Journal of Vocational Behavior 85 (2014) 75–84 Contents lists available at ScienceDirect Journal of Vocational Behavior journal homepage: www.elsevier.com/locate/jvb Career writing: Creative, expressive and reflective approaches to narrative identity formation in students in higher education Reinekke Lengelle a,⁎, Frans Meijers b, Rob Poell c, Mijke Post b a b c Faculty of Humanities and Social Sciences, Athabasca University, Canada School of Education, The Hague University of Applied Sciences, The Netherlands Department of Human Resource Studies, Tilburg University, The Netherlands a r t i c l e i n f o Article history: Received 1 March 2014 Available online 10 May 2014 Keywords: Creative writing Expressive writing Narrative career counseling LIWC Dialogical Self Theory Higher education a b s t r a c t This study investigates whether creative, expressive, and reflective writing contributes to the formation of a narrative career identity that offers students in higher education a sense of meaning and direction. The contents of writing done by students who participated in 2 two-day writing courses before and after work placements and of a control group were compared. Employers were also asked to evaluate students' performance. Writing samples were analyzed using the Linguistic Index Word Count program and an instrument based on Dialogical Self Theory. Work-placement self-reports were gathered, examined, and used as anecdotal evidence presented in the form of case studies. The results show that career writing can promote the development of career identity and holds promise as a narrative career guidance approach. © 2014 Elsevier Inc. All rights reserved. 1. Introduction Young people preparing for work need career guidance that will see them through unpredictable journeys. Traditional career-choice models are no longer working (Jarvis, 2014), as they do not successfully address the compounding factors of complexity, individualization, labor market insecurity, and the need for social and emotional competence in the workplace (Cherniss, 2000). As well, parents are frequently ill-equipped to help their children in making suitable career choices because of this increased complexity and the disappearance of the ‘grand narratives’ that used to guide choice making (Meijers, 2013). Studies also show that few students are intrinsically motivated in their studies and a majority have no idea of their career direction (Gatto, 2009; Holt, 1995). Dropout rates in European secondary and higher education are between 30 and 50% (Eurostat, 2008), which is partly a result of rather random educational choices. A Dutch study (Borghans, Coenen, Golsteyn, Hijgen, & Sieben, 2008) estimated that the societal costs of students taking longer to complete their studies as a result of unsuitable choices was 5.7 billion euros a year. Not only the ‘cafeteria of options’ (Guichard, 2009, p. 252) makes it difficult for young people to make choices but also the fact that the labor market has changed drastically in recent decades. There is a marked increase in contingent labor contracts and job insecurity (Goldstene, 2013). At the same time, the educational system has not kept pace with the resulting need for students to become more proactive; instead education continues to emphasize “rational–cognitive conceptions of knowledge and cumulative–linear models of learning” (Briton, 2012, p. 48). Employers are looking for intrinsically motivated employees who not only have technical skills and knowledge – which is where schools still focus most of their energy – but also possess the so-called ‘soft-skills’ (Hillage, Regan, Dickson, & McLoughlin, 2002; Lafer, 2004; Schulz, 2008). Indeed, the role of emotions in career learning is key for a number of reasons. First, there is a need to cultivate emotional competence in the workplace (Hochschild, 1983) because of the economy's focus on service provision. Second, affect ⁎ Corresponding author at: 8711 42 Avenue, Edmonton, Alberta, Canada. http://dx.doi.org/10.1016/j.jvb.2014.05.001 0001-8791/© 2014 Elsevier Inc. All rights reserved. 76 R. Lengelle et al. / Journal of Vocational Behavior 85 (2014) 75–84 effects choice making overall (Kidd, 2004; Meijers & Wardekker, 2002), and third, psychological factors must be attended to in the face of setbacks (Krumboltz, Foley, & Cotter, 2013) and insecurity (Pryor & Bright, 2011). Where guidance and education used to be focused on head and hands, there is now a requirement to also develop the heart (Claxton, 2006). We propose that career writing can help students cultivate affective development, improve intrinsic motivation, and contribute to their ability to make suitable choices. The idea posed by Lengelle and Meijers (2014) is that students must learn to construct flexible and emotionally salient personal narratives that replace the stable and standard ‘grand narratives’ of the past. In this qualitative study, we evaluate career writing's usefulness with a group of third-year Dutch bachelor's students. 2. Career writing Career writing is one of several narrative approaches to career construction (Cochran, 1997; Savickas et al., 2010) and draws its forms and practices primarily from the field of writing for personal and professional development (Bolton, 1999, 2010; Hunt & Sampson, 2002). It is intended to help individuals construct a career identity in narrative or poetic forms and does so by facilitating the exploration of life themes and life's disorientating dilemmas (Meijers & Lengelle, 2012). The three types of writing that fall under the umbrella of career writing are creative, expressive, and reflective writing. Table 1 shows the definitions and distinctions. Creative writing in the context of career writing refers to the writing of fiction or (fictional) autobiography, with the potential of gaining self-insight (Bolton, 1999; Hunt & Sampson, 2002). The idea here is that one's truths are often told in the form of “lies” made up to tell a story; the facts may not be accurate but the words reflect the thoughts and emotions associated with lived experience. Academics and practitioners of writing for personal development have found that fiction can be a way of exploring professional issues that are too problematic or not accessible enough to deal with in any other way (Bolton, 1994). Expressive writing research has been done in the past three decades into the therapeutic effects of writing in the face of job loss (Spera, Buhrfeind, & Pennebaker, 1994), workplace injustice (Barclay & Skarlicki, 2009) and other traumatic experiences. In this context, expressive writing refers to writing where an individual is encouraged to explore his/her deepest feelings and thoughts about an emotionally charged or negative life event (for an overview see Pennebaker, 2011). Reflective writing refers to putting on paper one's thoughts about life events in a way that can “take us out of our own narrow range of experience and help us to perceive experiences from a range of viewpoints and potential scenarios” (Bolton, 2010, p. 10). Reflective writing means one evokes the observer in one's work and words. This form of writing is in principle non-fiction and requires the questioning and de-construction of existing identifications that might stand in the way of one's agency (Lengelle & Meijers, 2009). In this context, existing identifications might refer to beliefs such as “I am not good at setting my own course …” or “My siblings are more successful than I am ….” Reflexive writing, an extension of reflective writing, involves ‘doubling the self’ (Hunt & Sampson, 2006, p. 4) and working directly with the self as multi-voiced, so that we are both ‘inside’ and ‘outside’ ourselves simultaneously while also “retaining a grounding in our familiar sense of self” (Bolton, 2010, p. 4). Where reflective writing refers to the presence of a conscious observer of our experiences, reflexive writing invites more voices to the table, not only to be observed but also to be embodied and tried out. In short, reflective writing allows us to watch the many (inner and outer) characters on the stage of our lives, while reflexive writing lets us get on the stage and try on the various selves. 3. Problem We proposed that 2 two-day career-writing courses – one before and one after a work placement – would foster the construction of career narratives among participants as compared with students who did not do the course (referred to here as controls). That career writing might indeed show beneficial change was based on research that showed that experiences alone do not lead to a person's learning and development (Bloch, 2005; Flum & Blustein, 2000), but that experiences and a dialogue about those experiences contribute to real career learning (Kuijpers, Meijers, & Gundy, 2011). Table 1 Creative, expressive, and reflective/reflexive writing. Category Creative Expressive Reflective/reflexive Brief definition The writing of fictional pieces for the purposes of personal and professional development. Fiction Artist Creativity Imagination Play exploration Superficiality, flights of fancy Writing about one's deepest thoughts and emotions surrounding a painful experience for the purpose of processing life events. Non-fiction Healer Expressiveness Openness Pain/resolution (re)connection Rumination, feelings of disempowerment and victimization Writing from life experience, reflecting to gain insight, constructing meaning and direction, and questioning pre-existing identifications. Non-fiction/inquiry Scientist/philosopher Reflexivity/examination structure Primary form Role or archetype Chief qualities Vital drive/goal Pitfall/dangers Insight/a sense of order Over-intellectualizing and rationalizing R. Lengelle et al. / Journal of Vocational Behavior 85 (2014) 75–84 77 4. Methods 4.1. Research design We explored the above research premise by analyzing texts gathered from two groups of third-year bachelor's students, both before and after the placement, using a linguistic analysis program or LIWC (developed by Pennebaker, Booth, & Francis, 2007) and a coding method (developed by Winters, 2012) based on Dialogical Self Theory (Hermans & Hermans-Konopka, 2010). We also asked employers to evaluate students in response to a 16-statement questionnaire. 4.2. Participants We recruited third-year bachelor's students from a Dutch university as this group had a scheduled five-month work-placement period. All students volunteered to participate, were asked for permission to report results, and signed to commit to participation as per research protocol in The Netherlands. We began with 20 participants in the experimental and 20 participants in the control group; we aimed for a group this size so we would have enough participation but keep the course to a maximum of 20 students. Sixteen members of the experimental group were able to take part in the first two-day course and nine completed all four days. The control group began with 19 active participants of which 14 submitted the required writing samples. The average age of participants was 23 years; there were eight females and one male in the experimental and 11 females and three males in the control group; participants, although Dutch speaking, came from a variety of cultural backgrounds. Most participants originally signed up to be part of the experimental group, however not all were able to take the course due to other commitments; many of the latter became part of the control group. Although all participants were recruited based on the expectations that they would do a work placement, two students from the experimental group completed final school projects within the period intended for work experience instead and we took that into consideration when we processed our results. We received permission from all participants to use names, but have abbreviated this information to preserve privacy. 4.3. The course The 2 two-day career-writing courses were aimed at getting students to enter into a dialogue about their work placements both before and after these took place through career writing exercises developed by Lengelle (Lengelle & Meijers, 2014). Each experimental-group participant received a booklet with instructions; writing exercises were explained thoroughly by the course instructor, and the writing was done within a face-to-face classroom setting. Fiction writing, poetry writing, journaling, and a method of written inquiry called “The Work” (Katie, 2002) were some of the activities that the students took part in; both courses were similar in content, and style, as was the order of activities completed. 4.4. Considerations for research It is important to note that the authors of this article were also involved directly in the study and anticipated positive results from the research. As we know, a researcher's partiality can affect results and therefore we mention this bias. One of the authors designed the course and taught it, while another observed the process. The coding on the data was done by two of the authors and later compared. As well an outside researcher who had designed the coding method, gave her advice and this likely increased the internal reliability. By using several methods to look at the same data we aimed to limit the effects of bias or expectation. We looked at whether results pointed in the same direction (or not). The DST coding was developed to analyze career guidance conversations (Winters, Meijers, Lengelle, & Baert, 2012; Winters et al., 2013) and LIWC (Pennebaker et al., 2007) was used in numerous larger-scale studies to examine patterns and trends in word use. 4.5. Data collection In total, two writing samples were collected from each of the 14 control group participants — one before work placements and one afterwards and six samples were collected from all nine of the experimental group students who completed both courses. The experimental group not only wrote before and after the work placements, but they also did so at the start, during and at the end of the first two-day course and at the start, during, and at the end of the second two-day course for a total of six times. The prompt for the writing samples collected from both groups was identical for all writing sessions; the wording was as follows: “Please write for 20 minutes about the work placement you're going to do/you just did and the significance of that work placement for your career and course of study.” For the two students who did not end up going on a work placement, we adjusted the prompt slightly and asked them to concentrate on the same question, but to replace the word “work placement” with “completing my final bachelor's project”. No further instructions were given. After completing their placements, students were also asked to hand in their official work-placement reports and we approached employers for their feedback. Of the experimental-group participants, we received four work-placement reports (of the seven students who actually did work placements) and five employer assessments. Of the original 19 from the control group, we received 14 work-placement reports and 13 employer assessments. 78 R. Lengelle et al. / Journal of Vocational Behavior 85 (2014) 75–84 4.6. Instruments and procedures As mentioned before, two instruments were used to analyze the writing samples that we received from students. The first was the Linguistic Index Word Count (LIWC) program developed by Pennebaker et al. (2007) with which we looked at the number and division of pronoun changes, emotion as well as insight and causal words. The specific types of words and the dynamics between them helped us identify whether a beneficial narrative was developing or not. A positive change in students' writing would show an increase in insight, causal and cognitive words, as well as in the presence of emotion words, and pronoun switching. A narrative that shows beneficial change contains more positive (e.g. accepting, good, fun) than negative emotion words (e.g. afraid, awful, tense) but does contain some negative words. As well, such a narrative shows pronoun switching, which means that if students use not only “I” but also “he, she, you”, alternatively, they are on the right path to constructing a beneficial piece. Those with healthier perspectives tend to “say something about their own thoughts and feelings in one instance and then explore what is happening with other people before writing about themselves again” (Pennebaker, 2011, p. 13). The second approach was a coding method based on Dialogical Self Theory (DST) (Hermans & Hermans-Konopka, 2010) and developed for career guidance conversations by Winters (2012); see also Winters et al. (2012) and Winters et al. (2013). Evidence of a developing career narrative in DST shows up if there are I-positions present (e.g. I wanted to work for a large company), expanded I-positions (e.g. I want to work for a large company because there you are able to network better), meta positions (e.g. I realize I am someone who thrives in a large company because there is a diversity of people to learn from), and finally promoter positions (e.g. My next step is to be more outgoing in approaching those colleagues whom I admire). In short, I-positions show what is important to an individual, expanded I-positions show what is important by extension or where I-positions are valued in other contexts. Meta positions show the ability to observe oneself and combine I-positions in integrated ways and promoter positions show an ability to take concrete action. Writing samples were coded by two of the researchers and compared for consistency and validity of coding. A 16-item statement-style questionnaire was also sent to employers so they could evaluate the student who had worked for them. The 16 statements represent core qualifications of Dutch secondary vocational education (Nijhof & Streumer, 1998). Statements included items like, “the student can do the work expected of him/her independently”, “the student listens to explanations given”, “the student enjoys the work”, “the student can work with others” and could be responded to with either, “excellent, very satisfactory, satisfactory, almost satisfactory, and unsatisfactory.” The questionnaire was not shared with students and it was not used to evaluate their academic performance or grade them. 5. Results The experimental and the control groups were comparable, both in gender, age, as in the writing that they did at the start (Lengelle, Meijers, Poell, & Post, 2013). The only notable exception is that the experimental group used more pronouns at the outset. Word count is indicated in the table below to give a sense of the size of the writing samples and the number of particular words relative to that. The groups differed slightly with regard to word count at the start, even though participants all wrote for the same length of time. We believe this can be attributed to the experimental group having to handwrite everything in class, whereas most controls typed their work on a computer keyboard. Interestingly, the experimental group did write more words by the end of the final course. 5.1. LIWC results Table 2 contains the results of the LIWC analysis whereby we looked at the average number of times that a particular type of word appeared in the writing sample of the members of each group. Note that this is not a quantitative study and numbers are not statistically significant, but are used here to point out changes. Both groups used the “I” pronoun the most in their work, with the control group using “you” once on average per writing sample. The experimental group increased their use of “you” at least once per writing sample by the end of the second writing course; their use of other pronouns also went up by an average of one additional pronoun (e.g. anyone, someone, its). The latter shows a very modest increase in pronoun switching. The more notable difference between the groups was the increase in anxiety, anger, and sad words among the control group members after work placements, even as general negative words (e.g. awful, disappointment, mess) seem to have decreased slightly. In the experimental group, we see a predominant increase in positive emotion words and an initial increase in some negative emotion words, but no further increase and no increase in the categories anxiety, anger, or sadness. There was no notable change in causal words (e.g. affects, change, because) in either group, however those in the experimental group showed cognitive (e.g. contemplate, correlate, acknowledge) word scores that seemed to stay higher. Both groups dipped slightly in their use of cognitive words after the work placements, but the experimental group participants ended on a higher note. This was also the case with regard to insight words; on average, the members of the experimental group all used at least one additional insight word as compared with controls, who showed no change. 5.2. DST results We anticipated that students who participated in the 2 two-day course would show more meaning and direction in their narratives as shown in the DST analysis by increasing and expanding I-positions, meta- and promoter positions. At the outset, 79 R. Lengelle et al. / Journal of Vocational Behavior 85 (2014) 75–84 Table 2 LIWC results before and after work placement and before, during, and after the career-writing course. Control group Word count Pronouns I We You Other pronouns Affective words Positive emotion Negative emotions Anxiety Anger words Sad words Cognitive words Causal words Insight words Experimental group Before After Before day 1 After day 2 Before day 3 After day 4 421 10 9 0 1 0 2 1 1 0 0 0 8 1 3 423 10 8 0 1 0 2 1 0 1 1 2 7 1 3 297 12 10 0 0 0 3 1 0 0 0 0 9 0 3 232 13 11 0 0 0 4 3 1 0 0 0 9 1 3 334 13 11 0 0 0 3 2 1 0 0 0 7 1 3 449 13 11 0 1 1 4 2 1 0 0 0 9 1 4 both groups wrote with similar numbers of the positions, though the experimental group showed a greater increase than the control group after the first two-day writing course (Lengelle et al., 2013). Table 3 shows the increases or decreases of both groups between the first measurement in January and the last one in June, when all students had finished their work placements and projects. If there was no change in the number of a particular position between January and June it is marked with a 0, decreases are marked by a −, increases by a +. For the experimental group double −− or ++ means that there was a change of more than three either down or up. So, if a student started with four meta positions and ended with seven or more positions, it would be marked thus: ++. However as the control group wrote only once in January and once in June, and the experimental group wrote three times in January and three times in June, the number of words produced by and analyzed for the control group was about a third of what was analyzed for the experimental group. Therefore, we decided to mark any increase or decrease of positions of more than two positions (e.g. a student started with two and ended with four or more meta positions, would be noted in the table as ++). An increase of one meta or promoter position might be coincidence, but a jump or decline of two is more likely to indicate an actual change. What we see in Table 3 is that those students who took part in the writing course tended to show an increase in meta- and promoter positions. The control group results show an increase in I and expanded I positions, but a decrease Table 3 Experimental and control group change in positions between January and June. Student I I's Experimental group (1) D.B. (2) S.B. (6) E.v.K. (7) T.K. (8) F.K. (p) (9) T.M. (10) R.M. (12) K.P. (p) (15) K.V. − − −− − − −− − − − + − − − + − − + + + Control group (22) L.P. (23) O.T. (24) D.D. (27) R.T. (28) T.Z. (31) K.H. (32) C.vd.K. (33) J.L. (34) U.G. (35) P.R. (36) T.H. (37) J.v.E. (38) S.R. (39) B.v.H. + + − − − − + − + − + + − 0 + − − + − + + + + + − + + 0 + − − + − − + − − + + + − − + + + + Meta Promoter ++ + 0 ++ −− + ++ ++ + − ++ + + ++ + + − − + − − + − 0 + 0 − − − + − − − 0 0 − − − −− − − − − − − − + − + − + − + 80 R. Lengelle et al. / Journal of Vocational Behavior 85 (2014) 75–84 or no change in promoter positions. Several control-group students showed an increase in formulating meta positions, but more than half showed a decrease. 5.3. Employer evaluations In response to the 16 statements that employers were asked to complete in order to evaluate the students who had worked for them, there were a total of 208 responses expected for the control group and 80 in the experimental group. However, two work-placement mentors of the control group each neglected (likely forgot) to fill in one question, therefore the total number of responses was 206. The experimental group scored “excellent” (i.e. the highest category) 67 times out of a total of 80 answers as seen below in Table 4 where absolute numbers are shown. This group had no scores in either the almost or the not satisfactory category. And although the control group also scored satisfactory or above on most workplace tasks, their scores showed fewer than half in the “excellent” category and a wider spread over very satisfactory and satisfactory. We have included two case studies from each group. The selection of cases from the experimental as well as the control groups includes only those students who handed in their work-placement reports, so that along with the writing samples, there was more material to compare and contrast. The inclusion of students' own words is fitting in the context of this writing study and gives a fuller and more concrete picture of what their narrative development entailed. 5.3.1. Experimental group: case study 1 D.B. did a work placement at a childcare organization where he had to set up a plan to check the quality of the care and perform duties as a communication professional (e.g. writing press releases, creating promotional material, doing market research, creating ads etc.). What we noticed in particular about D.B.'s writing samples from course days one and two as compared to course days three and four is that he developed many more expanded I-positions and meta positions. He wrote about gaining an expanded range of skills at work, but he did not focus only on the tasks associated with being a communication professional (e.g. writing press releases, contacting other parties). He also mentioned the importance of being able to contribute his own vision and ideas, his ability to be light and make his colleagues laugh, and the insight that his experiences at work had led to more integration of his school-based learning. He also explored his doubts and uncovered several personal patterns. For instance, one thing he noticed was his tendency to anticipate a negative event when things were going well. He called this his “yin-and-yang thinking” and described how he might be getting in his own way by always expecting the other shoe to drop. Although there was a slight decrease (from four to three) in the expression of promoter positions as compared with the initial course, he articulated a sense of continuing to expand his work experience with the goal of becoming a “full-fledged career professional”. In his official work-placement report, D.B. listed his goals, which included increasing his confidence level by taking on concrete tasks, thus expressing promoter positions. 5.3.2. Experimental group: case study 2 T.K. had just started her work placement with a broadcasting company that created radio programs when she embarked on the first of the 2 two-day writing courses. She tended to be philosophical in her all writing and she expressed meta positions readily. However, the meta positions that she expressed at the end of her work placement were more abundant and had a deeper quality to them. For instance, at the beginning of the study, she wrote about working with people who were “… nice, kind, and honest”. After the work placement, in the middle of the second two-day writing course, she wrote, “… the feeling of trust they had in me was what had me doing everything with pleasure. It seems that at first I have a strong need to feel that I belong, that they like me, that they give me the sense that I have really contributed something.” A metaphor that T.K. explored for herself by the end of the course, also shows a meta position that she expands upon, as if to explain to herself how her career path is taking shape, “A path is sometimes invisible, hidden in the light of dusk, hidden, but suddenly it becomes visible through light that arrives. That light has arrived on my path, it flickers now in any case, not strong, but also not weakly.” T.K. expressed two promoter positions in the third writing sample at the end of day two of the courses and four promoter positions in the writing samples that she did as part of the second course (days three & four). This more action-able attitude expressed itself in practical observations about her tasks in making radio programs but also in other insights. “During the work placement, I learned to accept things — you do something, it's done and then you go on to the next broadcast. You don't spend too much time thinking about it, otherwise you lose sight of the future.” In her work-placement report, she reiterated this insight into Table 4 Workplace evaluations for experimental and control groups. Evaluation Excellent Very satisfactory Satisfactory Almost satisfactory Not satisfactory Total Experimental group Control group (n = 5) (n = 13) 67 4 9 0 0 80 73 84 40 7 2 206 R. Lengelle et al. / Journal of Vocational Behavior 85 (2014) 75–84 81 a sound bite of advice to herself, “Don't lose yourself in the observations, but do something with them.” The promoter positions she expressed at the start of the study were also concrete, but they had more of an air of wishful thinking to them, “travel to places, start a café with good music and honest food.” In her work-placement report she gave many concrete examples of tasks that she had learned to do (e.g. using a computer program to create radio fragments to broadcast; phoning people to do interviews with and finding out about their expertise; writing press releases). Her promoter positions in that report were particularly focused on the insight that her shy and modest approach to her work and initiative taking could be balanced by being more bold and coming forward earlier with her creative ideas. In T.K.'s case, her tendency to philosophize seemed to be balanced by the experiences that she gained during her work placement. In her case, writing alone, as Pennebaker suggests (2011), if done in the spirit of philosophizing, could have turned into a form of ruminating and this may be more likely among people who already tend to “over think” things as T.K. observed about herself. Writing about something as concrete as a work placement might be particularly useful, whereas writing to analyze or reflect without this foothold might mean that writings' benefits could be lost. Students who tend to be less self-aware to begin with, might in that sense have more added benefit from writing, whereas more self-aware students such as T.K. benefit from writing about the doing. 5.3.3. Control group case studies To make a fair comparison possible between the two groups, we chose case studies from the control group of students who had handed in writing samples, work placement reports, and employer assessments. We considered this evidence of a commitment to the study, comparable to the experimental group. We chose the control-group students who had moved most in the direction that we hypothesized for the experimental group. 5.3.4. Control group: case study 1 C.vd.K. worked at a marketing firm as a communication professional helping to make suggestions about an update of the company's Intranet. In both her writing sample and work placement report she spoke about the insight that this work placement had taught her that she didn't want to continue to work in marketing, but that she was determined to be in a profession where she could help others. In her writing sample, she wrote, “The point is that I want to feel useful to other people. I really want to make a difference to other people.” This was a meta position — she had “learned something about myself.” Compared with the writing sample that she submitted at the start, this was something that came out of the work placement and she articulated for herself. The work placement report focused for the most part on concrete tasks that she had undertaken and practical insights, such as determining that she would be helped by becoming a better planner and remain more focused on the tasks that belonged to her, instead of being distracted by her tendency to help others when they interrupted her work flow. 5.3.5. Control group: case study 2 J.v.E. worked in the commercial sector during his placement as a communication professional in training. He did not describe the company he worked for in any detail or list the goal of their activities. His own learning aim was to improve his English-speaking skills and to receive feedback without defensiveness. The meta position that he expressed was the discovery that he wanted to work for a non-profit organization as he viewed the private sector as a place, “… where you have to go over dead bodies to achieve certain targets and that is not, in my view, how I'm put together.” In his work-placement report he expressed the insight that in receiving feedback he had the tendency to react too quickly to the person providing the feedback and that it was a learning objective for him to take the feedback in and “allow the person to finish speaking” and even to “ask questions so that the feedback can be deepened”. He listed the five rules of receiving feedback, which included the final one, “to do something about it” but he did not describe in detail what feedback was given that could lead to improvement in his performance. His employer assessment showed excellent and good scores, which was reflected in the comment that he mostly received positive feedback in his work doing interviews, taking notes, and taking part in meetings. 6. Discussion The results of the LIWC and DST analyses, the employer assessments, as well as the case studies show that there was indeed a difference between the experimental and control groups and that this change was in the anticipated direction. Those who did work placements and participated in career writing, expressed an increase in meta- and promoter positions overall. They also showed a more sustained increase in cognitive words and an increase in insight words and were evaluated more positively by their employers, though we are not claiming that the changes were numerically significant. It is also noteworthy that control-group students showed an increase in negative emotion words, including anxiety, anger, and sadness. This was not the case in the experimental group. And although studies show that the presence of negative emotion words is not necessarily a bad sign, as it signals the realization that life is not as ideal as one often hopes, an increase in negative emotion words without a rise in positive emotion words to accompany them is often an indication of a narrative that is not deemed beneficial (Pennebaker, 2011). Pennebaker (2011) also describes the importance of being in the process of constructing a story but not repeating one that is characterized by negative experiences as the latter resembles rumination. It is only when a story changes and moves from a tale of woe to a tale of hope that it serves the individual (Lengelle & Meijers, 2014). While the writing samples from the experimental group participants did show negative emotion words, the number of positive words increased also, indicating that on-the-job experience might bring up fear, doubt, and trepidation but that those feelings and responses can be mediated through an effective 82 R. Lengelle et al. / Journal of Vocational Behavior 85 (2014) 75–84 dialogue. Here effective refers to both emotions being felt and thoughts being articulated in a way that help individuals make sense of their experiences. The case studies further illustrate the results that we saw through the LIWC and DST analyses and employer evaluations more concretely. The experimental group showed a greater ability to identify what they had learned, from both inspiring as well as challenging experiences. They also examined more deeply the roles that they had played in their workplaces. And although both groups were focused on task-oriented skill development, those from the experimental group were more able to articulate specific self-insights. Even the two students who had done career writing but had not done work placements, showed an increase in either meta- or promoter positions; they identified vulnerabilities, and wrote about the learning that they had done as a result of both positive and negative experiences. The students in the experimental group also reported feeling ‘connected’ to one another, which may have contributed to the increase in positive-feeling words in this group. While they had encountered the same individuals who took part in the career-writing courses in earlier classes in their degree program, there was no capstone project or group gathering where they could reflect on the reality of completing their studies. In other words, no structured group setting in which to talk about work experiences, let alone express their uncertainties or explore gaps in their abilities to shape their careers. However, as seen in previous studies, reflection coordinated by college or university instructors doesn't necessarily lead to an insightful or helpful dialogue (Winters et al., 2012). Research would have to be done into other methods (e.g. storytelling, narrative group counseling) to see if similar results could be achieved with other group-focused narrative approaches. In the work-placement reports, there was evidence of new skills being acquired and learned by both groups, but the writing samples show that the experimental group reflected on more internal skills, such as reflecting on their identities more consciously and knowing where their weak spots were. The DST analysis shows a similar movement towards a more conscious (meta) positioning of attitudes and a subsequent ability to consider action (promoter) more readily. It is particularly relevant too that control group students more frequently expressed promoter positions as they related to specific skills (e.g. better planning, higher levels of competence in task-based skills like writing press releases and keeping meeting notes) while the experimental group more often explored learning-based themes like how to work more effectively with criticism, why being considered a valued colleague was useful and meaningful, and what kind of work environment suited their temperaments. This is a qualitative difference and points more to the development of competencies deemed necessary to succeed in the twenty-first century, which might explain in part why there was no increase in anxiety, anger, or sadness among this group. Those who only did work placements and did not receive career guidance through career writing did develop more expanded I-positions but decreased overall in meta- and promoter positions. This is an important finding because it indeed shows that students don't learn from experience alone as mentioned before. There must, in addition to hands-on experience and initial responses to that, be a dialogue about those experiences. It is precisely when this happens in combination, that we see students becoming truly reflexive and able to envision and plan actions in their careers (Kuijpers et al., 2011). The current study seems to indicate that work experience and standard reflection exercises (e.g. self reports) can help students develop different and expanded perspectives (i.e. expanded I-positions), but that they are not helped greatly by work experience alone to become truly reflective or to envision and plan their next steps. 6.1. Limitations Although this qualitative study shows promise that career writing is a viable narrative approach, the sample size was too small to draw final conclusions about its merits. There are also other weaknesses in our study and room for improvement in our methods. We did not, for instance receive all the items that we hoped to gather from students, although we were in touch with participants and requested missing pieces several times. Two students, who were very committed to the course work, did not hand in work-placement reflection reports; perhaps this is a sign that they felt that these were redundant – as research by Mittendorff (2010) and Mittendorff, den Brok, and Beijaard (2010) shows – or it could be a sign of waning interest in what they experienced during the courses. We anecdotally report also that two of the experimental group students told us that deeper questions had come up for them and that this was both exciting and disorienting. The realization that skills development on the job would not be enough and that they were being asked to look more deeply at themselves, might be experienced as a rude awakening or at the very least a wake-up call. And as we know from other studies, humans often prefer confidence borne of ignorance to being faced with conscious uncertainty (Hermans & DiMaggio, 2007). We also feel that follow-up interviews with participants from both groups might have added to the quality of our case-study reports and given more credence to our research. We also found that writing was not for everyone; two students who did not complete the second two-day course reported “not finding it useful” and decided that other activities had a higher priority than completing the study that they had signed up for. To evaluate career-writing's potential, we would also have to get more feedback from those who did not connect with the process and activities. 6.2. Recommendations What makes career writing attractive, as an alternative to one-on-one career counseling, is that it can be done in a classroom setting with a group of students, where they benefit from the presence of their peers. As such, it would be a method that schools and universities could implement as they realize more and more that they have a role to play in career guidance (Company, 2009). Their choice could be motivated in part by research as this, but also by related factors. For instance, the variety of stories R. Lengelle et al. / Journal of Vocational Behavior 85 (2014) 75–84 83 written and shared in a classroom setting becomes food for conversation. What students write and share is often cause for palpable excitement in the room. If intrinsic motivation and ‘soft skills’ are what schools and employers are after, this is one way to stimulate the internal and external dialogues that might foster their development. In short, career writing is both a time and cost effective approach without being automated or mechanical as many career tests are. Considering dwindling budgets and the marketization of career services, which has led to further impoverishment of what is available in schools (Hughes, Meijers, & Kuijpers, 2014), career writing might also be a service that can be affordably offered in or outside of schools. Parents might see such a course as they would a computer or driving course — something that gives their child additional or necessary skills to function in life. That said, it is important that teachers and those providing guidance services do not turn career writing into yet another educational hoop that students must jump through. We posit that its appeal and potential for enriching students' growth is made possible when it is taught with an appreciation of the creative process (e.g. the focus is not on grading or a ‘right’ outcome) and when those guiding students have a genuine interest in students' lives and an intention to be useful. Career professionals with an interest in narrative approaches can learn a lot about career-writing in a relatively short time; two days of thorough instruction and personal practice can already help some to get started. As more is written about the method and colleagues exchange experiences with one another, new exercises and insights will emerge. Whether professionals use it as their core method or add it to their existing practice is not as important as their experience of connecting with the approach themselves and imagining how it might even affect or inspire their own questions about meaning and direction. 6.3. Further research More research is needed to explore the ways in which career writing might play a role as a narrative-learning approach. Future research projects might include seeing whether an online course would foster career learning or whether students who did career writing instead of their regular career-guidance program (e.g. writing reflection reports; talking to work-place mentors) are better off. It would also be valuable to see how this method works for males as opposed to females, whether it is as suitable in other cultural contexts, if long-term benefits are apparent, and which groups are best served. Could career writing, for instance, be suited to working with long-term unemployed people, those retiring and wanting to revision their lives, or used to guide employees facing restructuring? We look forward to future studies to answer these questions. 7. Conclusion Career writing holds promise in helping students respond to the compounding issues of complexity, individualization, and insecurity on the labor market. It appears to help young people begin a dialogue about work experiences so that they may identify what motivates them intrinsically, what steps they might take to explore the territory of work, and why they might be interested in developing social and emotional skills both for themselves (e.g. for choice-making or navigating career direction) and in the workplace (e.g. to become a more competent employee in the service economy). Career writing engages them creatively, inviting both emotional and cognitive explorations of what is meaningful to them, how they might serve others, and what they might do to set a flexible course. References Barclay, L., & Skarlicki, D. P. (2009). Healing the wounds of organizational injustice: Examining the benefits of expressive writing. Journal of Applied Psychology, 94, 511–523. Bloch, D. P. (2005). Complexity, chaos, and nonlinear dynamics: A new perspective on career development theory. The Career Development Quarterly, 53, 194–207. Bolton, G. (1994). Stories at work: Fictional critical writing as a means of professional development. British Educational Research Journal, 20, 55–69. Bolton, G. (1999). The therapeutic potential of creative writing: Writing myself. London: Jessica Kingsley Publishers. Bolton, G. (2010). Reflective practice: Writing & professional development. London: Sage Publications. Borghans, L., Coenen, J., Golsteyn, B., Hijgen, T., & Sieben, L. (2008). Voorlichting en begeleiding bij de studie- en beroepskeuze en de rol van arbeidsmarktinformatie [Information and guidance in study and career guidance and the role of labour market information]. Maastricht: University of Maastricht/ROA. Briton, D. (2012). Making sense of why some learners choose failure over success. Proceedings of 31st annual conference of the Canadian Association for the Study of Adult Education (pp. 48–52). Ontario and Waterloo: Wilfred Laurier University and University of Waterloo. Cherniss, C. (2000). Social and emotional competence in the workplace. In B. Reuven, & J. Parker (Eds.), The handbook of emotional intelligence: Theory, development, assessment, and application at home, school, and in the workplace (pp. 433–458). San Francisco, CA: Jossey-Bass. Claxton, G. (2006). Expanding the capacity to learn: A new end for education? [PDF document]. Retrieved from. www.tloltd.co.uk/downloads/BERA-KeynoteUpdate-Feb10.pdf Cochran, L. (1997). Career counseling: A narrative approach. London: Sage Publications. Company, F. J. (2009). Vocational guidance and career counselling in the European Union: Origins and recent trends. In R. Maclean, & D. N. Wilson (Eds.), International handbook of education for the changing world of work. Bridging academic and vocational learning (pp. 2313–2329). New York: Springer. Eurostat (2008). Europe in figures. Eurostat yearbook 2008. Luxembourg: Office for Official Publications of the European Communities. Flum, H., & Blustein, D. L. (2000). Reinvigorating the study of vocational exploration: A framework for research. Journal of Vocational Behavior, 56, 380–404. Gatto, J. T. (2009). Weapons of mass instruction. Gabriola Island, Canada: New Society Publishers. Goldstene, C. (2013). The emergent academic proletariat and its shortchanged students. Dissent: A quarterly of politics and culture (Retrieved from http://www. dissentmagazine.org/online_articles/the-emergent-academic-proletariat-and-its-shortchanged-students). Guichard, J. (2009). Self-constructing. Journal of Vocational Behavior, 75, 251–258. Hermans, H. J. M., & DiMaggio, G. (2007). Self, identity and globalization in times of uncertainty: A dialogical analysis. Review of General Psychology, 11, 31–61. Hermans, H. J. M., & Hermans-Konopka, A. (2010). Dialogical self theory. Positioning and counter-positioning in a globalizing society. Cambridge, UK: Cambridge University Press. Hillage, J., Regan, J., Dickson, J., & McLoughlin, K. (2002). Employers skill survey 2002. Research report 372. Nottingham: UK: DfES. 84 R. Lengelle et al. / Journal of Vocational Behavior 85 (2014) 75–84 Hochschild, A. (1983). The managed heart: Commercialisation of human feelings. Berkeley, CA: University of California Press. Holt, J. (1995). How children fail. (rev. ed.). Reading, MA: Perseus Books. Hughes, D., Meijers, F., & Kuijpers, M. (2014). Testing times: careers policies and practices in England and The Netherlands (unpublished paper). Hunt, C., & Sampson, F. (2002). The self on the page: Theory and practice of creative writing in personal development. London: Jessica Kingsley Publishers. Hunt, C., & Sampson, F. (2006). Writing: self & reflexivity. New York: Palgrave. Jarvis, P. (2014). Career development: Key to economic development. In B. C. Shepard, & P. S. Mani (Eds.), Career development practice in Canada: Perspectives, principles, and professionalism (pp. 55–75). Toronto, Canada: CERIC. Kidd, J. (2004). Emotion in career contexts: Challenges for theory and research. Journal of Vocational Behavior, 64, 441–454. Krumboltz, J. D., Foley, P. F., & Cotter, E. W. (2013). Applying happenstance learning theory to involuntary career transitions. The Career Development Quarterly, 61. http://dx.doi.org/10.1002/j.2161-0045.2013.00032.x. Kuijpers, M., Meijers, F., & Gundy, C. (2011). The relationship between learning environment and career competencies of students in vocational education. Journal of Vocational Behavior, 78, 21–30. Lafer, G. (2004). What is ‘skill’? Training for discipline in the low-wage labour market. In C. Warhurst, I. Grugulis, & E. Keep (Eds.), The skills that matter (pp. 157–171). Basingstoke: Palgrave Macmillan. Lengelle, R., & Meijers, F. (2009). Mystery to mastery: An exploration of what happens in the black box of writing and healing. Journal of Poetry Therapy, 22, 59–77. Lengelle, R., & Meijers, F. (2014). Narrative identity: Writing the self in career learning. British Journal of Guidance and Counselling, 42, 52–72. Lengelle, R., Meijers, F., Poell, R., & Post, M. (2013). The effects of creative, expressive, and reflective writing in career learning. Journal of Vocational Behavior, 83, 419–427. Meijers, F. (2013). Monologue to dialogue: Education in the 21st century. International Journal for Dialogical Science, 7(1), 1–10. Meijers, F., & Lengelle, R. (2012). Narratives at work: The development of a career identity. British Journal of Guidance and Counselling, 40, 157–177. Meijers, F., & Wardekker, W. (2002). Career learning in a changing world: The role of emotions. International Journal for the Advancement of Counselling, 24, 149–167. Mittendorff, K. (2010). Career conversations in senior secondary vocational education. Eindhoven, The Netherlands: University of Technology Eindhoven. Mittendorff, K., den Brok, P., & Beijaard, D. (2010). Career conversations in vocational schools. British Journal of Guidance & Counselling, 38, 143–165. Nijhof, W. J., & Streumer, J. N. (1998). Key qualifications in work and education. Dordrecht, the Netherlands: Kluwer Academic Publishers. Pennebaker, J. (2011). The secret life of pronouns. New York: Bloomsbury Press. Pennebaker, J., Booth, R., & Francis, M. (2007). Operator's manual, Linguistic Inquiry and Word Count: LIWC 2007. Austin, TX: LIWC.net. Pryor, R. G. L., & Bright, J. E. H. (2011). The chaos theory of careers. New York and London: Routledge. Savickas, M. L., Nota, L., Rossier, J., Dauwalder, J. P., Duarte, M. E., Guichard, J., et al. (2010). Life designing: A paradigm for career construction in the 21st century. Journal of Vocational Behavior, 75, 239–250. Schulz, B. (2008). The importance of soft skills: Education beyond academic knowledge. NAWA Journal of Language and Communication, 2(1), 146–154. Spera, S. P., Buhrfeind, R. D., & Pennebaker, J. W. (1994). Expressive writing and coping with job loss. Academy of Management Journal, 37, 722–733. Winters, A. (2012). Career learning in vocational education: Guiding conversations for career development. Leuven, Belgium: Katholieke Universiteit Leuven. Winters, A., Meijers, F., Harlaar, M., Strik, A., Kuijpers, M., & Baert, H. (2013). The narrative quality of career conversations in vocational education. Journal of Constructivist Psychology, 26, 115–126. Winters, A., Meijers, F., Lengelle, R., & Baert, H. (2012). The self in career learning: An evolving dialogue. In H. Hermans, & T. Gieser (Eds.), Handbook of dialogical self theory (pp. 454–469). Cambridge, UK: Cambridge University Press.