-intervention essay. Results of the essays before the intervention and after the intervention were tabulated. A statistician used the Statistical Package for Social Sciences, version 18 (SPSS) tool to determine statistically significant differences in the mean scores. Students wrote a laboratory report before the intervention, and these were marked by the science tutors. The marks were recorded on a class list. Students wrote another laboratory report after the intervention, which were graded again. The results of the last report were also recorded. The statistician used the Compare Means and the Independent Samples t-test to determine the levels of significance for the pre- and post-intervention essay and laboratory report scores. This test is useful in determining whether there are differences in the performances of the same sample of students before and after the interventions. Independent Samples t-tests are 87 often used in situations where the number of subjects is comparatively small like in this study. 2009: The same procedure to collect comparative results of essays and laboratory reports was followed as described for 2008. The essay topic was different: Explain the importance of tertiary education in Namibia. The above mentioned information shows that this study has both, qualitative and quantitative data, and there will be an interpretive and a statistical analysis of data obtained before and after the interventions. 3.3.1.4 How the Action Plan and instruments were applied in my study Action research: practical application: The explanation below shows the steps followed to implement the two action plan cycles. Cycle 1: Step 1: Pre-thinking and needs analysis: The first step of the study was the problem identification, which was that FP students had ineffective academic writing abilities. The problem identification was not done with the deliberate purpose of conducting this study. My colleague and I noticed the weak academic writing abilities and lack of preparedness for tertiary studies of students in 2005, and deliberated on how the teaching practice could be changed and what materials would be suitable to improve the writing abilities of students. We also communicated with the Science lecturers on the FP course about the quality of students’ writing. At a visit to UNAM main campus in September 2006, I had the opportunity to talk to first year English Communication Skills lecturers as well as the Mathematics, Chemistry and Biology lecturers, who all confirmed that most first year students did not have the academic background to write effective academic essays and laboratory reports. After this informal reconnaissance mission my colleague and I did a needs analysis in 2006. Based on this needs analysis, the information I gathered from previous students’ writing abilities and comments from Science lecturers, I decided to change my teaching practice and write materials that I hoped would improve the writing abilities of students. I assumed that if FP students who attended the course in 2005, 2006 and 2007 had ineffective academic writing skills, then students in 2008 (and 2009) would have the same problem, since they came from the same educational background and had the same aim, namely to be admitted at main campus after doing the FP. I came to the conclusion that this writing ability intervention would be a suitable topic for the study. 88 Cycle 1: Step 2: November, December 2007, January 2008: Materials development: I wrote writing materials based on the process approach and the modeling/imitation approach. Extracts of the materials are provided in Chapter 4. Cycle 1: Step 3: March 2008: First investigation of AR cycle 1: I had to determine the extent of the problem of the 2008 students. I instructed students to write an essay on a specific topic: Explain how the current floods (January 2008 – March 2008) affect the living conditions of people in the north-central region of Namibia. Participants (class 1 and class 2) wrote the essay in class. The essays were marked and graded by two lecturers at Oshakati Campus. The essays were graded according to the Grading criteria grid illustrated in Appendix 2. No inter-rater reliability was done. I designed two questionnaires based on the same constructs and these were completed by students. The first questionnaire (5-point Likert scale) had close-ended questions, where students had to tick the box most relevant to their experience and attitude towards writing, as well as the nature of the writing materials they were exposed to at secondary school level (Appendix 6). The second questionnaire (Appendix 7) consisted of open-ended questions, where students had to give longer responses to questions about the same constructs dealt with in questionnaire 1. These questionnaires were not piloted. April 2008: Students wrote their first laboratory reports in Chemistry, Biology and Physics, and these were marked and graded by the respective science lecturers according to the marking grid in Appendix 4. The Biology lecturer marked the Biology laboratory reports, the Physics lecturer marked the Physics reports and the Chemistry lecturer marked the Chemistry reports. I felt that they would be the most suitable people to mark the reports as they conducted the experiments on which the reports were based. Cycle 1: Step 4: March - October 2008: Implementation of process and modeling/imitation approaches: I implemented the writing materials. Class 1 received the materials based on the process approach. Class 2 received materials based on the modeling/imitation approach. I did not have a control group based on ethical considerations. Cycle 1: Step 5: November 2008: Evaluation: At the end of their academic year on the FP, I had to determine the effect of the two writing approaches on the writing abilities of the 2008 students. I instructed students to write an essay on a specific topic: Explain how the current floods (January 2008 – March 2008) affect the living conditions of people in the north89 central region of Namibia. I gave the same topic as before the intervention to ensure reliability: another topic might have been easier or more difficult to write about which would compromise the reliability of the data. Participants (class 1 and class 2) wrote the essay in class time. The essays were marked and graded by two lecturers at Oshakati Campus. The essays were graded according to the Grading criteria grid illustrated in Appendix 2. No interrater reliability was done. The last laboratory reports for Chemistry, Physics and Biology were written and graded in October, and these marks were also used for data collection of my study. Students also had to complete the same questionnaires as in March. However, the wording of the questionnaires was changed slightly since the participants had to report on their experience and attitudes towards writing after having completed the FP. In the pre- intervention questionnaire the statements would read, for instance: I like the writing lessons at school. After the intervention, the same question would read: I like the writing lessons on FP. Cycle 1: Step 6: November 2008: Record keeping: I recorded the results of the questionnaires, essays and laboratory reports and did a preliminary analysis of the results to determine the effect of the writing interventions on students’ writing ability. Cycle 1: Step 7: December 2008: Critical reflection after cycle 1: At the end of the intervention in 2008, I conducted a critical review for each of the two classes. This was an in-depth process which allowed me to reflect on the quality and effectiveness of the two writing approaches respectively. I managed to determine the strengths and weaknesses of the intervention materials. This is a core feature of action research and the thinking involved in this phase is described fully in Chapter 4. Cycle 2: Step 1: December 2008, January 2009: Materials development: I wrote writing materials based on the process genre approach. I merged the materials of 2008 to get a hybrid approach which then formed the process genre approach. The materials are thus a combination of the recursive steps of the writing process as well as focus on the finished product in the form of example essays for different genres of academic writing (cause and effect essays, compare and contrast essays, classification essays, laboratory reports and argumentative essays). 90 Cycle 2: Step 2: March 2009: First investigation of the action research cycle 2: In 2009 with the new intake of students, I followed the same steps as in 2008. The essay topic was different: Explain the importance of tertiary education in Namibia. The essays were marked and graded by two lecturers at Oshakati Campus. The essays were graded according to the Grading criteria grid illustrated in Appendix 2. No inter-rater reliability was done. Students also completed closed-question and open-question questionnaires based on their school writing experience, attitudes, habits and materials. I also conducted interviews with the Biology, Physics and Chemistry lecturers with regard to students’ laboratory report writing abilities. Cycle 2: Step 3: March - October 2009: Implementation of process genre approach: I implemented the writing materials. The materials were used in exactly the same manner in class 1 and class 2. I had to use two classes as I could not exclude one class from the intervention for ethical reasons. This means class 1 and 2 students received the same modules which involved the process genre writing. I did not have a control group based on ethical considerations. Cycle 2: Step 4: November 2009: Evaluation: At the end of their academic year on the FP, I had to determine the effect of the process genre approach on the writing abilities of the 2009 students. I instructed students to write an essay on a specific topic: Explain the importance of tertiary education in Namibia. I gave the same topic as before the intervention to ensure reliability: another topic might have been easier or more difficult to write about which would compromise the reliability of the data. Participants (class 1 and class 2) wrote the essay in class time. The essays were marked and graded by two lecturers at Oshakati Campus. The essays were graded according to the Grading criteria grid illustrated in Appendix 2. No inter-rater reliability was done. The last laboratory reports for Chemistry, Physics and Biology were written and graded in October, and these were also used for data collection of my study. They also had to complete the same questionnaires as in March. However, the wording of the questionnaires was changed slightly since the participants had to report on their experience and attitudes towards writing after having completed the FP. For instance, in pre-intervention questionnaires a statement would read: There were useful materials relating to writing activities in school. The post-intervention statement would say: There were useful materials 91 relating to writing activities in FP English lessons. They were also asked to comment on the nature of the materials that they received on the FP. I conducted post-intervention interviews with the Biology, Physics and Chemistry lecturers with regard to students’ laboratory report writing abilities. Cycle 2: Step 5: November 2009: Record keeping: I recorded the results of the questionnaires, essays and laboratory reports and did a preliminary analysis of the results to determine the effect of the writing interventions on students’ writing ability. Cycle 2: Step 6: December 2009: Critical reflection after cycle 2: At the end of the intervention in 2009, I conducted a critical review for the 2009 classes. This was an in-depth process which allowed me to reflect on the quality and effectiveness of the process genre approach. I managed to determine the strengths and weaknesses of the intervention materials. This thinking phase is described fully in Chapter 4. 3.3.2 Data 3.3.2.1 Population and sample Qualitative research tools: Interviews 2009: Population: Four FP Science lecturers were interviewed. These lecturers were the existing Science lecturers of the FP, therefore convenience sampling was used to determine the sample for the interviews. The Science lecturers were two Physics lecturers, one Biology lecturer, and one Chemistry lecturer. These lecturers were chosen as they presented the Science lessons to the FP students, did the experiments with them and also marked their laboratory reports. Questionnaires 2008 and 2009: Population: Convenience sampling was used to identify the population. Because of the nature of the study and my responsibility towards the students I was not able to choose a subset from each class or one class only to carry out the intervention. The population used in all the research tools reported on below (qualitative and quantitative) was the FP students registered at UNAM Oshakati campus. For action research cycle 1 in 2008, the population was the FP class 1 and class 2 students. In 2008 there were 24 students in Class 1 and 25 students in Class 2, adding up to 49. For action research cycle 2 in 2009 there were 33 students in Class 1 and 33 in Class 2: 66 students all together. The total population for 2008 and 2009 is thus 115. 92 Sample: Open-question Questionnaire 2008: The sample is the students of Class 1 and 2 who were present on the day the questionnaire was done in lesson time. Open-question Questionnaire 2009: The sample is the students of Class 1 and 2 who were present on the day the questionnaire was done in lesson time. Quantitative research tools: Essay marks 2008 and 2009 pre- and post-intervention: The sample is Class 1 and 2 students who were present on the days the essays were written in class time. In 2008, for the preintervention essay, one student was absent, which brings the total number of essays handed in to 23. Four students were absent on the day the post intervention essay was written, with the total number of essays handed in 20. In 2009, only 27 out of the 33 Class 1 students handed in their essays, whereas 25 Class 2 student essays were handed in. Laboratory report marks 2008 and 2009 pre- and post-intervention: The sample consisted of the students that handed in the laboratory reports. Closed-question Questionnaire 2008 pre- and post-intervention: The sample is the students of Class 1 and 2 who were present on the day the questionnaire was done in lesson time. Closed-question Questionnaire 2009 pre- and post-intervention: The sample is the students of Class 1 and 2 who were present on the 2 days the questionnaire was done in lesson time. Next follows a detailed list of the number of students present on the days the data was collected: Essays: 2008 pre-intervention: Class 1: 23; Class 2: 22 2008 post-intervention: Class 1: 20; Class 2: 24 2009 pre-intervention: Class 1 and 2: 52 2009 post-intervention: Class 1 and 2: 52 Questionnaires: 2008 pre-intervention: Class 1: 24 responses; Class 2: 25 responses 2008 post-intervention: Class 1: 15 responses; Class 2: 18 responses 2009 pre-intervention: Class 1 and Class 2: 56 responses 93 2009 post-intervention: Class 1 and Class 2: 59 responses Laboratory reports: 2008 Biology pre-intervention: Class 1: 24; Class 2: 23 2008 Biology post-intervention: Class 1: 24; Class 2: 23 2008 Physics pre-intervention: Class 1: 24; Class 2: 19 2008 Physics post-intervention: Class 1: 23; Class 2: 19 2008 Chemistry pre-intervention; Class 1: 24; Class 2: 24 2008 Chemistry post-intervention: Class 1: 24; Class 2: 24 2009 Biology pre-intervention: Class 1 and Class 2: 66 2009 Biology post-intervention: Class 1 and Class 2: 66 2009 Physics pre-intervention: Class 1 and Class 2: 66 2009 Physics post-intervention: Class 1 and Class 2: 66, but 3 were absent 2009 Chemistry pre-intervention; Class 1 and Class 2: 66 with 3 absent 2009 Chemistry post-intervention: Class 1 and Class 2: 66 with 12 absent 3.3.2.2 Data: Explanation of the Data obtained Interviews: I set up the interview questions and the four Science lecturers were interviewed before the intervention and again after the intervention in 2009. The interviews were tape recorded. The data is qualitative and descriptive. The data obtained during the interviews focused on the quality of students’ laboratory reports before and after the intervention. I wanted to find out what the students’ strengths and weaknesses were regarding laboratory report writing. The interview questions can be separated into three different sections. The purpose of the first section was to gain an overall insight into the types of writing activities done in the Science lessons. It also focused on the students’ strengths and weaknesses in the writing skills to determine if there were any similarities compared to the quality of the students’ English writing skills. The second section aimed to find answers about the students’ laboratory report writing experiences as perceived by the science lecturers. It also dealt with the lecturers’ part in preparing the students to write laboratory reports and the strengths and weaknesses in the laboratory report writing skills of students as identified by the Science lecturers. In the third section the interview questions became more specific with regard to 94 students’ ability to use the appropriate vocabulary, style, grammar, cohesion techniques and structure when writing laboratory reports. The semi-structured interviews included the following questions: 1. What writing activities do students do in Biology/Physics/Chemistry lessons? 2. Which writing activities are students good in and which are weak? 3. How do you prepare students to write effective laboratory reports? 4. What aspects of laboratory report writing are students doing effectively? 5. What are they struggling with? 6. Do students use grammar effectively, like correct subject-verb agreement, or past tense passive voice in the procedure section? 7. Do students use words like ‘to determine’, ‘to test’, ‘to examine’ in the aim? The complete set of questions is found in Appendix 5. Closed-question questionnaire answers: The questionnaire was self-administered in class time under my supervision. The first closed-question questionnaire was administered before the intervention focusing on information before students started with the writing programme on the FP. The second closed-question questionnaire was done after the intervention with some adaptations. In 2008 I had only three sections on this questionnaire: section 1 – biographical and educational background; section 2: attitudes towards writing and section 3: writing experience. After personal reflection, I found that the questionnaires did not give me enough information on the materials and pedagogical purposes of writing. Therefore, I added two sections in the 2009 questionnaires. In 2009 the statements were divided into five sections. The first section of the data received from the application of the closed questionnaires in 2008 and 2009 will provide insight into students’ biographical and educational background. This helped me to understand the students’ past learning situation and socio-economic background which would have a bearing on the material planning and presentation, but will not be reported on in the data analysis and interpretation, as it was summarized in the background of the students in Chapter 1. The statements in the following sections relating to 95 the writing habits, attitudes towards writing, the type of materials they were exposed to and their expectations focus specifically on their writing experience and ability before and after the intervention. I will be able to determine whether any changes have taken place in their experience of, attitude to and ability of academic writing. The strength of this data is that I will have a complete set of all the participants present on the day of application since it was done in class time. The weaknesses relate to possible faking, acquiescence and misunderstanding the questions. The data is presented based on certain topics and in a numerical and/or statistical manner. The statements were rated on a five point Likert scale. The questionnaire can be found in Appendix 6. Open-question questionnaires: This self-designed questionnaire was applied to counter the weaknesses of the closed-question questionnaires and to allow students to give a richer description of their writing habits, experience of and attitude to academic writing. The data consists of sentence answers given by students. The data obtained by this questionnaire can be used in a quantitative manner based on topics and in a qualitative way, describing the experience, attitudes and expectations comprehensively. Some questions from questionnaire 2: 1. How did you feel when your teacher/lecturer instructed you to write an essay? 2. What type of essay questions did you get (e.g. articles, letters)? 3. How much time did you spend on average on writing an essay of about 200 words? 4. What are your weaknesses in writing? 5. What kind of writing do you expect to get on the FP? The complete questionnaire is in Appendix 7. Essay evaluations: The essay results are those marks that students scored after they wrote the pre- and post-intervention essays in 2008 and 2009. The essays were marked out of 20 and then the percentages were calculated. There are two sets of marks for each student, provided that the student was present on both days when the essays were written. The first mark is the pre-intervention result and the second mark is the post-intervention result. The marks were 96 recorded and an Independent Samples t-test was done to determine if a statistically significant improvement was achieved. Laboratory report evaluations: The first laboratory report marks obtained after the students did their first actual experiment in Biology, Physics and Chemistry were recorded by the Science lecturers and provided to me. These marks constitute the pre-intervention marks. The post-intervention marks are the marks of the last laboratory report that students wrote in the same year. That means there are 2 sets of marks for Biology, Physics and Chemistry each. The results were analysed using an Independent Samples t-test to establish if a statistically significant improvement had resulted. 3.3.3 Analysis 3.3.3.1 Analysis code 2008: Process approach: Class1 and Modeling/imitation approach: Class 2 Pre-test and post-test essays: quantitative: percentages: Independent Samples t-test and Bar graphs to indicate changes in marks. Pre-test and post-test laboratory reports: quantitative: percentages: Independent Samples ttest and Bar graphs to indicate changes in marks. Closed-question questionnaire: qualitative and quantitative: Bar graphs of significant issues and descriptions. Open-question questionnaire: qualitative and quantitative: Bar graphs of significant issues, verbatim responses and interpretive descriptions. Critical reviews: descriptions to indicate the strengths and weaknesses as experienced by the English lecturer (researcher) 2009: Process genre approach: Class 1 and 2 Pre-test and post-test essays: quantitative: percentages: Independent Samples t-test and Bar graphs to indicate changes in marks. Pre-test and post-test laboratory reports: quantitative: percentages: Independent Samples ttests and Bar graphs to indicate changes in marks. Closed-question questionnaire: qualitative and quantitative: Bar graphs of significant issues and descriptions. 97 Open-question questionnaire: qualitative and quantitative: Bar graphs of significant issues, verbatim responses and interpretive descriptions. Interviews: Themes and concepts are described qualitatively. Interviews were done qualitatively and results were analysed qualitatively, some verbatim responses are provided. 3.3.3.2 Description of the Methods of Analysis Essay results: SPSS was used to calculate and compute the results. The mean percentages of the pre- and post-intervention essays written in 2008 and 2009 were compared using an Independent Samples t-test, to determine a statistically significant improvement in the academic writing skills of students. In addition, the 2008 marks of Class 1 were compared to the marks obtained by Class 2 students, to determine whether either the process or the modeling writing approach was more useful in improving academic writing skills of students. In 2009, all students received the same writing instruction based on the process genre approach, therefore only the marks before and after the intervention were compared to determine any changes. However, the results of the three different approaches were also compared to each other to determine which of the three approaches yielded improvement and to what extent. The data is displayed in tables and Bar graphs to clearly illustrate whether any improvement has occurred. Laboratory reports: The first percentages of the first laboratory reports written in Biology, Chemistry and Physics (before the intervention) were compared to the laboratory report marks obtained by students after the writing intervention. Independent Samples t-tests were used to determine statistical data to indicate whether there was a statistically significant improvement in the students’ laboratory report writing skills. The 2008 marks underwent another comparison: the marks of Class1 (who did the process writing approach) were compared to the marks of Class 2 (who did the modeling approach). In 2009, there was only one comparison of marks to determine whether the students’ laboratory report writing abilities had improved, since both classes did the process genre approach. In addition, the results of the three different approaches were also compared to each other to determine which of the three approaches yielded improvement and to which extent. Graphs are used to show any differences in the percentages obtained by students. Closed-question questionnaire: (qualitative and quantitative): The results were analysed using a frequency count and indicating the pre-and post-intervention differences in a summative 98 table. Since the questionnaire was divided into themes, the data was categorized into the themes and thoroughly discussed. Open-question questionnaire: The answers of the students were analysed and categorized into themes and described to add depth and richness to the quantitative data. The most important information was provided in a descriptive format and supplemented with graphs or tables where necessary. Interviews: The data obtained from the interviews were categorised into themes and concepts to provide answers to the research questions and presented qualitatively, often including verbatim responses from the science lecturers. 3.4 LIMITATIONS Time was the biggest limitation. Action Research involves a longitudinal study with constant critical evaluation and re-thinking. Due to the space constraints of the dissertation, I omitted much detail relating to the needs analysis (Step 1, Cycle 1: 2008). I felt the project was too long and overwhelming, especially in 2009. The time lapse between the beginning of the study and the final writing up felt too long. Perhaps I was too ambitious with having so many research tools to gather the data, and by studying three different writing approaches over a period of two years. Secondly, some students were absent on the days when the essays, laboratory reports or the questionnaires were written. Also, in hindsight, regarding the questionnaires, I should have adopted the other sections as used by Ho (2006) on the Evaluation of the programme as well, useful information would have been provided if the students could give their opinion about the English course on the FP. However, I still have adequate data, since the majority of the students handed in their essays, laboratory reports and questionnaires. I believe that the data will still be useful to determine whether any improvements have been made in the FP students’ writing abilities, and which writing approach produced the best results. Also, I was unable to use a control group, which would have increased the internal validity of the study. As a result of my teaching context and ethical reasons I could not exclude one class from the intervention treatment. I used convenience sampling, which will impact on the transferability of my study and the generalizations drawn from the conclusions. 99 The study will provide information which will demonstrate the value of the three writing approaches in improving students’ writing skills. However, this study will not necessarily be transferable to other teaching contexts. 3.5 ETHICAL CONSIDERATIONS I was in a position to use my own classes in 2008 and 2009 as subjects in the study. Students were briefed about the purpose of the study and they gave their written consent. Another advantage was that there were no prescribed books and I had to write my own materials based on the syllabus objectives. I aligned the syllabus objectives with the writing skills of the process approach, the modeling approach and in 2009, the process genre approach respectively. The project thus contains a materials development aspect in addition to the research itself. These self-generated materials were evaluated by two groups of UNAM lecturers from the Language Centre at the main campus in Windhoek, including the Director of the Language Centre and three English lecturers respectively, for the purpose of quality assurance. The three FP Science lecturers and the laboratory assistant also agreed to participate in the study after I informed them about the aim of the study. 3.6 CONCLUSION In this chapter, I described and justified the use of an action research project using a mixed method research design. Information on research tools was provided, which was followed by an explanation of the population and samples used for every research tool. The data analysis, as described in this chapter, will provide information on the effectiveness of the three interventions. The chapter concludes with accounts of the limitations and ethical considerations. The next Chapter will focus on the data and data analysis. 100 CHAPTER 4: RESULTS 4.1 INTRODUCTION In this chapter I provide the data that I obtained during the study. The data include three reviews (2008 and 2009) of the impact of the three approaches. An extract of the three interventions as well as a meta-analysis in the form of track changes to illustrate how the materials are grounded in the three writing approaches follow. Then I present the data based on the essay and laboratory report results, the closed-question questionnaires, the openquestion questionnaires, as well as the individual semi-structured interviews. Statistical Package for the Social Sciences (SPSS) software was used to illustrate the context of the data in terms of numbers of participants, and the impact of the three approaches used in the interventions on essay and laboratory report results. Inferential statistics was employed to determine a difference among the results entered for each student for the essays and laboratory reports. Tests to determine the pre- and post-intervention mean scores were done, followed by Independent Samples t-test which indicate the statistically significant differences between the pre- and post-intervention essay and laboratory report results. After that, Bar graphs were produced to illustrate the results of the analyses. A summative, overall conclusion was created which indicates clearly the level of pre-intervention and postintervention performance of the students in 2008 and 2009. In addition, a conclusion could be reached about the extent of the three approaches’ potential to improve students’ writing skills with regard to the post-intervention results. SPSS was deemed useful for the analysis of the results, because it is simple (Hofstee 2006: 117) and it helped to make the data clear and understandable (Hofstee 2006: 151). All redundant information is removed, and only cumulative information that provides answers to the research questions is presented. The results of the closed-question questionnaires and open-question questionnaires were grouped according to the constructs given in the questionnaires. Only the most significant findings of the closed-question questionnaires are presented here in the form of frequency count tables, Bar graph displays and descriptions of the meanings concluded from the tables and graphs. Results of the open-question questionnaire are provided in descriptive form with verbatim responses from students’ questionnaires. The data which produce relevant information related to the research questions in Chapter 1 are presented. 101 The 2009 semi-structured interviews were analysed qualitatively to add more depth. The results of these interviews would indicate whether the science lecturers found the interventions beneficial with regard to students’ laboratory report writing skills. The results are described based on the most significant answers provided by the lecturers. The opinions of the science lecturers only relate to the 2009 intervention, meaning the process genre approach. Interviews were not conducted in 2008. The data presentation follows the following sequence. 4.1 The interventions: These are presented as a separate document (Addendum of the dissertation) as they contain track changes which alter the format of the document. The track changes contain a meta-narrative indicating my thinking behind the materials development. 4.2 4.1.1 The process approach 4.1.2 The modeling/imitation approach 4.1.3 The process genre approach Reviews: 4.2.1 Review of Class 1 (process approach) 2008 4.2.2 Review of Class 2 (modeling approach) 2008 4.2.2 Review of 2009 process genre approach 4.3 The context of the 2008 and 2009 studies and overall assessments are presented next. 4.4 Independent Samples t-test to indicate statistically significant differences in pre- and post-intervention results are provided as well as Bar graphs illustrating the effects of the interventions on the essays. The 2008 and 2009 pre- and post-intervention essay marks were used to show the impacts of the interventions. 4.5 The 2008 and 2009 marks of the pre-intervention and post-intervention laboratory reports were used to indicate whether the interventions had a cross-curricular effect. Independent Samples t-test to determine statistically significant results were used. Bar graphs illustrate the difference in results obtained by students and also the difference of results produced by applying the three different approaches. 4.6 The results of the 2008 and 2009 pre- and post-intervention closed-question questionnaire are given. 102 4.7 The results of the 2008 and 2009 pre- and post-intervention open-question questionnaire are provided. 4.8 Interview results: 2009 4.2 THE INTERVENTIONS Please see the separate document (Addendum). 4.3 REFLECTION REVIEWS Review refers to a reflection activity and is recommended as an important step in the action research cycle. I did formal reflection in 2008 and 2009 and the following three sections present the information resulting from the review activity. The 2008 reviews provided crucial input for the decision to explore the usefulness of the process genre approach to improve the writing skills of FP students. General information related to all students (2008 and 2009) involved in the study: In general, students could communicate orally in English and they passed their Grade 12 Science subjects which means their understanding must have been of intermediate to a high level. Their ability to understand diagrammes like tables and graphs was good, but in general they lacked the skill to explain the diagrammes fully. In terms of writing, most were struggling. They knew that an essay must have an introduction and a conclusion, and they could identify those two concepts, but they struggled to write them. Their content was a mixture of ideas, opinions, suggestions and recommendations with no logical order. Their ideas were generalized with no depth and very little evidence or examples for statements. There was no logical development in their essays. There was no logical, clear, meaningful paragraph structure. Students had difficulty with correct language usage as explained below. When students had sources to find ideas for their essays, they did not know how to extract relevant, essential points, make notes and summarize. In fact, they often plagiarized and failed to give references. 4.3.1 Review 2008: process approach After the introduction to the idea that writing is a process and not a once-off activity (the house-building analogy), it seemed as if students realised the lack in their approach to writing. Features of the process approach: 103 Analysing topics: This short unit to identify the key points in a topic was useful to most students. Whenever students wrote essays, this activity was done together in class time, not with every single topic, but with four or five. This helped them to provide relevant ideas and not miss the topic. However, in examination conditions, students seemed to forget what they learnt in class: how to actually circle and underline the key concepts in an essay topic. There were some cases where students missed the topic in exam conditions. Specific techniques to generate and organizing ideas: Students seemed to find this unit very interesting and they liked the group work activities to generate ideas. I introduced them to the different techniques one at a time. The first technique was free-writing, which they liked, but hardly any student used that technique when they had to actually generate ideas for their assigned essays. The second technique was spider diagrammes, which most students favoured and used when they had to do pre-writing activities in the writing units that followed. I think the simplicity and time efficiency of this technique lead to many students using it. However, their ideas were still very shallow. The tables and flow charts also seemed useful, but not to the same extent as spider diagrammes. Very often, the relationships between their ideas were not clear, but it did improve with time. For instance, when I showed them in the classification unit how to organize ideas of three or four items to be classified, they grasped the concept and applied it well, which lead to more meaningful ideas and organization. They wrote especially effective compare and contrast essays, and their cause and effect essays also showed a good understanding of the relationship between structure and content, as well as relationships between ideas, which starts with generating and organizing ideas. Paragraphs: Once students realized how simple it is to take the key points of the ideas and write them into topic sentences, they could apply this really well. On the other hand, some students persisted in giving sub-headings instead of topic sentences or their topic sentences often lacked a predicate. Writing strong and sufficient supporting sentences in each paragraph was more challenging. Even though they understood the concept of writing topic and supporting sentences, their ideas often lacked depth, but it was in general still better than at the 104 beginning of their FP year. As a result of organizing paragraphs into topic and supporting sentences, they had more ideas and could easily reach the word limits. First drafts: I checked all students’ first drafts to ensure that their content was valid and in relation to the topics, which they mostly managed well. However, their language usage was still a worrisome issue. Students were very appreciative of this action and eager to hand in their first drafts to receive help. Language usage: The biggest problem with language usage was their subject-verb agreement. Many students failed to add the plural form if their subjects were supposed to be plural and consequently the plural verb form was also wrong. They also often failed to add the –s at the end of a verb if the subject and the verb was singular. We did many grammar activities in class to practice this concept, and then students managed very well but when they wrote their essays independently they ‘forgot’ about it or did not know how to apply the rule successfully. This suggests that language items taught discretely might not transfer to the students’ independent writing. I employed different ways of explaining this issue, but failed to get all students to understand it. At the beginning of the FP, students also cut off their words at the end of the line at the wrong places. This means they did not know about syllables and the appropriate places to separate the words. Because of a lack of time to teach this concept, I took the expedient route and told them never to separate words at the end of a line and rather write the whole word correctly on the next line. This was understood quite well. Students also displayed linguistic characteristics of the language variety we call “Namlish”. This includes the overuse of the present continuous tense (I am having a book), the terms “used to”, “whereby”, “starting to...”, phrases like “Newton was the one who invented...”, “lions they are carnivores..”, “I am footing/moving to school”, the correct use of “again” and “also”, “both” and “all”, “this” and “these”, “a”, “an” and “the”, etc. One big problem was the spelling of words containing l and r. As a result of mother tongue interference, students confused these two letters and had many spelling errors, for instance: rular instead of rural. These issues were dealt with in error correction lessons and spelling tests and mostly 105 eradicated, but entrenched features of a developing variety of a language proved challenging, and some students persisted with these practices. Sentence structure was problematic at the beginning. Students tended to write very long sentences which then became confusing. Sometimes their paragraphs consisted of two sentences or even only one. We did practical activities in class to correct this, and their sentence structure became more effective. Punctuation, especially the use of capitals, was an issue at the beginning. Students started paragraphs and new sentences with small letters and even wrote names with small letters. This was also addressed in error correction lessons and most students improved. All in all, I felt the process approach writing intervention was effective in improving the academic writing skills of the students. The important principle, writing is a process, was implemented successfully, but mostly in lesson time and with my guidance. Students did pre-writing activities before the actual composing, they wrote one draft which was edited in class by themselves and their peers and they re-wrote. My role to provide input and assistance was very labour-intensive but rewarding. Laboratory reports: My observation was that the students improved their laboratory report results considerably, especially in Biology and Physics. They found the unit on laboratory report writing very useful as they could see the relevance clearly and after the unit they were very pleased with the improvement in their laboratory report marks. However, the recursive principle was not consistently evident in students’ actual composing process. Even though I would have wanted to agree with proponents (Goldstein and Carr 1996; Badger and White 2000; Emig 1971) of the recursive-principle, I cannot do that. In class, this principle was encouraged but it is not a very practical issue with large classes and students who are still dependent on a considerable amount of input and guidance from the lecturer. Lessons are usually structured and organized into steps and activities that follow logically. Perhaps, students used this principle when they continued with their essays at home. But, there is no conclusive evidence to support this. If recursiveness is understood as the ability to be flexible in the phases of planning, translating ideas into sentences, reviewing and re-writing then that is what was done in class, But, based on the explanations of recursiveness by Badger and White (2000), Emig (1971), it was not achieved in the English 106 lessons on the FP. For that reason, I decided that recursiveness as a principle of the process approach is not entirely useful in the English FP lessons. Another issue was time. I agree with Horowitz (1986) that time is a contentious issue. Time was not a problem when the essays were written for continuous assessment in the lessons. When students conducted the essay-writing activities in lesson time, it took at least six lesson hours of 55 minutes each, from analysing the topic to editing the first draft. Students were allowed to re-write the essay at home after the editing lesson and then hand in the following day. However, the essays that students wrote for the study, tests and examinations could only be written in one hour. Then students could not follow all the steps proposed by the process approach. This was reflected in their marks. Whereas their continuous assessments marks for essays were relatively good, the results of the essays for the study, tests and examinations were not as good. In conclusion, after the intervention, I still felt that the process approach had many features that could be used effectively to improve academic writing skills. Summary of salient themes/patterns that showed improvement using the process approach:  Analysing topics was beneficial.  Spider diagrammes were most successfully used to generate ideas. Idea generating techniques resulted in the following: more realistic ideas, more depth in ideas, more effective relationships between ideas.  Organising ideas improved: paragraph structure was more meaningful and effective, students managed to write effective topic and supporting sentences. The higher level of depth in paragraphs illustrated a deeper level of thinking about ideas. Students’ paragraph structure showed a clear relationship between content and structure.  Language usage: there was some improvement in subject-verb agreement, cutting of words decreased and spelling and punctuation improved considerably. Students managed to have a better sentences structure and sentence length.  Laboratory reports: There was a definite improvement in all respects, especially structure. Language usage improvement in the use of past tense passive voice and spelling was evident. Students also managed to use more appropriate vocabulary.  The following process approach features were helpful in the writing of academic essays, but mostly only during class time and not in the research pre- and post-intervention setting: recursiveness, review, sufficient time, peer-editing, role of the writing teacher. 107 4.3.2 Review 2008: modeling/imitation approach This group of students displayed the same inadequate academic writing abilities as the students in the process approach group. Introduction to Modeling/imitation approach: Students enjoyed playing “Simon says7...”, which is important as better learning takes place when having fun. The more serious approach to finding the relationship between imitation and writing and the dictionary work to find meanings was also done with enthusiasm, especially because most students did not have dictionaries at school and this was their first encounter with dictionaries. Features of the modelling/imitation approach: Reading and analysing weak and strong examples before the actual writing: Students first had to read the whole essay to get a general understanding of the text. Then the strong and weak essays were analysed in small steps starting with the introduction and ending with the conclusion. A comparison was drawn between the two essays and students had to determine the criteria based on the effectiveness of the strong essay. In my view, reading and analysing examples of a text before writing is beneficial, and a comparison of the preintervention and post-intervention essay results shows an average improvement of 12% of Class 2. The process approach comparison of pre-intervention and post-intervention essay results shows an average improvement of 13% for Class 1. This means that the improvements were fairly similar, an outcome also determined by Storalek (1194) in her study into the effectiveness of 5 different types of instructions. Proponents of the modeling/imitation approach (Bender 1993; Butler 2003; Farmer & Arrington 1993) assume that examples help students become more effective and independent. But the FP students failed to become totally independent with this skill. There was never a time when students comfortably and effectively analysed a whole example essay without my input. They still needed assistance until the end of the FP course. The modeling approach allowed for discovery learning: 7 A children’s game: one person is the leader and the group must imitate the actions done by the leader, but only if s/he announces “Simon says” before the action (e.g. waving arms). If s/he does an action without saying “Simon says” and some group members imitate the action, they are disqualified and have to sit. The last person standing is the winner. 108 I asked the questions regarding form, structure and language usage and students found the answers in the example essays. Students found the features demonstrated in the examples and could determine the functions and importance of the characteristics of the example essays, which they could model in their own essays. Reading and analysing examples was never a passive, mindless and unthoughtful learning method for FP students as explained by some apprehensive critics (Flanigan 1980; Harwood 2005; Kim & Kim 2005). The examples of laboratory reports were especially welcomed by students as they had written their first reports and received low marks. After reading and analysing the example reports in the English lessons their marks of the following reports improved considerably. Generating and organizing ideas: When students had to write their own essays, I compared the form of the essay to a specific template that would help students to structure their ideas meaningfully. This helped them to generate more relevant, meaningful and realistic ideas and organize them successfully into effective paragraphs. I modeled a number of techniques to generate ideas, like diagrammes, notes and free writing. Most students preferred the spider diagramme or note making technique. Their essays were no longer jumbled lists of ideas. Students managed to produce paragraphs with better organization and more depth as proposed by Coe (1987). It was interesting to witness how attention to form resulted in more effective idea generation and structure. Students’ positive experience with regard to generating ideas more effectively had an affirmative effect on their skills as academic writers. Writing first drafts: Even though multiple drafts is not a principle of the modeling approach per se, I applied it in the lessons. I checked all the students’ first drafts to determine whether their writing was related to the topic and the expected structure. The content and structure of the essays improved, but their language usage did not have the same level of improvement. It appeared as if the examples were not as effective as hoped to improve students’ language usage. This could be related to Watson’s (1982) apprehension regarding the questionable extent of input that leads to intake. In the case of this study the input had a limited effect on the language usage of students. Students of Class 2 showed similar deficiencies as the Class 1 students. To avoid repetition I will not discuss it here again. 109 Laboratory reports: Class 2 students enjoyed doing the simple experiments in the lesson and using the examples related to the experiments seemed useful. They felt more comfortable writing their own reports. Their marks improved which created a positive attitude towards laboratory report writing. On the whole, the modeling approach was useful in improving students’ academic writing abilities. Under my supervision FP students used the examples of model essays effectively to develop their writing skills. It would be interesting to know whether students continue to use this technique independently in a different learning context. My observation was that the examples demystified the academic texts students were expected to write. It illustrated what was expected of them, which reduced their apprehension. My observations were similar to those of Charney and Carlson (1995): using model texts have an impact on certain aspects of students’ writing skills, like generating more relevant ideas they might have omitted without the examples, organization of ideas and, style and rhetoric, and structure or form. The only aspect that was not addressed satisfactorily by the modeling approach was language usage as explained earlier. As a result of these observations of the two approaches used in 2008, I decided to use the process genre approach in 2009 to determine whether it would have a greater effect on the level of improvement of FP students. These reviews prompted me to re-think the effectiveness of the two approaches used in 2008: process and modeling/imitation. While a measure of success was evident, I would have liked more students to write more effective essays. Therefore, I decided to use the process genre approach in 2009. I believed that if the beneficial features, strategies and techniques of the process- and the modeling/imitation approach were combined, more success could be achieved with regard to writing improvements of FP students. Summary of salient themes/patterns that showed improvement using the modeling/imitation approach: Improvement was seen in:  Reading and analyzing weak and effective essay examples: students were very appreciative when they received examples of model essays. Model essays seemed to be beneficial and helpful for using in the class. Analysing example essays taught students how to read with more concentration and what features and conventions to look out for and to apply in their own writing. 110  Discovery learning: the example essays helped students to understand the relationship between form, style, language usage and content.  Imitation: students realized the advantages of reading effective essays, and using the essays effectively to write their own criteria and their own essays.  An imitator is a synthesizer: students managed to use what they learned from the example essays. They added their own content and, therefore, became more effective in synthesizing information, form, language usage and style.  Generating ideas: spider diagrammes were mostly used. The depth of their ideas improved as well as the relationships between ideas.  Laboratory reports: students used the example laboratory reports to help them write their own reports. The structure of students’ laboratory reports showed improvement. Their content was also better because they determined what information is required under each of the subheadings in laboratory reports. Language improvement was found in the use of past tense passive voice, spelling and writing the conclusion. 4.3.3 Review 2009: process genre approach I felt that the unique situation presented by the FP students called for an eclectic approach to improve their writing skills. The process genre approach as I applied it is a combination of the process and modeling approaches with an extra focus on genre. The educational background of the 2009 students was similar to that of the 2008 students, thus I will not repeat it here. Introduction to Process genre approach: The meaning of process genre approach was explained by using the same introductory materials of the process approach (writing process) and the modeling approach (Simon says, and dictionary work). An added feature was the genre. In hindsight I realized that a considerable amount of writing theory was presented to students, and I am not sure whether they fully understood the relationship between the three key concepts: process, modeling and genre. Features and stages of the process genre approach: Models as input: The models served multiple purposes. They were used to analyse topics and determine the relationship between the title, the purpose, audience, structure, content, language usage and style. For each genre of essay students had to write for continuous assessment pruposes in the English class on the FP, they received example texts or essays: classification, cause and 111 effect, laboratory reports, compare and contrast, and argumentative essays. The reading and analyzing of the examples in class time helped students to understand how academic texts are written. They were able to see how ideas were stated and clarified by the use of topic and supporting sentences in the input texts. In addition, the variety of effective introductions and conclusions encouraged students to write more effective introductions and conclusions in their own essays. Before the intervention, they normally wrote summaries as their conclusions which were often just repetitions of a collection of sentences in the essay and therefore not very effective. During and after the intervention, students realized that there are many different ways to begin and end an essay, and their own introductions and conclusions improved with regard to content and structure. The examples were helpful to illustrate how theoretical, abstract concepts like purpose, audience, structure, content, language usage and style are practically incorporated in the writing process. However, since model essays were not available during the pre-and post-test essay writing activity for this study, the true effect of this feature cannot be commented on with confidence. Generating and organizing ideas: I modeled to students how to generate ideas using the same techniques as in the process approach. The different techniques were dealt with individually and we also did joint ideageneration as class activity, or group- and pair work. This unit was done at the beginning of the programme to allow students to choose the technique(s) that appealed to them most. Students were not expected to write essays after they generated the ideas in this unit. Students felt that they wasted their time generating ideas and then just stopping abruptly to learn a new technique without doing something with their ideas. I believe I should have done this differently, by introducing a new idea generation technique for each actual essay that students were expected to write. On the other hand, there is a disadvantage of not introducing all idea-generating techniques at the beginning of the course: students might only at the end of the course find the technique most suitable for them. Writing paragraphs: We looked at example essays to identify topic sentences, and to determine how the supporting sentences are related to the topic sentence of each paragraph. Students analysed the paragraphs to determine the features of the paragraphs based on the genre, purpose and audience. Then students had to write their own paragraphs based on similar topics. Even though most students were able to draw the relationships in the example essays, they needed 112 a considerable amount of guidance to be able to do it themselves and to reach a certain level of depth in their own paragraphs. The benefit of initially working with paragraphs instead of whole essays was that I could closely monitor each student to determine weaknesses and address them. When students managed to write their paragraphs more effectively, they had more confidence to write longer essays. Consolidation: After the units about topic analysis, generating and organising ideas and writing welldeveloped and structured paragraphs, a class essay was written. A genre was chosen, for instance compare and contrast and then a topic: tigers and lions. The topic was analysed by looking at the meanings of compare (similarities) and contrast (differences). Then in groups students generated the ideas which were recorded by me and then the ideas were grouped into two sections: similarities and differences. I used the ideas to write an essay which students copied. Although time-consuming, these structured activities were useful to students as they became familiar with the genre, the methods and techniques of the writing process and in the end they had a model essay which they could imitate. Re-inforcement: Another example essay was provided and analysed based on content, structure and language. The purpose of this activity was to make sure that students understood all the important aspects and their relationships and what a whole essay looks like when all the steps were followed. I believe that this was useful to students as their continuous assessment essays showed improvement. Writing the essay: The stages of writing the own essays were done based on the stages of the process approach: pre-writing (analysing or choosing topic, generating and organising ideas); composing (writing the first draft); editing and revising; writing the final draft. These activities were normally done in class time and took about 6 to 8 lesson hours. Students did these activities quite independently and I only offered assistance when asked by individuals. I believe that the examples showed students what was expected of them and the steps showed them how to get to an effective product. The criteria also helped students to stay on track, edit and evaluate their own essays. 113 Writing laboratory reports: My general reflection about laboratory report writing is that students gained more confidence after the laboratory report writing unit in the English lesson. However, it would have been better if they could do the experiments in class like in 2008. Nevertheless, the majority of students managed to improve their laboratory report writing skills and that was also reflected in their marks. My opinion of the process genre approach is that it provided students with all the tools necessary to write more effectively: genres, examples and steps. The time issue as explained in the reflection of the process approach is also an issue here and will therefore not be discussed again. However, in my view, the activities and steps of the process genre approach take more time than only following the recursive steps of the process approach alone or simply using models to imitate. The same can be said for the language issue. My observation was that the content and structure of students’ essays improved noticeably when they had enough time to write, but the language improved only to a small degree, regardless of time. The outcome of my study based on the process genre approach can be compared to the study of Foo (2007), who also found that after the implementation of the process genre approach, students managed to improve their skills to communicate their ideas and to develop more relevant ideas. Foo did not see great improvement in organising ideas and language usage. In my study students managed to improve in content, and structure, but failed to a large extent to improve in language usage. Furthermore, writing teachers aiming to use the process genre approach are required to know what types of texts students are expected to write. If these texts are stated in the syllabus, or the course objectives, teachers should be able to determine the genres and incorporate these in their writing syllabus. However, if writing teachers are uncertain about what writing requirements are needed in the students’ future discourse community, they might make inaccurate assumptions to the detriment of the writing students. One might argue that writing teachers could attempt to determine the writing rhetoric anticipated for future studies by doing a survey or merely asking future teachers or lecturers about the objectives. But do writing practitioners always have the time, the confidence, and the experience and skills to do that? Do the writing teachers know exactly which institutions or work places the students go 114 to in future? Are all the rhetorical communities of all educational institutions the same? This means in my view that the process genre approach can only be used effectively if the writing teacher knows what genres to focus on. The disparities in the real classroom situations at students’ current or future institutions make it difficult for a writing teacher to decide which genres to choose. In addition, the question that begs an answer is: how well are writing teachers equipped to implement and maintain this approach? How do they know that they have translated the process genre approach successfully into their own teaching situation? How do they know that they are doing the right thing? Do they have to wait until the results of students’ writing attempts show that the approach has been effective? What do they do if the approach does not have the desired outcome? These questions could just as well be asked when other approaches are introduced. But they are still valid questions, and they are still drawbacks for almost any writing approach. Summary of salient themes/patterns that showed improvement using the process genre approach:  Models as input: The students found the reading and analyzing of model texts valuable. It helped them to understand how academic texts are written. It was useful in improving their continuous assessment marks for FP English essays, but not so useful in the essay writing session for this study or for examinations. However, by that time their writing skill has overall improved to an extent and they produced better post-intervention essays than their preintervention essays.  Generating ideas: students managed to choose a technique that suited them and the writing situation best. The different techniques also helped students to generate more meaningful ideas.  Organising ideas: students were able to organize their ideas more logically than before the intervention. Their essays had a higher degree of cohesion and coherence.  Writing paragraphs: this was a useful activity since paragraphs are like short essays and the structure of topic and supporting sentences in each paragraph helped students to have more ideas and write more effective sentences.  Consolidation: students found the consolidation activity helpful as it practically showed them a framework of how to deal with all the steps when writing essays.  Writing the actual essay: students were able to write more effective essays with less assistance from me. However, some students were still uncertain about how to narrow a topic and generate and organize ideas.  Laboratory reports: this unit has mainly helped students to improve their laboratory writing report skills with regard to structure and content. Language issues that improved after the laboratory writing unit were past tense passive voice, sentence structure and spelling. 115 In conclusion, one interesting point of the three review reports is that I have found the same issues as being challenges as students have indicated in their questionnaires: language usage like tenses, and subject-verb agreement. My perception was also that it is easier to improve content and structure, than grammar. Even though students are encouraged to focus on their grammar issues at the end, it is nevertheless an important aspect in writing as effective language usage is part of communicative competence. Language usage affects students’ marks, just as one student stated in the open-question questionnaire: “If my grammar is bad, my marks are low, it is discouraging”. The reviews indicate that there were some successes with regard to teaching and learning in the writing approaches of 2008 and 2009. But many challenges still remained, for instance students coming to the programme with F in Grade 12 English, and failing to attend the extra-classes offered or office consultations. For those students there was very little hope of showing sufficient improvement as 7 months is not enough to improve all the communicative aspects relevant to academic writing. 4.4 CONTEXT OF THE STUDY The information below gives the context of the study. It indicates the number of students who took part in the study with regard to the three approaches used in the intervention over two years. In 2008, two different approaches were used: process approach with Class 1, and modeling/imitation approach with Class 2. In 2009, the process genre approach was used with both classes. Important to mention here is that the class sizes differed. In 2008, Class 1 had 24 students, while Class 2 had 25 students. In 2009, both classes had 33 students each. Thus, the adapted intervention (2009) based on the critical reflection after Cycle 1 in 2008, had twice as many students than either group in 2008. The higher number of students in 2009 affected the study mainly in the practical application: I could not provide as much individual assistance to students as in 2008. In terms of the results, it might have an effect: a higher number of students might either get better or weaker results. However, the marks are provided as percentage scores in the tables and graphs that follow. It should therefore not have an effect on the calculations to determine significance levels of mean scores. 116 Figure 1: Overall context of the study Frequen cy 24 Valid Process Modelin 25 g Process 66 Genre Total 115 Perce nt 20.9 21.7 57.4 100 A total of 115 students were taught and monitored in the AR study, which consisted of two cycles spanning over two years: 2008 and 2009. The second cycle was employed after critical reflection on the two 2008 writing approaches. In the first cycle in 2008, 21.7% used the modeling approach and 20.9% used the process approach, whereas 57.4% were instructed in cycle 2 (2009) using the process genre approach. In 2008 the researcher monitored 42% (N=49) of the students while 57% (N=66) of the students were monitored in 2009. 4.4.1 Overall Assessments The results presented here give answers to research question (f). (f) How effective have the interventions been to improve students’ writing results in academic essays and laboratory reports? The results are presented first and then an interpretation of the findings is provided. 117 4.4.1.1 Pre- and post-intervention results compared: essays and laboratory reports (2008 and 2009) The results of pre-intervention essays and laboratory reports of all three approaches (2008 and 2009) were compared to the results of post-intervention results of essays and laboratory reports of all three approaches. The pre-intervention results of all three approaches were added together to be compared to the combined post-intervention results. Time of Assessment Marks scored Pre Intervention N Mean 427 60.87 Std. Deviation 14.953 Std. Error Mean .724 Post 403 71.51 13.889 .692 Intervention Table 1: Pre- and post-intervention results of essays and laboratory reports compared The table illustrates the difference in pre-intervention and post-intervention means of essays and laboratory reports. It covers both years to give an overall indication of the effect of the three writing approaches on the academic writing ability of 2008 and 2009 students. It is evident from the table above that the average score at post-intervention assessment (71.51%) was considerably higher than the pre-intervention score (60.87%). The next table provides data which determined the level of significance with regard to the improvement. 118 Independent Samples t-test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Difference Sig. (2- F Marks Equal scored variances 2.174 Sig. t df .141 -10.604 Std. Error tailed Mean Differenc ) Difference e Lower Upper 828 .000 -10.639 1.003 -12.608 -8.669 -10.626 827.792 .000 -10.639 1.001 -12.604 -8.674 assumed Equal variances not assumed Table 2: Independent Samples t-test of pre- and post-intervention essay and laboratory report results In a bid to establish whether the interventions were effective in improving the students’ performance on their essays and laboratory reports, a t-test for the equality of means was performed. The null hypothesis was that the interventions had no effect on the academic writing abilities of students. It was shown at the 5% significance level that there was a significant improvement in the performance of the students as a result of the interventions. This conclusion was reached since the p-value of 0.00 is less than the significance level of 0.05. The null hypothesis was rejected. This means that the three interventions had a positive effect on the academic writing abilities of 2008 and 2009 FP students. The following graph illustrates the differences in pre- and post-intervention results. 119 Figure 2: Bar graph: pre- and post-intervention results of essays and laboratory reports Results show that the performance of students improved post-intervention. Where only 83 out of 427 students managed to achieve the Excellent result pre-intervention, 182 out of 403 students received an Excellent result post-intervention. Interestingly, 77 students achieved Good results in the pre- and post-intervention measurement. Whereas 104 students received Above Average results pre-intervention, 79 were recorded in the same category postintervention. 52 Students had Weak8 results pre-intervention compared to just 16 postintervention. Furthermore, 26 students were categorized as receiving Poor results preintervention compared to only 4 recorded post-intervention. The conclusion of this finding is that there was a distinct improvement post-intervention if results of all three approaches are added together. The overall results presented above indicate that most students of 2008 and 2009 benefited from the interventions in terms of achieving better marks for their essays and laboratory reports, even if all did not get Good or Excellent results. This holistic method of presenting the findings creates a background to the individual results for each class and/or each process. Tables 1 and 2 and Figure 2 clearly show that the three 8 In the Bar graphs Weak and Poor pre-intervention means a higher number of students had 49-0%, whereas Weak and Poor post-intervention means a lower number of students had 49-0%. This means the performance post-intervention is better. 120 interventions collectively were instrumental in improving students’ academic essay and laboratory report writing skills. 4.5 ESSAYS One of the aims of this study was to determine whether the interventions had any effect on the academic writing skills of the students. In addition, it was stated in Chapter 1 that the study aimed to determine which writing approach would be the most beneficial with regard to improvement of academic writing skills. The tables and graphs below shed light on the two above-mentioned aims. The findings also produce insight into my teaching practice. 4.5.1 Comparison of overall pre-intervention and post-intervention essay results All the pre-intervention essay results of all three approaches were compared to the postintervention essay results. Group Statistics Time of Assessment N Marks scored Pre Intervention Mean 97 46.60 Std. Deviation Std. Error Mean 13.041 1.324 Post 96 57.08 11.463 Intervention Table 3: Pre- and post-intervention overall essay mean score 1.170 It is clear from the table that the post-intervention mean score of 57.08% is higher than the pre-intervention mean score of 46.60%. Thus, there is a difference of 10.45% in the pre- and post-intervention results. However, this is not conclusive to indicate clearly that the interventions had a statistically significant effect on the academic writing skills of 2008 and 2009 FP students. Thus, a test for the equality of means was performed and the results are shown in Table 4 below. 121 Independent Samples t-test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Difference Sig. (2- F Marks Equal variances scored assumed Sig. .395 Equal variances t df .531 -5.930 Mean Std. Error tailed) Difference Difference Lower Upper 191 .000 -10.485 1.768 -13.973 -6.998 -5.934 188.380 .000 -10.485 1.767 -13.971 -7.000 not assumed Table 4: Independent Samples t-test: for pre- and post-intervention overall essay results The two-tailed test illustrated above shows that there is sufficient evidence at the 5% significance level to conclude that the interventions were very effective in improving the essay writing skills of the 2008 and 2009 FP students. This conclusion was reached since the p value of 0.00 is less than the significance level of 0.05. Therefore, the null hypothesis was rejected which stated that the performance before the intervention is equal to the performance after the intervention. The following Bar graph illustrates the effect of the interventions on essay results (2008 + 2009). 122 Figure 3: Bar graph: 2008 and 2009 pre- and post-intervention essay result comparison Only 7 students achieved Excellent results pre-intervention compared to 11 who achieved Excellent marks post-intervention. There were no Good results pre-intervention. However, 6 Good marks were recorded post-intervention. The Above Average category had an increase of 19 students, from 9 pre-intervention to 28 post-intervention. A high number of students (35) received Average post-intervention, with only 13 and 3 receiving Weak and Poor respectively after the interventions. The post-intervention results indicate a good argument for the three writing interventions with regard to FP students’ essay writing skills. Tables 5 to 10 and Figures 4 to 6 present individual data on students’ achievements with regard to the three approaches respectively. These tables and figures provide data which allow a comparison with findings from other studies relating to the three approaches. 123 4.5.2 Essay Class 1: process approach – 2008: The tables and the graph below compare the pre-intervention results with the postintervention results of Class 1 students in 2008. Students wrote these essays in examination conditions in only one hour. Group Statistics Time of Assessment N Marks scored Pre Intervention Std. Deviation Mean 23 49.35 Std. Error Mean 14.326 2.987 Post 20 59.75 12.615 Intervention Table 5: Pre- and post-intervention Class 1 essay mean scores 2.821 A difference of 10.40% was found between the pre-intervention and post intervention essay scores. The mean pre-intervention was found to be 49.35% and that for the post-intervention score was found to be 59.75%. It was further tested if this difference is significant and the results are shown below. Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Difference Sig. (2- F Marks Equal variances scored assumed Equal variances Sig. .264 t .610 -2.509 df tailed) Mean Std. Error Difference Difference Lower Upper 41 .016 -10.402 4.146 -18.775 -2.029 -2.532 40.990 .015 -10.402 4.109 -18.700 -2.105 not assumed Table 6: Independent Samples t-test for significant difference in Class 1 essay mean scores The table above shows that Class 1 students managed to improve their academic essay writing skills as a result of the process approach intervention. This is evidenced by a p value of 0.016 which is less than the 5% significance level, allowing the rejection of the null hypothesis which states that the mean scores are the same. Figure 4 presents the data in the form of a Bar graph. 124 Figure 4: Bar graph: Comparison of Class 1 (2008) pre- and post-intervention essay results The results show that the post-intervention marks are better than the pre-intervention marks. In Class 1, 13% (3) of students received an Excellent score pre-intervention which increased to 20% (4 students) post-intervention. No Good results were recorded pre-intervention and only 6% - or 1 student - managed to receive Good post-intervention. The Above Average result increased from 8% (2 students) pre-intervention to 20% (4 students) post-intervention. The Average result showed the highest increase from 21% (5 students) pre-intervention to 46% (9 students) post-intervention. The Weak results decreased from 7 (30%) to 2 students and there were no Poor results recorded for post-intervention as opposed to 30% of the preintervention results. These results are an indication that the process approach has been useful in improving writing skills of most of the Class 1 students. Figure 4 shows that students have managed to improve their academic essay writing skills even though 46%, or 9 students out of 20, still have Average results. Researchers citing a positive effect of the process approach on the writing skills of students include Goldstein and Carr (1996), Mahon and Yau (1992), Talshir (1998), and Urzua (1987). However, these studies did not provide comparable quantitative information specifically based on improvements in marks. On the other hand, Ho’s findings (2006) indicated quantitative information in the form of pre-and post-intervention comparisons of marks at six schools in Hong Kong. Her results showed post-intervention improvement at the six schools involved 125 in the study. The results were based on a mark out of 50 for pre-and post-test essays. Improvements ranged from 4.1 points increase out of 50 to a 20 point increase. Only one school received an average 20 point increase, the other schools increased with 11.4, 4.4, 4.1, 5.59 and 5.34 average points out of 50 respectively. These improvements concur with the results of my study. 126 4.5.3 Essay Class 2: modeling/imitation approach – 2008: These tables and graph compare the pre-intervention results with the post-intervention results of Class 2 students in 2008. Students wrote these essays in examination conditions in only one hour. Group Statistics Time of Assessment N Marks scored Pre Intervention Std. Deviation Mean 22 53.18 Std. Error Mean 16.514 3.521 Post 24 61.67 11.948 Intervention Table 7: Pre- and post-intervention Class 2 essay mean scores 2.439 The table above shows that a mean score of 53.18% and 61.67% was observed for pre- and post- intervention essays in Class 2 respectively. This resulted in a difference of 8.49% in mean scores. In order to determine if the difference in mean scores was significant a t-test was used, as illustrated in the following table. Independent Samples t-test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Difference Sig. (2- F Marks Equal variances scored assumed Equal variances 1.831 Sig. t .183 -2.009 df tailed) Mean Std. Error Difference Difference Lower Upper 44 .051 -8.485 4.224 -16.998 .028 -1.981 37.999 .055 -8.485 4.283 -17.155 .186 not assumed Table 8: Independent Samples t-test for significant difference in Class 2 essay mean scores This table indicates that at the 5% significance level the 8.49% difference is not statistically significant since the p value (0.051) is greater than the level of significance (5%). Thus the null hypothesis, which states that the performance is the same, was not rejected. This means that the difference as result of the intervention is insignificant, thus the modeling approach writing intervention did not improve the performance of the class 2 students on essay writing 127 skills at a statistically significant level. Despite these results the following Bar graph shows that there was a positive change in the essay results of Class 2 students. Figure 5: Bar graph: Pre- and post-intervention Class 2 (2008) essay results This graph indicates that in Class 2, the biggest improvement was in the Average category with 14%, or 3 out of 22, students receiving Average pre-intervention and 38% (9 out of 24) post-intervention. There was a slight post-intervention increase in the Excellent section with 7%, from 18% pre-intervention to 25% post-intervention, translated into 6 students, receiving Excellent. No good results were recorded pre-intervention compared to the 13% (3 students) post-intervention. The results show a decrease in the Weak section: from 37% (8 students) pre-intervention to 7% (2 students) post-intervention. No poor results were recorded postintervention. The results in the graph show that the modeling/imitation approach has affected the writing skills of many of Class 2 students, even though 38% of the students still fall in the Average category. The results illustrated in Figure 5 indicate a similar tendency to the results achieved for the process approach: there is a slight improvement in Excellent (18% pre-intervention to 25% post-intervention), and Good (no recording: pre-intervention to 13% post-intervention). In the Above Average category the results decreased slightly from 18% pre-intervention to 17% post-intervention. There is a huge difference in Average, from 14% (3 students) pre- intervention to 38% (9 students) post-intervention, the same phenomenon as observed in the 128 results of Class 1. Two studies conducted by researchers, Stolarek (1994) and Twomey (2003), on the effectiveness of the modeling/imitation approach reported similar results. Stolarek did not quantify her data, but Twomey indicated that out of the 49 students in her study 6 students scored an A post-intervention, 27 scored B’s, 12 got C’s and 3 got D’s. Both contested to the positive impact of the modeling/imitation approach on students’ essay marks. A favourable effect of the modeling approach on the academic writing skills of Class 2 students can also be seen in the results of Figure 5. 129 4.5.4 Essay Class 1 and 2: process genre approach – 2009: The following tables and graph compare the pre-intervention results with the postintervention results of Classes 1 and 2 students in 2009. Students wrote these essays in examination conditions in only one hour. Group Statistics Time of Assessment N Marks scored Pre Intervention Std. Deviation Mean 52 42.60 Std. Error Mean 8.993 1.247 Post 52 53.94 9.918 Intervention Table 9: Pre- and post-intervention 2009 essay mean scores 1.375 The table above shows that the mean score of Class 1 and 2 pre-intervention results was 42.60%. The post-intervention essay result mean score was 53.94%. There was a difference of 11.34%. It was further investigated if this difference was significant using a t-test as illustrated below. Table 10: Independent Samples t-test for significant difference in 2009 essay mean Independent Samples t-test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Difference Sig. (2- F Marks Equal variances scored assumed Equal variances .540 Sig. .464 t -6.111 df Mean tailed) Difference Std. Error Difference Lower Upper 102 .000 -11.346 1.857 -15.029 -7.664 -6.111 101.037 .000 -11.346 1.857 -15.029 -7.663 not assumed scores The significance level of .000 shows at the 5% significance level the difference of 11.34% is statistically significant. This means that the process genre intervention was successful in improving the performance of the students in essay writing skills. The Bar graph below 130 shows the pre-intervention results compared to the post-intervention results of the students doing the process genre approach. Figure 6: Bar graph: 2009 Pre- and post-intervention essay results The graph indicates that no marks were recorded in the Excellent or Good categories preintervention, meaning no student managed to score above 70% pre-intervention. However, only 3 students managed to score above 70% post-intervention, with 2% (1 student) in Excellent and 4% (2) in Good. The 39% (20 students) in post-intervention Above Average compares well with the only 6% (3) pre-intervention. All the pre-intervention marks are from Above Average (5%) to Poor (12 out of 52 students), with the highest result in Weak (24 students), compared to the 18% (9 out of 52 students)in Weak post-intervention. The results in Poor reduced from 23% pre-intervention to 6% (3 students) post-intervention. The steady post-intervention increase from Poor to Above Average is positive in terms of the question whether the process genre approach has benefited students’ academic writing skills. Researchers in different situations have also reported on an improvement in writing skills as a result of the process genre approach. Nordin, Halib, and Ghazali, (2010: 46) reported in their quantitative study at Univesity Teknologi in Malaysia that the process genre approach resulted in a positive outcome for the students. They compared the pre-test and post-test results and found that 79.6% of the students doing the process genre approach received a score between 5.00 and 5.63 out of 6 compared to the 23.1% of the students in the control group who received genre approach instruction, and scored between 5.00 and 5.25, and 8.6% of the students received scores between 3.00 – 3.63. In addition, a study done by Nihayah in 131 2009 on the effectiveness of the process genre approach indicated an improvement postintervention. The researcher found that students’ mean scores (with a maximum score of 4) taken before and after the study improved with 1.29 to 3.15 on content, 1.62 to 3.01 on organization, and 1.55 to 2.98 on language use. The results of the current study indicate a comparable positive impact on the results of essay writing abilities of students who were instructed using the process genre approach. 4.5.5 Comparison of the essay results of the three approaches: 2008 and 2009: Most students receiving post-intervention Excellent marks were from the modeling approach at 25% (6 students) with the lowest Excellent achievement recorded in the process genre approach at 2% (1 student). The modeling approach equally recorded the highest Good results (13% - 3 students) with the process approach recording the lowest at 4% (1 student). 39%, (20 out of 52 students) of the students instructed using the process genre approach received above-average marks. However, the highest record in all last three scores namely Average (33%), Weak (27%) and Poor (5%) were recorded in the process genre approach. These results indicate that, if Excellent, Good, and Above Average results are taken into consideration, the modeling approach produced the best essay results (55%), with the process genre approach following in second place (45%). The process approach is not far behind with 44% if the Excellent, Good and Above Average percentages are added together. On the other hand, the differences in mean scores of the three approaches indicate that the process genre approach had the biggest difference with 11.34%, followed by the process approach with 10.4% and lastly the modeling approach with 8.49%. This finding means that the process genre approach has had the biggest impact on the academic writing abilities of students, albeit marginally. The anticipated improvements resulting from the implementation of the process genre approach were based on the considerations after Cycle 1 of the action research. Despite the fact that there were more students per class in 2009, and that their pre-intervention results (no Excellent or Good recordings and only 6% for Above Average) illustrated weaker writing abilities, there was considerable individual improvement, which resulted in a higher post-intervention mean score. Usually, more students per class translates to less individual assisting, monitoring, input and feedback from lecturers to students, which means students have to be more independent. Interestingly, the process genre approach appeared to have helped the students with their abilities to construct and write academic essays more independently than before, which I believe resulted in a higher mean score than the modeling – and process approach. 132 4.6 LABORATORY REPORT RESULTS The effectiveness of academic writing skills is important in writing for the sciences. As the aim of the FP is to equip students with the necessary skills to cope at tertiary level in mainly science-related fields, it is important to help the students write effective laboratory reports. Therefore, students wrote laboratory reports and consequently the pre-intervention laboratory report marks were compared to the post-intervention laboratory report marks to determine any statistically significant difference in their writing ability. The Bar graphs provide information on the effectiveness of the interventions in relationship to the laboratory report writing skills of students in 2008 and 2009. 4.6.1 Overview of how the approaches were used to teach the laboratory report writing unit in the English lessons: Class 1: 2008: process approach: Students received information on how to go about writing the laboratory report based on the steps of the process approach. For an introduction, a simple experiment (The volume and mass of irregular shaped objects) was done in the English lesson, and then the laboratory reports were written as class activity. The lecturer guided the whole process with leading questions and instructions and application activities. Peer-editing and editing was done regularly. No examples of laboratory reports were provided. A Biology experiment (Osmosis) was also done in the English lesson and a laboratory report written in different stages as group work. Special focus was placed on past tense passive voice, how to write in the different sections of a laboratory report and how to use discourse markers, action words and hedging words effectively. Students received the criteria and marked their own laboratory reports. Ohm’s Law was done in the Physics lesson but the process and the steps of writing the laboratory report were explained in the English workbook and students used those to write their laboratory report. The experiment, The correlation between the height of a slope and distance travelled by a ball, was done in the English lesson and reports were written based on the instructions and steps provided in the English workbook and with constant monitoring and input from the lecturer. In effect, this means three experiments were actually done in the English lesson and one in the Physics lesson and four laboratory reports were written in the English lessons. 133 Class 2: 2008: modeling/imitation approach: In brief, Class 2 students receiving laboratory report instruction using the modeling approach received examples of laboratory reports: Titration and Volume and mass of an irregular shaped object. They had to read and analyse the reports based on structure, content, style, and language (this included past tense passive voice, discourse markers, action and hedging words). Criteria were determined. Then students received the instructions for an experiment and had to conduct the experiment in the English lesson. After the experiment they had to write a report (Volume and mass of an irregular shaped object: the displacement method). The instructions for another experiment (The effect of Protozoan Parasites on growth rates in tadpoles) were provided and students had to write a laboratory report based on the instructions, the model answer was provided afterwards for students to mark their own reports. The third report was based on the instructions for a Biology experiment (Osmosis) and students had to write a report, the model answer was provided for editing and correcting. To summarize, Class 2 (2008) analysed two model laboratory reports, did two actual experiments in the English lesson and wrote the reports on it and wrote one report based on instructions only. Model answers were provided for all experiments. Class 1 + 2: 2009: process genre approach: The 2009 groups received the laboratory report unit instruction using the process genre approach. The lessons started off with a brief observation- and- writing activity based on a burning candle. Students received a template of a laboratory report and had to complete the template after the simple candle experiment. Students also looked at the reasons for writing laboratory reports, collected vocabulary for action words in the Aim section, discourse markers and hedging words and did Past tense, Passive Voice activities. Then two examples of laboratory reports (Titration, Volume of an irregular object) were read and analysed according to guiding questions focusing on all the important issues of a laboratory report. No actual experiments were done in the English lessons. Students set up their own criteria. As a whole class activity an experiment was written together focusing on different steps (Volume of an irregular object: displacement method). As a next step students received the instructions and a model answer of a laboratory report of The effect of Protozoan Parasites in growth rates in tadpoles. Students had to determine the effectiveness of the report as compared to the instructions. Lastly, students wrote a laboratory report regarding Osmosis based on the instructions of the experiment and writing steps given in the workbook. They 134 received a model answer later to edit and correct their work. The last report to be written for the English laboratory report unit was based on an experiment conducted in the Physics lesson: Ohm’s Law. Students had to follow the steps in the guidelines to write the report after they did the experiment. In brief, this means students did not do actual experiments in the English lessons, they received two examples and two model answers and the writing steps were clearly outlined in the English workbook and followed in the lessons. All the laboratory reports used in this study were written after the experiments were conducted in the Physics, Biology and Chemistry practical lessons. Students had one week to work on the laboratory report in their own time before handing in to the Physics, Biology or Chemistry lecturer for grading purposes. The marks of the first laboratory reports (preintervention) were compared to the marks of the last laboratory reports (post-intervention) written in each of the three subjects mentioned. 135 4.6.2 PHYSICS 4.6.2.1 Physics laboratory report results of 2008 Class 1: process approach The tables and Bar graph illustrate the Class 1 pre-intervention and post-intervention results. Group Statistics Time of Assessment N Marks scored Pre Intervention Std. Deviation Mean 24 71.25 Std. Error Mean 8.999 1.837 Post Intervention 23 77.17 7.043 1.469 Table 11: Pre- and post-intervention Class 1 2008 laboratory report mean scores The pre-intervention mean scores (71.25%) on this table show that students started out with above average marks in their Physics laboratory reports. However, after the intervention there was an increase in their results as this table indicates: a difference of 5.82% on the mean scores for the pre- and post-intervention measurement. It was further tested if this difference is statistically significant and the results are shown in Table 12 below. Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Difference Sig. (2- F Marks Equal variances scored assumed Equal variances Sig. .313 t .578 -2.506 df tailed) Mean Std. Error Difference Difference Lower Upper 45 .016 -5.924 2.364 -10.685 -1.162 -2.519 43.303 .016 -5.924 2.352 -10.666 -1.182 not assumed Table 12: Independent Samples t-test for significant difference in Class 1 2008 Physics laboratory report mean scores The Independent Samples t-test was used to determine the significance of the difference. It is evident from the table above that p=0.016 is less than alpha=0.05, and therefore the null hypothesis was rejected at the 5% significance level. Thus the process approach intervention was effective in improving the performance of the 2008 Physics Class 1 students’ laboratory report writing abilities. The following graph illustrates the improvement of the 2008 Class 1 students, doing the process approach. 136 Figure 7: Bar graph: Pre- and post-intervention Class 1 2008 Physics laboratory report results The results on this graph show that 33% of students (8 students) managed to achieve Excellent results pre-intervention. However, the Excellent result increased post-intervention to 69%, translating to 16 students out of 23. Eleven students achieved Good pre-intervention (46%), with no Average results for post-intervention. No post-intervention marks were recorded for Average, Poor and Weak. This graph indicates that the process approach has contributed to an improved Physics laboratory report writing result for Class 1, 2008. 137 4.6.2.2 Physics laboratory report results of 2008 Class 2: modeling/imitation approach The tables and graph below illustrate the Class 2 pre-intervention and post-intervention results. Group Statistics Time of Assessment N Marks scored Pre Intervention Mean 19 Std. Deviation 51.58 Std. Error Mean 8.983 2.061 Post Intervention 13 80.77 9.541 2.646 Table 13: Pre- and post-intervention Class 2 Physics laboratory report mean scores Table 13 clearly shows that a mean score of 51.58% was achieved before the intervention was applied. As a result of the intervention the mean score was found to be 80.77%. There is a 29.19% difference in the pre- and post-intervention mean scores. It was further investigated if the intervention was effective and the results are shown in Table 14 below. Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Difference Sig. (2- F Marks Equal variances scored assumed Equal variances Sig. .068 t df .796 -8.805 tailed) Mean Std. Error Difference Difference Lower Upper 30 .000 -29.190 3.315 -35.961 -22.420 -8.703 24.873 .000 -29.190 3.354 -36.100 -22.281 not assumed Table 14: Independent Samples t-test for significant difference in Class 2 2008 Physics laboratory report mean scores It was investigated whether the difference between the pre- and post-intervention score was statistically significant. It is evident from Table 14 above that this difference was statistically significant, (p=0.00< alpha=0.00) meaning that the modelling intervention was very effective in improving the performance of the Class 2 students in Physics laboratory report writing. The Bar graph below shows the post-intervention improvement. 138 Figure 8: Bar graph: Pre- and post-intervention Class 2 2008 Physics laboratory report results Class 2 results have undergone quite dramatic changes. No pre-intervention Excellent mark was achieved by students, but there was an 77% (10 students) Excellent recording for postintervention. In addition, there were no Average or Weak results post-intervention. All postintervention results are in the Above Average, Good and Excellent section. Overall, the modeling/imitation approach has added value to Class 2 students’ laboratory report writing ability. 139 4.6.2.3 Physics laboratory report results of 2009 Class 1 + Class 2: process genre approach The tables and graph below illustrate the Class 1 and 2 pre-intervention and post-intervention results. Group Statistics Time of Assessment N Marks scored Pre Intervention Std. Deviation Mean 66 60.00 13.700 Std. Error Mean 1.686 Post Intervention 63 71.11 9.691 1.221 Table 15: Pre- and post-intervention 2009 Physics laboratory reports mean scores It was found that there was a difference of 11.11 in the mean scores as recorded before and after the intervention: process genre approach. The approach contributed to an improvement in the laboratory report writing skills of students in 2009. The following table will indicate whether the difference in mean scores was significant. Independent Samples t-test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Difference Sig. (2- F Marks Equal variances scored assumed Equal variances 4.029 Sig. t .047 -5.295 df Mean Std. Error tailed) Difference Difference Lower Upper 127 .000 -11.111 2.098 -15.263 -6.959 -5.337 117.231 .000 -11.111 2.082 -15.234 -6.988 not assumed Table 16: Independent Samples t-test for significance difference in 2009 Physics laboratory report mean scores The table above presented the following finding: since the p value was found to be 0.00, which is less than the level of significance (0.05), it was concluded at the 5% significance level that the process genre intervention was effective in improving the laboratory report writing performance of the 2009 Physics students. The effect is graphically illustrated in the following figure. 140 Figure 9: Bar graph: Pre- and post-intervention 2009 Physics laboratory report results The graph indicates that the 2009 Physics pre-intervention laboratory report results show a range from Excellent to Poor. The highest pre-intervention result was for Above Average (33% or 22 out of 66 students). There was a shift to Good and Excellent post-intervention with 43% (27 students), and 33% (22 out of 63 students) respectively. Fifteen (15) students received average results pre-intervention, which decreased to 2 post-intervention. No poor results were recorded post-intervention. The results indicate that the process genre approach was effective in improving the laboratory report writing skills of the majority of students with the biggest group achieving from Excellent to Above Average post-intervention. 4.6.2.4 Comparison of approaches: Physics Class 2, who did the modeling approach, managed to achieve a 78% for Excellent, which is quite remarkable considering that there were no pre-intervention Excellent results. 68% of Class 1 students (process approach) achieved Excellent, compared to the 36% of the process genre approach. The modeling approach produced the most Excellent results. However, if all the post-intervention results for Excellent, Good and Above Average of the three different approaches are added up respectively, the process and modeling approaches are equal at 100%, with the process genre approach lagging behind by only 5% at approximately 95%. Practically, this could translate into a conclusion that the three approaches were effective in improving students Physics laboratory report writing results. 141 However, the mean scores indicate that the modeling approach had the highest level of improvement. The differences in pre- and post-intervention mean scores are as follows: the process approach: 5.92%, the modeling approach: 29.19% and the process genre approach: 11.11%. The issue is that the pre-intervention results of the Class 1 students who did the process approach were already good (mean score of 71.25%) whereas the Class 2 students had a pre-intervention mean score of 51.58%. But, the post-intervention mean score of the modeling approach (80.77%) was still higher than the mean score of the process approach (77.17%). Therefore, the modeling approach could be argued to have been more effective than the process approach and the process genre approach in improving students’ physics laboratory report writing abilities. These results also suggest that the core issue is not necessarily which approach produces the best results, but rather that all three approaches have strengths that can all be employed to assist students to write better laboratory reports. This confirms the current notion of applying an eclectic approach that suits the level and needs of the students as well as the objectives of a programme. 142 4.6.3 CHEMISTRY 4.6.3.1 Chemistry laboratory report results of 2008 Class 1: process approach The tables and graph below show the results students achieved pre-intervention compared to the post-intervention results. Group Statistics Time of Assessment N Marks scored Pre Intervention Std. Deviation Mean 24 64.79 10.982 Std. Error Mean 2.242 Post Intervention 24 73.27 12.154 2.481 Table 17: Pre- and post-intervention Class 1 2008 Chemistry laboratory report mean scores The mean scores before and after the intervention, in Class 1 of 2008, were found to be 64.79% and 73.27% respectively. There was a difference of 8.48% in the mean scores as recorded before and after the intervention: process approach. It was further tested if this difference is significant and the results are shown in Table 18 below. Independent Samples t-test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Difference Sig. (2- F Marks Equal variances scored assumed Equal variances .239 Sig. .627 t -2.536 df tailed) Mean Std. Error Difference Difference Lower Upper 46 .015 -8.479 3.344 -15.210 -1.749 -2.536 45.535 .015 -8.479 3.344 -15.211 -1.747 not assumed Table 18: Independent Samples t-test for significant difference in Class 1 2008 Chemistry laboratory report mean scores This table indicates that the p value was 0.015, which is less than the level of significance (0.05). Therefore, the null hypothesis was rejected. It was concluded at the 5% significance level that the process approach intervention was effective in improving the laboratory report 143 writing performance of the Class 1 Chemistry students. The following Bar graph indicates the level of change among the Chemistry laboratory report writing results of Class 1 students. Figure 10: Bar graph: Pre-and post-intervention Class 1 2008 Chemistry laboratory report results A relatively high percentage (29% or 7 out of 24 students) for pre-intervention Excellent was recorded, but it increased by 21% to 50% (12 students) post-intervention. Interestingly, preintervention and post-intervention Good received exactly the same percentages (27%, 4 students). A similar situation occurred with Above Average with 35% (6 students). There was a very small recording (4% or 1 student)) for post-intervention Average and then the same percentage (4%) pre- and post-intervention Weak. No Poor results were recorded preand post-intervention. Class 1 post-intervention results indicate a slight improvement with regard to Chemistry laboratory report writing. 144 4.6.3.2 Chemistry laboratory report results of 2008 Class 2: modeling/imitation approach Table 19 below illustrates the results Class 2 students achieved pre-intervention compared to the post-intervention results. Group Statistics Time of Assessment N Marks scored Pre Intervention Std. Deviation Mean 22 65.68 9.167 Std. Error Mean 1.954 Post Intervention 23 69.48 7.309 1.524 Table 19: Pre- and post-intervention Class 2 2008 Chemistry laboratory report mean scores This table provides data with regard to pre- and post-intervention mean scores. The pre- and post-intervention scores were recorded as 65.68% and 69.48 % respectively. It was found that there was a difference of 3.80% in the mean scores as recorded before and after the intervention. A test for the equality of means was done to see if this difference is significant and the results are shown in table 20 below. Independent Samples t-test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Difference F Marks Equal .499 Sig. t .484 -1.540 df Sig. (2- Mean Std. Error tailed) Difference Difference Lower Upper 43 .131 -3.796 2.466 -8.769 1.176 -1.532 40.137 .133 -3.796 2.478 -8.805 1.212 scored variances assumed Equal variances not assumed Table 20: Independent Samples t-test for significant difference in Class 2 2008 Chemistry laboratory report mean scores This t-test to determine the significant difference in mean scores illustrates that the p value was 0.131, which is greater than the level of significance (0.05). It was therefore concluded 145 at the 5% significance level that the null hypothesis was not rejected. The modeling intervention was not effective at a statistically significant level in improving the laboratory report writing performance of the Class 2 Chemistry students. The Bar graph below indicates the changes observed as a result of the modeling intervention. Figure 11: Bar graph: Pre- and post-intervention C