Designing Your Thesis/Dissertation Proposal and Converting Them Into Full-Blown Research Projects

Assistant Professor 4 Chief, Graduate School Research Office Polytechnic University of the Philippines

Elmer G. De Jose, MAP, MMgt.

GOALS OF RESEARCH
To obtain new information To measure certain specific characteristics of certain population To observe some phenomena (and causes of these phenomena) in the environment To assess the outcomes of some programs or activities being undertaken

What is Thesis Writing?

- formal, systematic and intensive process of carrying a scientific method of analysis - generally looks into original or raw data. - requires independent thinking both in perceiving and analyzing problems and in finding solutions to them

A question strikes the mind of the researcher. The researcher converts the question into a clearly stated research problem. The researcher makes a sound guess or a temporary answer to the problem. This is called hypothesis.

Recommendations.

A conclusion is drawn. The hypothesis is tested whether it is true or not. The facts are subjected to analysis to comprehend their meaning. The researcher gathers the facts. Literature are searched to look into some relevant facts and possible answers. The researcher designs a methodology to be used in terms of the goal of the research and the needed facts.

THE PROBLEM AND ITS BACKGROUND

It is in this section where you discuss the overview of your research problem, the factors that lead you to the conceptualization of the problem and the justification for considering your chosen topic as a problem for research.

Criteria for the selection of a research problem

Internal/Personal
Training and personal qualification of the researches Interest, intellectual curiosity and intrinsic motivation Costs Time

Criteria for the selection of a research problem

External
Novelty of the problem Significance in the field Availability of data and methods Work conditions Hazards and handicaps

Theoretical/Conceptual Framework
Concepts - abstractions; ideas Theories - sets of interrelated concepts, constructs, definitions and propositions that present systematic view of the phenomenon by specifying relationships among variables with the purpose of explaining and predicting the phenomena.

Theoretical/Conceptual Framework

Framework - basic supporting part or structure Conceptual/Theoretical Framework made up of concepts and theories that form the basis of the study. A schematic diagram that shows the relationship of the variables in the study.

clearly stated goals and purposes (policy manual) Relation between philosophy and outcomes

broad curricula

Relevant courses and programs

conducive school atmosphere

Students entering the graduate programs

Policies on Admission, Comprehensiv e Examination, and Thesis/ Dissertation writing

effective counseling Institutional Quality

quality outputs (thesis and dissertation)

clearly defined administrative and academic responsibilities

stable resources

Graduates who are effective leaders and researchers

Manual on Graduate Admission Policies

extent of implementation

continuous intellectual and professional staff development

responsive board of trustees

adequate physical facilities and learning resources

effective on-going institutional self-study and planning procedure

GRADUATE PROGRAMS

C-I-P-P Model
Context

Input
Vision, mission, policies in research Research capability of faculty Faculty involvement in research Adequacy of research training and management personnel facilities funding and incentives

Process
Assessment on; Extent of attainment of vision, mission, policies in research level of research capability of faculty Extent of faculty involvement in research level of adequacy of research training and management personnel facilities funding and incentives

Product
Strategic Plan to enhance research orientation and activities at the Polytechnic University of the Philippines

OVPRD Structure University Research Manual

Revised University Research Manual

Feedback

Statement of the Problem

Statement of the objectives or purposes of the study: - main objective - specific objectives (specific problems)

Statement of the Problem This is where you state the general and specific problems for the research you choose to work on and hopes to achieve.

Hypothesis
A hypothesis is a temporary answer to the specific problems stated. A research hypothesis is usually based on the findings of previous studies. When it becomes necessary to test this hypothesis using some statistical tools, you need to state it in the null form.

Scope and Delimitation

Discuss inclusively the coverage of the study in terms of the population, time frame, variables.

Discuss if there are some specific items which should be excluded in the study.
(E.g. For instance, if the respondents are teachers, it may be specified that only full-time teachers are to be included in the study. Those who have part-time teaching loads will not be part of the study.

Significance of the Study

This part discusses the groups of persons, institutions or offices which will be benefited by the results of the study. How these groups are benefited by the results must likewise be discussed. The contribution of the study must be discussed in relation to the goals of education, institutions such as your school, the community and, to some extent, the country.

Definition of Terms The main variables of the study including terms, words or phrases which have special or unique meanings in the study are defined.

Definition may be taken from encyclopedia, dictionaries, books, other scholarly publications but the researcher must acknowledge his source. Definitions should be brief, clear and unequivocal.

Definition of Terms

Acronyms should always be spelled out fully especially if they are not commonly known or they are used for the first time.
Include the operational definitions which are specifically the meaning of the terms as they are used in the study.
E.g For instance, academic performance has a very broad meaning. But in your study its meaning may be confined to the evaluation by the teacher as reflected in the final grades in all subjects. These definitions will make clear to the reader the meaning of the terms as they are used in the text of the research paper.

REVIEW OF RELATED STUDIES AND LITERATURE

This contains the literature reviewed and the synthesis of the review. Foreign and local literature reviewed may be presented in separate sections.

Purposes of Review of Literature

1. Provides the refinements the research needs in terms of formulating the research questions and the framework for the investigation to be done.
2. To establish the facts which are already known from previous researches. 3. To have a relatively wide range of information about the topic.

4. To get a clearer vision of the additional research that needs to be done.

5. To enable the researcher to develop a theoretical/ conceptual framework for his/her research.

RESEARCH METHODOLOGY
1. 2. 3. 4. 5. 6.

Method of research to be used Population frame and sample size Description of respondents Instrumentation Data-gathering procedure Statistical treatment of data

I.

Method of Research

Methods of Research
Qualitative Methods

Quantitative Methods

Archival/Historical Case study Ethnomethodology

Descriptive method Experimental study Quasi-experiment

Qualitative Methods of Research

Archival/Historical - the systematic and objective location,
evaluation, and synthesis of evidence in order to establish facts and draw conclusions about past events; an act of reconstruction undertaken in a spirit of critical inquiry designed to achieve a fruitful representation of a previous age.

Case Study - typically observes the characteristics of an individual

unit; the purpose is to probe deeply and to analyze intensively the multifarious phenomena that constitute the life cycle of the unit with a view of establishing generalizations about the wider population to which the unit belongs.

Qualitative Methods of Research

Ethnomethodology - concerned with the world of everyday life;

how people make sense of their everyday world; directed at the mechanisms by which participants achieve and sustain interaction in a social encounter - the assumptions they make, the conventions they utilize, and the practices they adopt.

Quantitative Methods of Research

Descriptive method - involves the collection of data in order to
test hypotheses concerning the current status of the subject of the study.

Experimental method - investigates the cause-and-effect

relationship between variables; involves the manipulation of the independent variable/s and assigning subjects randomly into different treatment conditions.

Quasi-experimental method - resembles the true experiment

except the inability to assign people to conditions randomly.

Method of Research
Example 1: The method of research employed in this study was the descriptive method. According to Gay (1976, p. 69) descriptive method involves the collection of data in order to test hypothesis and answer specific questions concerning the current status of the subject of the study.

Method of Research
In relation to this research, the descriptive method was utilized because it is the most appropriate with regards the purpose of the researcher, that is, to gather data and answer specific problems pertaining to the changes in the physico-physiological, cognitive and social attributes of the elderly respondents, how they adjust to these changes, their personality profile, their stressors and abuses committed to them, and their stress-reducing mechanisms.

Method of Research
Example 2:
This study utilized the descriptive method of research. Descriptive research according to Martinez (1998) involves collection of data needed in testing the hypothesis or in answering questions concerning the current status of the subject of the study. It involves description, recording, analysis and interpretation of the conditions that are presently in existence. Martinez further said that descriptive method of research describes contemporary events and that the research questions are raised based on an appreciation of the present state of the system that is rooted in the past and may affect the future system.

Method of Research
Similar to Martinez, Calderon and Gonzales (1993) stressed that a descriptive research is a fact-finding study with adequate and accurate interpretation of findings. It describes with emphasis what actually exists, such as the current conditions, practices, solutions, and other phenomena. In this study, the researchers attempt to describe the current capability and effectiveness of the PNP in conducting intelligence operations against terrorism in Metro Manila. Based on the findings, a sound action plan will be designed to help in the improvement of the system and avert terrorism in the country today.

II. The Respondents of the Study

Population and Sampling Technique

Parameters characteristics of the population (numerical descriptive measures of a population). Statistics characteristics of a sample (numerical descriptive measures computed from a sample).

Population and Sampling Technique

Population a group of persons (institutions, events, or other subjects of study) that one wants to describe or about which one wants to generalize. Sample a group of persons (institutions, events, of other subjects of the study) that is used to represent and make a generalization about the population.

Population and Sampling Technique

Sampling Techniques:

Random or Probability Sampling Non-Random or Non-Probability Sampling

Population and Sampling Technique

Random Sampling Techniques:

Simple Random Sampling Systematic Random Sampling Stratified Random Sampling Proportional Stratified Random Sampling Multi-Staged Sampling

Population and Sampling Technique

Non-Random Sampling Techniques:

Purposive Sampling Convenient Sampling Quota Sampling Snowball Sampling

Determining the Sample Size

Determining Formula the Sample Size: N 1 + N (e) The Slavins

n
where: n N 1 e

=
= = = =

the sample size the population constant the margin of error

Determining the Sample Size

The Slavins Formula: Sample Application 777 1 + 264 777 (.05)2

=
=

Example: Two state universities were selected to compose the locale of the study the Polytechnic University of the Philippines (PUP) located in Sta. Mesa, Manila, and the Philippine Normal University (PNU) located at Ayala Ave. cor. Taft. Ave., Manila. The researcher utilized the purposive sampling technique in determining the respondents of the study. Purposive sampling technique according to Vogt (1999) is a technique where subjects are selected deliberately because the researcher thinks certain characteristics are typical or representative of the population. The technique was used since the respondents of the research were composed of only those who are preretirees and retirees teaching in the aforementioned state universities in Manila.

The researcher wrote to the human resource departments of the selected institutions and asked for the complete lists (population) of their fulltime faculty members (the pre-retirees) who are aged fifty-five to sixty (55 to 60) years old, and those who are sixty (60) years old and above who have already retired from the service but are still teaching on a part-time basis in their respective institutions. The data that were requested by and provided to the researcher are based on the master list of faculty during the 2nd Semester of SY 1998-1999. Please see Appendix A. Table 1 presents the population of the study.

Example 2: Population or universe refers to the total number of people that are to be studied while a sample refers to the subset that represents the population (Birion and De Jose, 1998). The population of the study is the 91 PCOs and the 680 PNCOs (a total of 777 police officers) from all the intelligence units/stations in Metro Manila. Using the following Slavins formula, the sample size was be computed.

Out of 777, two hundred sixty four (264) police officers participated in the study. Using the proportional stratified random sampling, the population of the study was subdivided or stratified according to the type of Officership or rank (whether they belong to the PCO or the PNCO groups). Proportional stratified random sampling according to Vogt (1999) is a type of probability sampling where participants are drawn from particular categories or strata of the population being studied. The technique works when individuals within the strata are highly similar to one another and different from individuals in other strata. When the characteristics of the population under study are somewhat heterogeneous, a method of stratification, that is, subdividing the population into homogeneous subgroups (strata) needs to be done first before sampling (Birion and De Jose, 1998).

Proportional stratified random sampling according to Vogt (1999) is a type of probability sampling where participants are drawn from particular categories or strata of the population being studied. The technique works when individuals within the strata are highly similar to one another and different from individuals in other strata. When the characteristics of the population under study are somewhat heterogeneous, a method of stratification, that is, subdividing the population into homogeneous subgroups (strata) needs to be done first before sampling (Birion and De Jose, 1998).

Example:

Table 3 Distributions of the Population and Sample of the Study N 97 680 n 33 231 p .34 .34

Group PCO PNCO Total

777

264

.34

In this study, the sample size which is equal to 264 is equivalent to .34 or 34% of the population. Hence the proportion of the sample size to the population size per stratum is equal .34. Table 3 presents the distributions of the population and sample of the study. After determining the sample size of each stratum, the simple random sampling was conducted to identify the actual respondents of the study. The conclusion derived from the data gathered through random sampling was generalized for the whole population.

Random sampling was done through the following procedures:

1. 2. 3.

4. 5.

6. 7. 8.

Getting a list of the total population or universe; Cutting pieces of paper to small sizes that can be rolled; Putting a number on each piece of paper corresponding to each number in the list of the total population or universe; Rolling each small piece of numbered paper; Putting the rolled pieces of paper in a bowl to give equal chances for everyone in the universe to be chosen as sample; Picking out one rolled paper at a time; Unrolling the paper and recording the number; and Repeating the process until the desired sample size is obtained.

III. The Instrument/s Used

Instrument/s Used
Three Options:
1. Look for previously validated questionnaire which has been used in the previous study. 2. Evolve an adaptation of a previously used or validated questionnaire. An adaptation results when a researcher modifies, replaces or deletes some items to make the set of items relevant to the study.

3. Construct a questionnaire based on the information needed for the study.

The Instrumentation

The process of developing and validating questionnaires through which variables of the study are measured.
1.

2.

Researcher-Made Instrument/Questionnaire Item-generation Face validation Content validation Item-analysis Reliability Psychological Tests Validity Reliability

Culture-fair!

Instrumentation
Characteristics of a good instrument
1. It deals with a significant topic, a topic the respondent will recognize as important enough to warrant spending his time in completing. The significance should be and carefully stated on the questionnaire, or in the letter that accompanies it. 2. It seeks only information which cannot be obtained from other sources such as school reports or census data. 3. It is as short as possible, only long enough to get the essential data. Long questionnaires frequently find their way into the wastebasket.

Instrumentation
4. It is attractive in appearance, neatly arranged, and clearly duplicated and printed. 5. Directions are clear and complete. Important terms are defined, each question deals with a single idea, all questions are worded as simply and as clearly as possible and the categories provide and opportunity for easy, accurate, and unambiguous responses. 6. The questions are objective, with no leading suggestions as to the responses desired. 7. Questions are presented in logical order, proceeding from general to more specific responses. 8. It is easy to tabulate and interpret.

Instrumentation
Preparing and administering the questionnaire:
1. Get all of the help that you can in planning and constructing your questionnaire . Study other questionnaires, and submit your items for criticism to other members of your class or your faculty, especially to those who have had experience in questionnaire construction.

2. Try out your questionnaire on a few friends and acquaintances.

Instrumentation
3. Choose respondents carefully. It is important that questionnaires be sent only to those who possess the desired informationthose who are likely to be sufficiently interested to respond conscientiously and objectively.

4. If schedules or questionnaires are planned for use in a public school, asking for the responses of teachers or pupils, it is essential that approval of the project be secured from the principal, who may then wish to secure approval from the superintendent of schools.
5. If the desired information is delicate or intimate in nature, consider the possibility of providing for anonymous response. The anonymous instrument is most likely to produce objective responses.

Instrumentation
6. Try to get the aid of sponsorship. Recipients are more likely to
answer if a person or organization, or institution of prestige has endorsed the project. 7. Be sure to include a courteous, carefully constructed cover letter to explain the purpose of the study. 8. Recipients are often slow to return completed questionnaires. To increase the number of return, a vigorous follow-up procedure may be necessary.

Instrumentation
Example:
The researcher gathered all the necessary reading materials in order to develop the rough draft of the questionnaires. These include books in developmental psychology, human development, general psychology, geriatrics, health care, etc., journals, magazines, and newspapers that featured articles related to the multidimensional attributes of aging. The researcher also interviewed individuals belonging to both the pre-retiree and the retiree age brackets to find out what were the changes they noticed in the different attributes, which were the concerns of this study. From these various sources, the researcher was able to develop the rough draft of the questionnaire.

Instrumentation
Rough drafts were submitted for expert judgment (content validation) by authorities in the field of psychology and psychometrics. The comments and suggestions for revisions were incorporated and the instruments were dryrun to twenty (20) senior faculty members at the Polytechnic University of the Philippines and the EARIST for face validation. Their ages range from 50 to 57 with a mean age of 53.6 or 54. The dry-run helped improved the questionnaires by eliminating the redundant questions and inconsistencies in the instrument. After careful analyses and necessary modifications were made, the researcher framed the final draft of the instrument.

Instrumentation
The faculty members who participated in the dry-run of the questionnaires were not included as respondents of the study. The following are the different instruments used in the study: the Multidimensional Attributes Questionnaire, which identified the changes in the physico-physiological, cognitive, and social changes experienced by the respondents, and how they adjust to them; the Personality Scale, which measured their personality attributes; the Stressor Survey, which gathered information regarding their stressors in family relationship, peer relationship, changes in financial status and work situations, stress due to injury, illness and bereavement, and the abuses committed against them; and the Stress-Reducing Mechanism Inventory, which determined the various stress-reducing mechanisms utilized by the respondents. Please see Appendix A.

(1)

(2) (3)

(4)

IV. The Data-Gathering Procedure

Data-Gathering Procedure
the process by which information is gathered.

Main types 1. Observation 2. Interviewing 3. Focus group discussion

4. Documentary analysis
5. Evaluation and examination 6. Instrument administration

Data-Gathering Procedure
Example: The researcher wrote official communication to the respective heads of the two educational institutions requesting permission to allow their faculty to participate in the study. After approval of the concerned authorities, and after the lists of faculty members who composed the population of the study has been released by the Human Resource Management Departments of both schools, the researcher identified the respondents of the study. The researcher administered the instruments and conducted interviews, when necessary, individually or in groups depending on the convenience on the part of the participants. Some of the questionnaires, however, were distributed through the help of the deans or heads of the departments where the faculty respondents belong. The questionnaires were coded to identify the respondents who participated in the study. This aided in the speedy retrieval of the floated questionnaires.

V. Statistical Treatment of Data

Statistical Treatment of Data

determining the appropriate statistical techniques to be used in analyzing the collected information/data

SOME TERMINOLOGIES

STATISTICAL PROCEDURES: GENERAL CLASSIFICATIONS

Descriptive Statistics procedures for summarizing, organizing, graphing, and in general, describing quantitative information. Inferential Statistics Statistics that allow us to draw conclusions or inferences from data. Usually this means coming to conclusions (such as estimates, generalizations, decisions, or predictions) about a population on the basis of data describing a sample.

FREQUENCY AND PERCENTAGE DISTRIBUTIONS

Frequency Distribution a tally of the number of times each score occurs in a group of scores. More formally, a way of presenting data that shows the number of cases having each of the attributes of a particular variable.
Percentage Frequency Distribution a frequency distribution that shows the percentage (not the number) of cases having each of the attributes of a particular variable.

MEASURES OF CENTRAL TENDENCY

Measures of Central Tendency - numerical descriptive measures that locate the center of a distribution of measurements. Arithmetic Mean the sum of scores or measurements divided by the number of measurements in a set. Median the middle score or measurement in a set of ranked scores or measurements; the point that divides a distribution into two equal halves. Mode the measurement that occurs most often in the set.

MEASURES OF VARIABILITY
Measures of Variability a statistic showing the amount of variation of spread in the scores for, or values of, a variable. When the dispersion is large, the scores or values are widely scattered; when it is small, they are tightly clustered. Range the difference between the largest and the smallest measurements. Variance the mean of the sum of the squared deviations from the mean score divided by number of scores Standard Deviation the square root of the variance.

MEASURES OF RELATIONSHIP
Measures of Relationship or Correlation a statistic showing the extent to which two or more variables are related to one another (co-related). Correlation Coefficient the statistic that indicates the extent or degree of relationship between or among variables. A positive value depicts a direct relationship while a negative value suggests an inverse relationship.

LEVELS OF MEASUREMENT
Levels of Measurement a term used to describe measurement scales in terms of how much information they convey about the differences among values the higher the level, the more information.
Nominal involves categorizing cases according to the presence or absence of some attribute. Ordinal categorizes and ranks cases on some variable, the differences between ranks need not be equal.

LEVELS OF MEASUREMENT

Interval categorizes, ranks and describes variables in such a way that the distance between any two adjacent units of measurement is the same.

Ratio possesses all the attributes of the lower levels of measurement and has a true zero point.

VARIABLES AND TYPES

Discrete Variable made up of distinct and separate units or categories; it can only take on a finite value. Dichotomous Variable a categorical variable that has been divided into two categories (e.g. pass and fail).

VARIABLES AND TYPES

Continuous Variable can be expressed by a large (often infinite) number of measures. Dichotomized Variable a continuous variable that has been divided into two categories (e.g. poor and not poor).

VARIABLES AND TYPES

Independent Variable the presumed cause in a study; a variable that can be used to predict or explain the values of another variable. Dependent Variable the presumed effect in a study; the variable whose values are predicted by the independent variable.

VARIABLES AND TYPES

Intervening Variable a variable that explains the relation, or provides the causal link between other variables (also known as mediating variables). Moderating Variable a variable that influences or moderates the relation between two other variables and thus produces an interaction effect.

TYPE OF TESTS BASED ON NUMBER OF SAMPLES

One-Sample Test procedures used to test statistics for only one sample. Two-Sample Test two samples or groups of subjects. K-Sample refers to statistical tools used for more than two samples.

TYPE OF TESTS BASED ON NUMBER OF SAMPLES

Independent Samples groups are unrelated to one another (mutually exclusive), that is, measurement of subjects in one group has nothing to do with measurements of subjects in the other group.

Correlated Samples two or more samples in which members of the separate samples share characteristics or relationship with one another.

HYPOTHESIS TESTING
Null Hypothesis (H0) The hypothesis that a researcher hopes to reject, thereby substantiating its opposite. Often, the hypothesis that two or more variables are not related or that two or more statistics are not the same. Alternative Hypothesis (Ha) The research hypothesis which the researcher hopes to indirectly substantiate by rejecting the null hypothesis.

HYPOTHESIS TESTING
Alpha Level ( ) -The chance a researcher is willing to take of committing an alpha error or Type 1 error, that is rejecting a null hypothesis that is true. The smaller the alpha level, the more significant the finding because the smaller the likelihood that the finding is due to chance (random error).

TESTING SIGNIFICANCE OF RELATIONSHIP

FOR NOMINAL MEASURES:

Chi-Square Test ( 2 ) Chi-square is not a measure of association even though it is used in the numerator of several measures of association (Leonard, 1976) for categorical variables. Contingency Coefficient ( C ) Short for Pearsons contingency coefficient which is a measure of association for categorical variables usually displayed in a contingency table (nominal).

TESTING SIGNIFICANCE OF RELATIONSHIP

Phi-Coefficient ( ) Measure of association between two variables used when both are categorical and one or both are dichotomous (nominal)

TESTING SIGNIFICANCE OF RELATIONSHIP

FOR ORDINAL MEASURES:

Spearmans Rank Order Correlation Coefficient ( ) A statistic that shows the degree of monotonic relationship between two variables that are arranged in rank order (ordinal).

TESTING SIGNIFICANCE OF RELATIONSHIP

Kendalls Tau Correlation ( ) a measure of association between two ordinal variables. Generally considered better than the Spearmans rho because of the way it deals with tied ranks. Kendalls Coefficient of Concordance ( W ) a nonparametrical statistical tests of the agreement among sets of rankings.

TESTING SIGNIFICANCE OF RELATIONSHIP

FOR INTERVAL/RATIO MEASURES:
Pearson Product Moment Correlation Coefficient ( r ) a measure of relationship between two variables that have been measured on interval or ration scales.

TESTING SIGNIFICANCE OF RELATIONSHIP

Assumption testing for correlational analysis (Pearson r):
1. Related pairs
- data must be collected for related pairs: that is, if you obtain a score on an X variable, there must also be a score on the Y variable form the same participant.

2. Scale of measurement - data should be interval or ratio in nature. 3. Normality - the scores for each variable should be normally distributed. 4. Linearity - the relationship between the two variables must be linear. 5. Homoscedasticity - the variability in the scores for one variable is roughly the same at all levels of the other variable. That is, it is concerned with how the score cluster uniformly about the regression line.

TESTING SIGNIFICANCE OF DIFFERENCE

FOR NOMINAL MEASURES Chi-Square Test ( 2 ) of Independence chisquare test applied to two nominal-variable measures.

TESTING SIGNIFICANCE OF DIFFERENCE

Assumption testing for chi-square:
1. Scale of measurement - the data should be at nominal level of
measurement.

2. Random sampling - observations should be randomly sampled from the population of all possible observations.
3. Independence of observations - each observation should be generated by a different subject and no subject is counted twice 4. Size of expected frequencies - when the number of cells than ten and particularly when the total sample size is the lowest expected frequency required for a chi-square five. However, the observed frequencies can be any including zero. is less small, test is value,

TESTING SIGNIFICANCE OF DIFFERENCE

FOR ORDINAL MEASURES:
Wilcoxon Signed-Rank Test a non-parametric test of statistical significance for use with two correlated samples, such as the same subjects on a before-andafter measure (ordinal). A nonparametric equivalence of t test for related sample.

TESTING SIGNIFICANCE OF DIFFERENCE Mann-Whitney U Test a test of statistical significance between two groups where data are measured on an ordinal scale. A nonparametric equivalence of t test for independent sample. Kruskal-Wallis Test ( H ) a nonparametric test of statistical significance used when testing more than two independent samples.

TESTING SIGNIFICANCE OF DIFFERENCE

t Test for Related Sample - a parametric test of
statistical significance for use with two correlated samples, such as the same subjects on a before-andafter measures, or with matched groups (interval/ratio).

t Test for Independent Sample - a test of statistical

significance between two groups where data are measured on an interval/ratio scale.

TESTING SIGNIFICANCE OF DIFFERENCE

Assumption testing for t test:
1. Scale of measurement - the data should be at interval or ratio level of measurement. 2. Random sampling - the scores should be randomly sampled from the population of interest.

3. Normality - the scores should be normally distributed in the population.

TESTING SIGNIFICANCE OF DIFFERENCE

One-Way ANOVA Test ( F ) a parametric test of statistical significance used when testing more than two independent samples. Used with only one independent variable or factor. Two-Way ANOVA Test ( F ) analysis of variance with two independent variables. It is used to study the effects of two independent variables separately (main effects) and together (Interaction effect).

TESTING SIGNIFICANCE OF DIFFERENCE

Assumption testing for ANOVA:
1. Random Selection - the same should be independently independently and randomly selected from the population of interest. 2. distribution.

Normality

each population data should have a normal

3. Homogeneity of variance - the different populations of scores should have homogenous variances.

TESTING PREDICTIONS
Regression Analyses used for predicting the values of variable/s from values of the other variable/s.
Simple Regression Analysis Multiple regression analysis

Type of Data

Question

Number of Groups or Variables

Scale Continuous Rank Dichotomous

Independence

Test Pearsons r Spearmans r Point biserial, etc.

2 variables
Relationships Multiple Variables

Independent 2 groups

Multiple Regression Two-sample t (Mann-Whitney)

Measurement

Dependent

Related-sample t (Wilcoxon)
One-way ANOVA (Kruskal-Wallis)

Differences Independent

Factorial ANOVA Data Multiple groups or variables Dependent Repeated-Measures (ANOVA) (Friedman) Chi-square Chi-square

Categorical

1 level of categorization 2 levels of categorization

REFERENCES
Best, J. W. (1981) Research in Education. 4th Ed. Englewood Cliffs, NJ: Prentice-Hall.
Birion, J. C. & De Jose, E. G. (1998) Glossary of Statistical Terms for Statisticians, Researchers, and Beginners. Quezon City: Rex Bookstore, Inc. Birion, J. C., De Jose, E. G., Dayrit, B., & Mapa, C. M. (2005) Thesis and Dissertation Writing Without Anguish. Valenzuela: Mutya Publishing House, Co. Bloom, M. (1986). The Experience of Research. New York, NY: Macmillan Publishing Co. Cruz, C. U (?) Principles and Methods of Research. A Simplified

Approach

REFERENCES
De Jose, E. G. (1999). Multi-Attributes of the elderly teaching in selected universities in Manila: a gerontology. Unpublished Thesis. Polytechnic University of the Philippines, Manila.
Diekhoff, G. (1992) Statistics for the Social and Behavioral Sciences: Univariate, Bivariate, Multivariate. Wm. C. Brown Publishers Leonard II, W. M. (1976) Basic Social Statistics. New York: West Publishing Co. Miranda, D. (2004) Extent of the intelligence capability and effectiveness of the Philippine National Police in combating terrorism in Metro Manila. Unpublished Thesis. Philippine Public Safety College. Vogt, W. P. (1999) Dictionary of Statistics and Methodology. A Nontechnical Guide for the Social Science. SAGE Publications, Inc.