SOCIAL RESEARCH METHODOLOGY Generating information for programme design and evaluation - 5 Day Course Presenter: Dr Susan Ziehl

Introduction Accreditation: Stellenbosch University Short Course Number 2788: A 12-credit Social Research Methodology SAQA level 8 short course adhering to SAQA and HEQC approval, quality control, individual assessment and accreditation requirements and with official certificates of competence to successful participants. Background: The goal of this five day social research methodology course is to create a better understanding of social research enquiry and to train participants in methodologies and techniques around generating information for programme design and evaluation. In this course we will firstly set the context by giving participants an introduction to social inquiry and the application of social research within their organisation and specifically the research unit. Secondly we will focus on the methodological paradigms underlying social research. Thirdly we will focus on quantitative & qualitative research design, methodologies and data collection techniques and elementary analysis of information. Participants will also be taught how to present their findings in a research report. Target participants: Public service (departments, entities, municipalities and non-governmental managers) tasked with research, programme evaluation and writing research reports.

Course objective, topics and outcomes: The overarching course objective is to create an awareness of the nature of academic research, establish a basic knowledge of different research designs and methodologies and to understand the logic & structure of a research report The following outcomes are to be achieved: 1.Understand the difference between quantitative and qualitative data, 2.Show a basic understanding of the different research designs and methodologies, 3.Be able to conduct basic research enquiries 4.Describe the process of data entry and coding, 5.Demonstrate a basic understanding of descriptive statistics, 6.Be able to analyse data and make basic comparisons, and 7.Be able to present data in a report.

Assessment and credits: After attending this course, participants should be able to fulfil the following assessment criteria through the indicated methods of assessment: Assessment criteria:

i

Each participant must, through the assessments, provide proof that he / she mastered the content in term of the satisfactory level of knowledge required and that he/she is able to apply this within the workplace setting. Assessment methods: Participants will be assessed by means of the following methods: 1. Group work that will be assessed in class and that will comprise 20% of the course mark and; 2. An assignment to be submitted 30 days after the conclusion of the course (80% of the course mark. Please note that the policy on submission of assignments for all Executive Development Programmes specifically states that assignments must be submitted four weeks after commencement of the course or the date agreed between the facilitator and the participants during the contact session. Assignments must be submitted as follows: 1. E-mail ONLY to Ms Tharia Uys: [email protected]; 2. Clearly marked with YOUR NAME, NAME OF THE COURSE and DATE OF THE COURSE. 3. Please contact Ms Uys at 021-918 4121 if you have submission related questions. Late submission of assignments is not accepted in the light of the accreditation requirements and Stellenbosch University policies. Where exceptional circumstances, however, prevent you from submitting on time, late submissions may only be accepted up to two weeks after submission date with a mark penalty of 5 % per week. Such a penalty can only be waived upon receipt of written authoritative proof of such exceptional circumstances. After two weeks, no further assignments will be accepted and a certificate of attendance will be issued. Assignment marks and final results for the course will be available as from 12 weeks after the commencement of the course. No marks, final results or certificates will be released until invoices have been settled in full. Please note that late settlement of invoices will delay the release of assignment marks, final results and the issuing of certificates. Certificates are printed by the Certificate Division of SU and are prepared and printed four weeks after all assignment marks and final results have been received from the SPL. Certificates will therefore, at the earliest, be dispatched 16 weeks after the course by SU. Certificates are sent by registered mail to the address noted on the registration form or to the coordinator who arranged the course with the SPL. Any certificates returned to the SU or the SPL due to non-collection will be retained and will only be sent or may be collected after suitable arrangements and payment have been made for collection from SU/SPL. Lost certificates: Please ensure the proper safekeeping of certificates. Kindly note that, in order to prevent fraud, SU cannot reissue lost certificates.

Course co-ordinator: Ms Adéle Burger Private Bag X1, Matieland, Stellenbosch E-mail: [email protected] Tel: 021 918 4412, Fax: 021 918 4123 ii

References: Babbie, E. & Mouton, J. (2001). The Practice of Social Research-South African Edition. Cape Town: Oxford University Press. Breynard, P.A. & Hanekom, S.X. (1997). Introduction to Research in Public Administration and Related Academic Disciplines. Pretoria: JL van Schaik. De Vos, A.S., Strydom, H., Fouché, C.B., & Delport, C.S.L. (2005) Research at grass roots, for the social sciences and human service professionals. 3rd edition. Pretoria: JL van Schaik. Gabers, J.G. (ed.) (1996). Effective research in the human sciences. Research management for researchers, supervisors and master’s and doctoral candidates. Pretoria: JL van Schaik. (Also available in Afrikaans) Henning, E., van Rensburg, W. & Smit, B. (2004). Finindg your way in qualitative research. Pretoria: JL van Schaik. Henning, E., Gravett, S. & van Rensburg, W. (2002). Finding your way in Academic Writing. Pretoria: JL van Schaik. Higson-Smith, C., Parle, J., Lange, L. & Tothill, A. (2000). Writing your research proposal, A workbook for first time and inexperienced researchers in the social sciences and humanities. Pretoria:

National

Research

Foundation

(NRF).

4

March

2004.

http://www.nrf.ac.za/yenza/research/proposals.htm Mouton, J. (1996). Understanding social research. Pretoria: JL van Schaik. (Chapter 26: Writing the research report) Neumann, W.L. (2000). Social research methods: qualitative and quantitative approaches. Boston: Allyn & Bacon, 4th edition. Vogt, WP., Johnson, R.B. (2011) Dictionary of Statistics & Methodology. A Nontechnical Guide for the Social Sciences. 4th Edition. Los Angeles: SAGE Publications. Welman, JC., Kruger, SJ. & Mitchell, B. (2005). Research Methodology. 3rd Edition. Cape Town: Oxford University Press. USEFULL WEBSITES List of current and completed research projects: http://www.nrf.ac.za/nexus Writing the report/assignment: http://www.sun.ac.za/library/eng/help/Infolit2002/report.htm iii

African Digital Library: http://www.AfricaEducation.org/adl/ SABINET online journals: http://www.sabinet.co.za YENZA, Start your research: http://www.nrf.ac.za/yenza/research

iv

Source: Palama (2010: 26-27) Research Methodologies for SMS in the Public Service, Section C

v

Daily Programme of Activities: Times 08:00 to 08:30

Day 1 Registration, tea, coffee

Day 2 Tea, coffee

08:30 to 10:30

Setting the Context: An introduction to social inquiry and the methodological paradigm of social research Clarifying the role and purpose of the Research Unit Identifying possible research projects

The Literature review: As a nonempirical design type How to go about organising a review Citations and referencing

Break Planning your research: Understanding how to develop a research problem statement: The origin of research ideas Transforming a research idea into a research problem Research question/hypothesis statement and research objectives

Break Quantitative Research Design: What is quantitative research? Experimental designs Quasi-experimental Non experimental design types

Lunch Understanding how to develop a research proposal How to go about choosing the correct research design: 4 dimensions to any research design Different design types Group work

11:00 to 13:00

13:45 to 15:45

15:45 to 16:15

Day 3

Day 4

Day 5

Tea, coffee Non experimental Research Design: Survey research Questionnaire construction methods for administering surveys Theoretical and Methodological Approaches to Quantitative Research Quantifying Data- Data entry, Coding and Data cleaning Break Qualitative research design: What is qualitative research? Ethnographic studies Case studies Life Histories

Qualitative data gathering methodologies: Focus group interviews Observations Personal documentation How do we go about sampling? Non-probability sampling

Introduction to computer aided data analysis Guidelines for reporting findings Ethical considerations when conducting research

Break Basic Quantitative data analysis: Data analysis tools: Elementary Analysis: Univariate Analysis, Subgroup Comparisons, Bivariate Analysis Social Statistics: Descriptive statistics, Level of Measurement, Inferential statistics

Lunch How do we go about sampling? Probability sampling Identifying our unit of analysis

Lunch Qualitative data gathering methodologies: Basic individual interviews Depth individual interviews

Group work

Group work

Lunch Basic Qualitative data analysis: Capturing and coding of qualitative data Content analysis Discourse analysis Grounded Theory Group work

Break The research report: Some basic considerations Organization of the report Guidelines for reporting findings How to present your findings and the role of the audience. Course closure Lunch

i

SETTING THE CONTEXT Here we are confronted with the following question. What is social research? Social research is the systematic analysis of research questions by using empirical methods (e.g. of asking, observing, analysing data etc.). Its aim is to make empirically grounded statements that can be generalized or to test such statements. Various approaches can be distinguished and also a number of fields of application (health, education, poverty etc.). Various aims can be pursued, ranging from an exact description of a phenomenon to its explanation or to the evaluation of an intervention or institution.

UNDERSTANDING THE THREE WORLDS FRAMEWORK WORLD 3: WORLD OF METASCIENCE Philosophy of science, research & research methodology

WORLD 2: WORLD OF SCIENCE o o o o

Theories, models Concepts & definitions Findings, data Instruments, scales, questionnaires

RESEARCH PROCESS PROBLEM-DESIGN-METHODOLGY-CONCLUSION

WORLD 1: EVERY DAY LIFE Social world: individual human beings; actions and events; organisations; institutions; interventions; collectives and social/cultural objects. Physical world: plants; animals; atoms etc

1

Everyday knowledge and practices vs. science and research Everyday knowledge

Science and research Relief from a pressure to act

Context of knowledge (production)

Pressure to act Analysing of problems is the Solving of problems is the priority: - routines are not put to question priority: - systematic analysis - reflection in case of practical problems

Ways of knowledge (production)

State of knowledge Relation of everyday knowledge and science

-

routines are put to question and broken down

Intuition

Use of scientific theories

Implicit development of theories

Explicit development of theories

Experience-driven development of theories

Methods-driven development of

Pragmatic testing of theories

Methods-based testing of theories

Check of solutions for problems Concrete, referring to the particular situations

Use of research methods

Everyday knowledge can be used as starting points for theory development and empirical research

theories

Abstract and generalizing Everyday knowledge is increasingly influenced by scientific theories and results of research

What characterises social research in dealing with everyday life issues 

Social research approaches issues in a systematic and above all empirical way.



For this purpose, you will develop research questions.



For answering these questions, you will collect and analyse data.



You will collect and analyse these data by using research methods.



The results are intended to be generalized beyond the examples (cases, samples etc.) that were studied.



From the systematic use of research methods and their results, you will derive descriptions or explanations of the phenomena you study.



For a systematic approach, time, freedom and (other) resources are necessary.

What Can You Achieve with Social Research? 

explore issues, fields and phenomena and provide first descriptions



discover new relations by collecting and analysing data



provide empirical data and analyses as a basis for developing theories



test existing theories and stocks of knowledge empirically



document the effects of interventions, treatments, programs etc. in an empirically based way 2



provide knowledge (i.e. data, analyses and results) as an empirically grounded basis for political, administrative and practical decision-making.

UNDERSTADING HOW TO DEFINE A RESEARCH PROBLEM Problem statement (theoretical/practical problem): Refers to some difficulty that the researcher experience in the context of either a theoretical or practical situation and to which he/she wants to obtain a solution Research question: Formulating the research problem as a question that you want to answer through the research process. Research aim: A general statements of intent or purpose of what the researcher plans to achieve, in its entirety Hypothesis: A tentative assumption or preliminary statement about the relationship between two or more things (variables) that need to be examined. An explanation of a research problem and the task of the researcher are to examine it. Variables: is a characteristic or an attribute of the study object (individuals, groups, organizations, human products or events, conditions to which they are exposed) It is important to be able to distinguish between an independent variable (x): which is a factor which the researcher selects and manipulate in order to determine its effect on the observed phenomena (the problem that is being investigated) and an dependant variable (y): which is a factor which the researcher observes and measure to determine what effect the independent variable has on it. Factors which appear disappear or vary as the researcher introduces, removes or varies the levels of the independent variable. Research objectives: Specific statements of purpose related to the concrete/measurable epistemic outcomes what the research wants to achieve. Research objectives are usually written as activity statements, using strong verbs such as “to describe” or “to explain” OR can be written as output statements, using noun words such as “to construct a model” or to build a “narrative”

3

IDENTIFYING OUR UNIT OF ANALYSIS Once we have decided on a design type we have to obtain our research participants (unit of analysis) according to your chosen sampling procedure. What could be our unit of analysis? “The ‘what’ or ‘whom’ you want to make deductions about”

Individuals

Institutions/ Organizations

Interventions

UNIT OF ANALYSIS Social artefacts/cultural objects

Social actions

Groups

Notes: __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ _______________________________________________

4

INTRODUCTION TO DIFFERENT RESEARCH DESIGNS & METHODOLOGIES Methodological Paradigm Here we are confronted with the following question What is a paradigm? A model or framework for observations and understanding, which shapes both what we see and how we understand it An accepted tradition and set of beliefs/values that guide research -Thomas KuhnBefore we can start with the research process we have to understand the Methodological Paradigms of Social Research Quantitative Research

Qualitative Research

Participatory Action Research

Quantificationassigning numbers to perceived quality of things  role of variables  control for sources of error

“Insider” perspective Describing and understanding “Stay close” to research subject

Closer relationship between researcher and research subject Participation of research subjects in research Political dimension

5

Differences between qualitative and quantitative research Quantitative research

Qualitative research

Theory

As a starting point to be tested

As an end point to be developed

Case selection

Oriented on (statistical) representativity, ideally random

Case selection

sampling

Purposive according to the theoretical fruitfulness of the case

sampling

Data collection

Standardized

Open

Analysis of data

Statistical

Interpretative

Generalization

In a statistical sense to the population

In a theoretical sense

Out of these paradigms we can identify certain research designs and operationalise them into specific research methodologies.

Research Design “Plan or blueprint” Focus on end product: What kind of study is being planned and what results are aimed at? Your point of departure is always your = Research problem Focus on logic of research: What kind of evidence is required to address the research question adequately? Four dimensions to a research design 1.

Empirical vs. Non-empirical

2.

Using primary vs. Existing/secondary data

3.

The nature of the data-numerical or textual

4.

The degree of control- highly structured vs. natural field setting

Research Design Types 

Participant observations



Case Studies



Participatory Action Research (PAR)



Surveys 6



Comparative, cross cultural & cross-national studies



Evaluation research: Implementation (process) evaluation



Evaluation research: experimental & quasi-experimental outcomes studies



Secondary Data Analysis (SDA)



Content Analysis

Quantitative Research Design Quantitative analysis- The numerical representation and manipulation of observations for the purpose of describing and explaining the phenomena that those observations reflect. (Babbie & Mouton, 2001) We can distinguish between three types of Quantitative research Designs: Experimental

Quasi-Experimental

Social Scientific Experiments  independent & dependent variables (defined)  pre & post testing  experimental & control groups  random assignment of NON subjects EXPERIMENTAL DESIGNS:

 independent and dependent variables (defined)  Natural environmentno control over intervention  No random assignment of subjects

Non Experimental

 independent and dependent variables (more than one)  No random assignment of subjects  No planned intervention

 survey design Examines the relationships that occur between different variables, without any planned intervention. Measurement at a single time (correlation-, criterion-group- , cross-section design)

7

Correlation- a single group of units of analysis is individually measured on two or more variables at about the same time; and the relationship (correlation) between these variables is analysed. Criterion-group- samples are drawn from the population representing the different levels of the independent variable. Then we investigate whether these groups differ in terms of the dependent variable. Cross-section- We take different criterion groups typically comprising of for instance different age groups which we examine in term of one or more variables at approximately the same time. Longitudinal design (panel, cohort and trend) Same group is examined at different time intervals (Time consuming & expensive) Panel- measurements are obtained at different points in time on one or more dependent variable(s) for the same sample that is more or less representative of the relevant population. Cohort- does not involve a representative sample from some population. Trend- measures different samples from the same population, rather than the same sample. Prediction studies (retrospective, prospective) (Time dimension, different variables are measured at different times-usually only two, subjects are divided into criterion groups) Retrospective- we know the criterion group of subject, but want to determine in terms of which possible predictor variable they have differed in the past. Prospective- we measure unit of analysis in terms of the supposed predictor variables and eventually their criterion group membership is checked.

8

Sampling Now that we have identified ‘what’ or ‘whom’ we are going to study, how would we go about sampling our research group? We get two types of sampling:

Probability

Non-probability

Simple random sampling

Cannot determine probability of an element being included -Useful in pilot studies -More economical and less complicated Accidental

Purposive

Systematic Snowball

Stratified Quota

Cluster Probability proportionate to size stratified Notes: __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ ______________________________________________________________ Before we explain some of these techniques we need to know the basic terminology to understand sampling:

9

Element- the unit about which information is gathered, often the dame as unit of analysis Population- theoretically specified aggregation of study elements Study population- aggregation of elements from which the sample is selected Sampling Unit- that elements or set of elements considered for selection in some stage of sampling (primary- secondary and final sampling units) Sampling frame- the actual list of sampling units from which the sample or some stage of the sample is selected. Parameter- summary description of a given variable in a population. (Mean income and age distribution). Statistic- summary description of a given variable in a sample. Sample statistics are used to make estimates of population parameters. Sampling error- estimated degree of error to be expected for a given sample design. Confidence levels and intervals- we express the accuracy of our sample statistics in terms of a level of confidence that the statistics fall within a specific interval from the parameter. Simple random sampling exercise

In a random sample of say, 50 homeowners from the population of homeowners in the Delft Housing Scheme, each homeowner irrespective of sex, age and race have an equal opportunity of being selected.

Step 1: We identified our population and drew a sampling frame. In this case our sampling frame would be a list of all the homeowners in the Delft Housing Scheme. Step 2: we now need to assign a single number to each element in the list, not skipping any number. A table of random numbers (Appendix A) is then used to select elements for the sample. The Delft Housing Scheme has 6 000 homeowners, which you now list form 0001- 6 000, we need to use 4 digit numbers to give everyone a chance to be selected.

10

Step 3: We now turn to Appendix A. you will see rows of five digits. We only want four digits. What we can do is to only look at the left four digits. We could decide on any plan, as long as we stick with it. Next we have to decide what pattern we are going to follow. Are we going to move down each column, across or diagonally? Say for this case we are going to move down the columns. Where do we start? Here we can apply various strategies, form closing our eyes and stick the pencil anywhere on the page to assigning numbers to columns and rows. Say we are going to start in column 2 row six (06907). The first number that will be included into our sample would then be no 690. We then move down the column and when we get to the bottom we move to the top of the next column, until we have selected 50 participants for our sample. Remember to watch out for numbers that lie outside the range of our sample frame, in this case 6 000. Ignore that number and move on the next one that falls within the range.

Systematic sampling exercise In systematic sampling every N/nth element is included in the sample. (N/n is a whole number) Suppose we need to select a systematic sample of 600 homeowners from a total population of 6 000 homeowners in the Delft housing scheme. 6 000 In this instance N/nth =

600

= 10.

To obtain a systematic sample, we first have to draw the first homeowner randomly from the first 10 on the list and after that we choose every 10th element of the remaining 6 000 homeowners. Stratified sampling exercise Suppose the population is composed out of clearly recognisable, non-overlapping subpopulations (strata) that differ from one another mutually in terms of the variable. The division may be in terms of one variable (sex – two strata male & female) or a combination of more than one variable (sex and age- strata: young adult males, young adult females, middle aged adult males, middle aged adult females etc.)

11

Let’s take our homeowner example. Let’s assume that there are four times more women homeowners in Delft than men. If in a single random sample of an equal numbers of men and women the proportion of homeownership is not separated we would seriously underestimate the instances of homeownership amongst women and overestimate amongst men.

Through simple random sampling we obtained a sample of five coloured males, three coloured females and two black male homeowners. There were no black females in our sample, though we know that they make up 6% the population. Black homeowners make up a total of 12% of the population, and if we had drawn a random sample according to such strata, Coloured vs. Black, there would have been a probability of 0, 5 (50%) of a black female being included into our sample of 10. Cluster sampling exercise When there is no list available (impossible or impractical) to draw our sample from we use this technique. Then we draw our sample from pre-existing heterogeneous groups called clusters, out of which we then draw random or stratified samples. Say for example we want to draw a sample of township dwellers in South Africa, we would in cluster sampling 

first draw a sample out of the nine provinces or regions



then draw a number of townships within each province or region



next draw a few street blocks within each township



and from these street blocks draw a random sample of individuals to be included into your sample

Now that we know how to draw our sample we need to learn how to draw up a questionnaire and the rules relating to survey research.

Survey Research: Principles of constructing questions What do we need to keep in mind when we construct a questionnaire?  Questions and statements  Open-ended and closed-ended questions

12

 Make items clear (vagueness leads to obscure answers) Avoid the following word in formulating questions: “Sort of”/ “In general’/ ‘Often”/ “Very”/

“On the whole”/ “About”/ “Why’/

“Basically”  Avoid double –barrelled questions (ambiguity)  Respondents must be competent to answer  Respondents must be willing to answer  Questions should be relevant  Short items are best  Avoid negative items  Avoid biased items and term and loaded questions  Avoid leading questions  Avoid hypothetical questions  Translation What information do we require? Factual questions- socio-demographic and personal information. Opinions or attitude questions probe for feelings, convictions, ideas and values related to a subject. Opinion- feelings and thoughts of respondents at a specific time and on a specific subject. Attitude- attempt to determine the integrated attitude-system underlying a particular opinion. Information questions- discover what respondents know about certain events. Questions on behavior- can be analyzed in terms of five dimensions: 

presence or absence



nature



frequency



degree of behavior on termination



degree of importance

What format can the questions we ask take? 13

Open ended questions (unstructured or free response questions) Respondent can formulate and express his/her response freely, since this type of question does not contain any fixed response categories. Advantages: Impose no restriction and researcher can determine exactly how the respondent has interpreted the question. Appropriate where researchers knowledge of subject is limited or where these is uncertainty about what response the question would elicit. Appropriate when a wide range of opinions is anticipated. Appropriate for pilot studies, with the view to compile answer categories for structured questions. Disadvantages: Time consuming, uneconomical and limits the number of questions. Success of response depends on competency of interviewer. Can be misleading; create the impression that the researcher is acquiring profound information about the complex motives and feelings of respondents. Does not necessarily produce more specific responses. Result in lower returns, considerable thinking involved. They are easy to ask, difficult to answer and even more difficult to analyze. Closed ended questions (structured questions) A question that contains specific, mutually exclusive response categories, from which the respondent selects one category that best, suits his/her response. Advantages: Easy to administer, coded beforehand. Economically and less time consuming to administer. Disadvantages: Can lead to loss of rapport and frustration if respondent feels that the response options do not accommodate their opinion.

14

Less subtle and respondents can discern the intentions behind questions, which leads to a subjective opinion.

Questionnaire construction General format Decide between the marking of responses in a box and the circling of pre-defined response codes. Try and keep the cost of production and reproduction as low as possible and make the questionnaire as attractive and respondent friendly as possible. The researcher often has to make a compromise between the two. Layout: You questionnaire should be spread out and not cluttered. Items should be adequately spaced to ensure that none are missed out. Normal typeface can be used, but use bold for instructions. Length of questions- should be as short as possible. Contingency questions Often a questionnaire contains questions that would be relevant to some participants and irrelevant to others. This sort of question arises when you want to ask a series of questions on a specific topic. The subsequent questions in a series of questions would be called contingency questions.

15

There are several formats for contingency questions. The clearest and most effective way of asking these types of questions is illustrated in the following example. 20

Have you ever smoked dagga? 1 Yes 2 No

If yes: About how many times have you smoked dagga? 1 once 2 2-5 times 3 6 or more times 4 11-20 times 5 more than 20 times

Here is an example of a more complex contingency question. 23

Have you ever attended a national soccer match where Bafana Bafana played? 1 Yes 2 No

If yes a Do you generally approve or disapprove of using vuvuzelas at big soccer matches 1 Approve 2 Disapprove 3 No opinion b Have you ever used a vuvuzela at a soccer match? 1 Yes 2 No

If yes: Why do you use it? ______________________

16

Another way of doing this is to give the respondent instructions to skip certain sections. 10

Have you ever voted in a national election? 1.

Yes

(please answer questions 11 to 20)

2. No (please skip questions 11 to 20 and go directly to question 21 on page 5) Notes: __________________________________________________________________________________ __________________________________________________________________________________ _____________________________________________________________ Matrix questions Often you would like to ask several questions that have the same set of answer categories. This is typically the case whenever the Likert response categories are used. This format uses space efficiently, is faster to complete, and the format increases the comparability of responses. This technique could also lead to certain response sets or patterns being formed by some respondents and some respondent may misread the statement because they wrongfully assume that they represent the same orientation. Example: 17 Beside each of the following statements presented below, please indicate whether you Strongly Agree (SA), Agree (A), Disagree (D), Strongly Disagree (SD) or are Undecided (U). (question asked to those who did not pay their municipal bill last month) SA

A

D

SD

U

1. It is important that people pay their municipal accounts

1

2

4

5

3

2. We stopped paying our municipal account when we saw that other people were not paying

1

2

4

5

3

3. We will start paying our municipal accounts once we know that other people are paying

1

2

4

5

3

17

again 4. At the moment we will be able to pay next month’s municipal fees

1

2

4

5

3

Notes: __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ _______________________________________________ Ordering items in a questionnaire The order of questions in a questionnaire can influence the refusal rate, as well as the quality of responses obtained, particularly when sensitive subjects are involved. Question order effect. Begin questionnaire with easy, non-threatening, interesting questions. This helps put the respondent at ease. Sensitive questions should be asked last, because if the respondent then does not complete the questionnaire little information would be lost. Group questions according to subject. Arrange questions logically, enabling the respondent to understand the relationships between them. When changing form one subject to another, introductory remarks should be made explaining what the following set of questions embraces. They should be so phrased and positioned that they contribute to the flow of the interview. Group items requiring similar responses together. Grouping can eliminate repetition of response categories for every question, particularly in the case of scale items. Vary questions with regard to length, response format and question type. A useful arrangement of questions is the funnel method. This option places the more general questions first and follows with more specific questions. This helps prevent early questions influencing responses to later questions on the same subject. Demographic questions can be seen as valuable ice-breakers at the start of your questionnaire. Response effects can be categorised as: ›

Consistency effects 18

›

Fatigue effects

›

Redundancy effects

›

Contrasting effects

We have now developed our questionnaire; pre tested, piloted and finally administered the questionnaire. How do we go about entering and coding the responses?

Coding responses The first step in the process would be to quantify your data. How do we go about quantifying our data? The first step in the process is coding your responses. For each response we need to develop a code category. This would mean that each data item (answer) would have a numeric code attached to it. It is important that you set up a codebook for your questionnaire. Different coding options:  Transfer sheet  Edge-coding  direct data entry  data entry by interviewer  optical scan sheets It is important that you clean your data set before you start entering your data into the computer. Example of a data sheet in Statistical Package for the Social Sciences (SPSS)

19

Example of coded data

20

Entering numeric values

21

22

23

24

Elementary Analysis & Social Statistics Before data analysis can take place it is important to understand what type of analysis will be done. If only one variable is involved in our study we call it univariate analysis, if two variables are involved it is called bivariate analysis and if more than two variables are involved it is called multivariate analysis. UNIVARIATE ANALYSIS: The analysis of a single variable, for purposes of description. Frequency distributions, averages, and measures of dispersion would be examples of univariate analysis. UNIVARIATE STATSITICS MODE: An average, the most frequently observed value or attribute. If a sample contains 1000 Protestants, 275 Catholics and 33 Jews. Protestants is the modal category. MEAN: An average, computed by summing the values several observations and dividing by the number of observations. If you now have a grade point average of 70% based on four courses and you get a 90% in your research methods course, your new (mean) average will be 72% (280 + 90/5). MEDIAN: An average, representing the value of the “middle” case in a rank-ordered set of observations. If the ages of five men are 16, 17, 20, 54 and 88, the median would be 20 (the mean would be 39). CONFIDENCE INTERVAL: The rage of values within which a population parameter is estimated to lie. A survey, for example, may show 40% of a sample favouring Candidate A. Although the best estimate of the support existing among all votes would also be 40%, we would not expect it to be exactly that. We might therefore, compute a confidence interval (35 to 45%) within which the actual percentage of the population probably lies. Note that we must specify a confidence level in connection with every confidence interval. CONFIDENCE LEVEL: The estimated probability that a population parameter lies within a given confidence interval. Thus we might be 95% confident that between 35 and 45% of all voters favour Candidate A STANDARD DEVIATION: An indication of how closely values are clustered around the mean. MINIMUM: The smallest value obtained for a variable. MAXIMUM: The largest value obtained for a variable. RANGE: The distance from the lowest to the highest values. Obtained by subtracting the highest form the lowest values.

25

SKEWNESS: An indication if the distribution of values are symmetric or not. A negatively skewed distribution has a mean that tends towards the higher values. KURTOSIS: Measures the “peakness’ of a distribution by looking at the flatness of the tail-ends of the distribution. If the kurtosis is high, then the distribution has a heavy-tailed distribution with a large number of scores that are very high and very low. If the kurtosis is low, than the distribution is more “peaked” and has relatively few values that are very high and very low. The following examples are taken from the MOONSTATS programme. It is a very basic data analysis package and is good to use for beginners. FREQUENCY TABLE: A table listing the values or scores and the frequencies with which they occur. GRAPHS- BAR, PIE, HISTOGRAM BIVARIATE ANALYSIS: The analysis of two variables simultaneously, for the purpose of determining the empirical relationship between them. The construction of a simple percentage table or the computation of a simple correlation coefficient would be examples of bivariate analysis. CROSSTABULATION: Shows how many cases with particular values on one variable have particular values on another variable. CORRELATION: Gives the relationship between variables. -

Pearson’s product moment for normally distributed data

-

Spearman’s rank order for not-normally distributed data or when the distribution is unknown

The p-value provides an indication of the significance of the relationship. The scatterplot gives a visual representation of the relationship between variables. Correlations are relatively sensitive to outliers. T-TEST: Is used to evaluate the differences in means between two groups. - Independent t-test: An example of independent groups are randomly selected experimental/treatment and control groups. - Dependant t-test: Dependant groups are based on related subject samples, for instance when a group of subjects are tested before and after an intervention to see if there is an improvement in their scores. The p-value reported with a t-test represents the probability of error involved in accepting the research hypothesis about the existence of a difference in the means.

26

MULTIVARIATE ANALYSIS: The analysis of the simultaneous relationship among several variables. Examining simultaneously the effects of age, gender and social class on religiosity would be an example of multivariate analysis. Before one can select any statistical test to be performed on a dataset it is important to determine the level of measurement. It is important to know that attributes of variables can be operationalised as mutually exclusive and exhaustive but also in terms of different levels of measurement. Four levels can be distinguished: NOMINAL Distinguishability (2 is different form 1)

Male

Female

1

2

ORDINAL

3

Chief Officer

Order or rank (2 has a higher rank than 1) Officer

2 Assistant Officer

1 INTERVAL Equal interval between successive higher

IQ

numbers (0…1…2…3…4…)

85

27

100

105

115

RATIO

INCOME

Absolute size( 0-1=-1=1-0=1)

0

R10

R20

R30

R40

The following are some statistics can be used for: nominal measurements

Mode Frequencies Percentages or proportions Chi-square (X2)

ordinal measurements

Median Frequencies Spearman’s rank-order correlation Chi-square (X2) Mann-Whitney u test Kruskal-Wallis test

interval & ratio (scale or non-categorical) Mean measurements

Variance & Standard deviation z-scores Pearson’s product moment correlation t-test and analysis of variance (F-test) Regression

DESCRIPTIVE STATISTICS This is a method for presenting quantitative data in a manageable form. It is a summary/description of a set of sample observations Descriptive statistics as used for mainly data reduction and measures of association 28

Descriptive statistical techniques include options like: MODE: The most frequently observed value or attribute. If a sample contains 1000 Protestants, 275 Catholics and 33 Jews. Protestants is the modal category. MEAN: An average, computed by summing the values of several observations and dividing it by the number of observations. If you now have a grade point average of 70% based on four courses and you get a 90% in your research methods course, your new (mean) average will be 72% ((280 + 90)/5 x 100). The mean is sensitive of outlier or extreme values. If there are extreme values, the median will be a better reflection of the most typical value. MEDIAN: Representing the value of the “middle” case in a rank-ordered set of observations. If the ages of five men are 16, 17, 20, 54 and 88, the median would be 20 (the mean would be 39). STATISTICAL SIGNIFICANCE: the non-probability that observed relationship between variables or the sample result can be ascribed to sampling error  Based on probability theory- need probability samples expressed ito probabilities (significant on e.g. 95% level- researcher choose level)  Level of significance of observed relationships = probability that relationship is only due to sampling error, e.g. p=0,05 – 5% probability that the relationship found between variables is due to chance CONFIDENCE INTERVAL: Can apply to sample statistics from a random or probability sample. The range of values within which a population parameter is estimated to lie.  Confidence level linked to probability/ significance level (e.g. p=0,05 confidence level 0,95 or 95%)  Confidence interval (CI): Boundaries within which the true population value will fall (use 95% CIs)  If the mean represents the true mean well, the CI of the mean should be small. A survey, for example, may show 40% of a sample favouring Candidate A. Although the best estimate of the support existing among all voters would also be 40%, we would not expect that all random samples of the same size from the same population will be exactly that. We might therefore, compute a 95% confidence interval indicating the range of values in which 95% of samples of a

29

particular size will produce results. Note that we must specify a confidence level in connection with every confidence interval. CONFIDENCE LEVEL: The estimated probability that a population parameter lies within a given confidence interval. Thus we might be 95% confident that between 35% and 45% of all voters favour Candidate A STANDARD DEVIATION: An indication of how closely values are clustered around the mean. MINIMUM: The smallest value obtained for a variable. MAXIMUM: The largest value obtained for a variable. RANGE: The distance from the lowest to the highest values. Obtained by subtracting the highest form the lowest values. SKEWNESS: An indication if the distribution of values is symmetric or not, i.e. whether there are outlier or extreme values. A negatively skewed distribution (or skewed to the left/ i.e. a few outlying low values) has a mean that will be lower than the median. The opposite for a positively skewed distribution. A normal distribution is not skewed. KURTOSIS: Measures the “peakness’ of a distribution by looking at the flatness of the tail-ends of the distribution. If the kurtosis is high, then the distribution has a heavy-tailed distribution with a large number of scores that are very high and very low. If the kurtosis is low, than the distribution is more “peaked” and has relatively few values that are very high and very low. FREQUENCY TABLE: A table listing the values or scores and the frequencies with which they occur. Frequency

Percent

HIV negative

27

90.0

HIV positive

3

10.0

30

100.0

Total

30

GRAPHS- BAR, PIE, HISTOGRAM, SCATTER PLOT Bar Chart (for nominal/ordinal data)

Pie Chart (for nominal/ordinal data)

31

Histogram (used for interval/ratio data

Notes __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ ________________________________________

32

CROSSTABULATION: Shows how many cases with particular values on one variable have particular values on another variable. I.e. one can study the relationship between two variables:

Constructing crosstabs:  Columns  Rows  Cell information ›

Frequencies or counts

›

Column %s – use when Independent Variable (IV) (grouping variable) is column variable (also called down percentaging)

›

Row % s – use when IV is row variable (across percentaging) 33

 Column & row totals are needed  Reading of a crosstab: If IV is column variable read across, if IV is row variable read down Table Error! No text of specified style in document..1: Backyard structure used for living purposes by area Backyard structure used for living purposes

Area Southern Cape larger Boland towns larger George, towns Mossel Paarl, Bay, City of Worcester, Knysna, Cape Town Wellington Plet

Yes

N Col %

No

N Col %

Total

N Col %

Eden Cape excluding Winelands George, excluding Mossel Paarl, Bay, Worcester, Knysna, Wellington Plet West Coast Overberg

Central Karoo

Total

514

77

163

32

14

120

52

4

976

39.9%

50.7%

41.6%

17.7%

18.9%

24.3%

26.8%

6.7%

34.4%

775

75

229

149

60

373

142

56

1859

60.1%

49.3%

58.4%

82.3%

81.1%

75.7%

73.2%

93.3%

65.6%

1289

152

392

181

74

493

194

60

2835

100.0%

100.0%

100.0%

100.0%

100.0%

100.0%

100.0%

100.0%

100.0%



Chi-Square = 121.644 / df = 7 / sig. = .000



Uncertainty Coefficient = .014

CORRELATION: Describes the linear relationship between variables. Estimates the extent to which change in one variable is associated with change in the other variable. A positive correlation explains a direct relationship, where an increase in the one variable will correspond with an increase in the other variable. Two variables that are inversely related would produce a negative correlation where an increase in the one variable would be associated with a decrease in the other variable. This is expressed in term of a coefficient r of -1,00 (perfect, inverse relationship) and a coefficient r of +1,00 indicates a perfect, linear relationship. A coefficient close to 0 indicated no relationship at all. -

Pearson’s product moment for interval/ratio variables

-

Spearman’s rank order for ordinal variables

The p-value provides an indication of the significance of the relationship (used for random or probability samples only). 34

The scatter plot gives a visual representation of the relationship between two scale (interval/ratio) variables. Correlations are relatively sensitive to outliers.

Scatter plots depicting different correlation coefficients (r) between two variables ranging from r = 1,00 (perfect correlation) to r = 0,00 (no correlation). (A relatively strong correlation is evident in the scatter plot to the left above and hardly any correlation in the plot on the right above.) Notes __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ ________________________________________

35

INFERENTIAL STATISTICS The body of statistical computations relevant to making inferences form findings based on random/probability sample observations to some larger population If we want to investigate a research hypothesis by means of inferential statistics methods we must transform our hypothesis into a statistical hypothesis. Such a statistical hypothesis consists of two statements known as: A null hypothesis (H0) (stating that there is no difference between groups in terms of the dependent variable) – this is the hypothesis that your test of statistical significance tests and you are not prepared to take a chance greater than p=0.05 (or 5%) (level of statistical significance) that the observed statistic or relationship is due to chance (sampling error). Alternative hypothesis (H1). H1 typically reflects the research hypothesis, e.g. stating that two variables are related or that there is a difference between groups (based on the so-called independent variable) based on their position on a dependent variable. In testing a model (or hypothesis) the following are important components: Between group variation (numerator)

reflects variance in dependent variable which is due to the independent variable

Within group variation (denominator)

reflects error variance caused by other variables systematic sources of variance: affects the scores of all subjects in a predictable manner unsystematic source of variance: affects subjects in an accidental unpredictable manner

Statistical significance The statistical significance of a relationship observed in a set of sample data is always expressed as a probability and refers to the H0. Significant at the 0.05 level (p≤0.05). This means that the probability of a relationship as strong as the observed one being attributable to sampling error alone is not more than 5 in 100. Three levels of significance are often used in research reports: 0.05, 0.01 and 0.001. This translate to a 5/100, 1/100 and 1/1000 change that this sample can be attribute to sampling error alone. Convention in the social sciences to use p = 0.05. 36

Tests of statistical significance Two groups: Parametric & Non-parametric Criteria for deciding parametric or non-parametric tests: 

Level of measurement of variables (categorical or scale data)? 

categorical data - use non-parametric tests



Ordinal & interval/ratio data - decide appropriateness of par/non-par test



Comparing proportions or means?

 Do variances differ? - use non-par. tests – debate, robust tests 

Related or unrelated/independent groups?



2 or more subgroups?

CHI-SQUARE: is a frequently used test for significance in social sciences. It is based on the null hypothesis. It involved measuring participants in terms of categories such as male- female. By using the chi-square test we can determine if (males versus females) have different preferences to for instance a particular product. T-TEST: Is used to evaluate the differences in means between two groups.  Independent t-test: An example of independent groups are randomly selected experimental/treatment and control groups.  Dependent t-test: Dependent groups are based on related subject samples, for instance when a group of subjects are tested before and after an intervention like in an experiment to see if there is an improvement in their scores. The p-value reported with a t-test represents the probability of error involved in accepting the research hypothesis about the existence of a difference in the means. Notes __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ ________________________________________

37

COMPUTER AIDED QUANTITATIVE DATA ANALYSIS Quantitative data analysis is done by using computer assisted programmes. Some useful programmes are: 

SPSS PASW (sophisticated analysis)

 Moonstats (elementary analysis)  Statistica A demonstration on the use of the SPSS PASW programme will be provided in class. Useful modules available with SPSS PASW

PASW Data Collection PASW Data Collection allows users to take control of the entire data collection lifecycle, from survey creation, through survey deployment and management to survey reporting. Users can create anything from simple web-based surveys to sophisticated data collection projects administered in multiple languages and multiple modes (on the web, on the phone, on paper, face-to-face, etc.). Users can enforce data quality through survey logic, real-time validation and calculations, sample management and quotas. PASW Data Collection uses an author once, deploy anywhere and in any language approach to survey development ensuring that you can easily leverage multiple means of reaching an audience without compromising survey quality. For web-based, phone-based and managed face-to-face surveys, PASW Data Collection includes robust survey management and administration tools online to ensure visibility into the survey process. Survey results can be made available at any stage of data collection through flexible reporting facilities. PASW Text Analysis for Surveys Text Analysis for Surveys enables survey researchers to code open-ended responses more quickly and reliably than manual-only methods. It does so by offering advanced linguistic technologies that eliminate the need to read verbatim responses word-for-word, and manual controls, which make verifying, refining, and overriding automated results easy. Users never lose control of the coding process or the quality of their results. Notes: __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ ___________________

38

Qualitative Research design Qualitative research can be distinguished for other types in terms of the following key features:  Research is conducted in the natural setting of the social actors  It focus on process rather than outcomes  Insider perspective (“emic” view) is being emphasized  Primary aim is in-depth description and understanding of events and actions  Its concern is to understand social action in terms of its specific context rather than attempting to generalize to some theoretical population  The research process is often inductive in its approach, resulting in the generating of new theories  The qualitative researcher is seen as the main instrument in the research process Inductive reasoning- is a form of reasoning where genuine supporting evidence can at best lead to highly probable conclusions (lends support to the conclusion the researcher makes). In

qualitative

research

we

have

three

main

research

designs

we

will

discuss.

Ethnographic studies

Case studies

Life histories

Originally= Cultural anthropology - direct observations of behaviour - uses interview techniques and participant observations- data gathering

Intensive investigation of a single unit (multiple variables) - individual cases - community studies - social groups - organisations/institutio ns - events, roles and relationships

Intensive observations of a subject’s life - concern for subject reality - focus on process and change - a perspective on totality - historical tool

Design principles of qualitative research:  Conceptualization  Contextual detail 39

 Multiple sources of data  Analytical strategies

Sampling There are two types of sampling option researchers take into consideration when working in the qualitative paradigm. 1.

Develop important criteria for the inclusion or exclusion of

before doing your fieldwork. This would number of 2.

respondents in your study

narrow your research to a focussed, much smaller

potential respondents.

Wait to be informed by your study as to which categories are less

in setting out the criteria for your sample.

or more important

= Theoretical sampling

Remember that sampling in studies where qualitative methods are used are almost always by means of purposive sampling

Qualitative data gathering methodologies Criteria for selecting respondents:  Thorough enculturation  Current involvement  Adequate time available Let’s discuss some basic data gathering techniques used during qualitative research. Basic individual interviews Open interview- allows the subject to speak for him/herself. Flexible The interviewer has a general plan of inquiry or an interview guide rather than a specific set of questions. This can be seen as a directed conversation with specific topics being pursued. Remember that asking questions may seem simple, but wording questions is a tricky business. Probes are a good way of getting answers in more depth without biasing later answers. We can distinguish seven steps in the interview process. (Kvale, 1996:88) STEP 1: THERMALIZING Clarifying the purpose of the interviews and the concepts to be explored 40

STEP 2: DESIGNING Laying out the process through which you’ll accomplish your purpose, including a consideration of the ethical dimension. STEP 3: INTERVIEWING Doing the actual interview. STEP 4: TRANSCRIBING Writing a text of the interview. STEP 5: ANALYSING Determining the meaning of gathered materials in relation to the purpose of the study. STEP 6: VERIFYING Checking the reliability and validity of the material. RELIABILITY The quality of measurement method that suggest that the same data would have been collected each time in repeated observations of the same phenomena (same result each time) VALIDITY A descriptive term used of a measure that accurately reflects the concept that is intended to measure (reflects real meaning of concept) STEP 7: REPORTING Telling others what you have learned. Depth individual interviews The researcher is not that interested in the content of the conversation, but rather the process by which the content of the conversation came into being. Coming to understand how another person’s frame of meaning is constructed. There is several ways in which you can do this: ›

Ask occasionally why questions.

›

Look out for apparent contradictions in what they are saying

This is an advanced technique and is not recommended for people who do not have interviewing experience. 41

Focus group interviews Normally comprises out of 8 to 12 participants. Always over-recruit by 20% to compensate for members not showing up. Takes about 2-2½ hours to conduct a focus group. Try and have 4-5 groups. The more heterogeneous your groups are you may want to stratify in terms of certain characteristics. You may want to record the conversations so you would need a reliable tape recorder and then you need someone to transcribe the tapes if you cannot do that yourself. The researcher acts as the group facilitator and can also take notes, but it is recommended that you have a separate person taking notes. The researcher would manage the process and each participant would get a chance to speak. The researcher would then end up with the individual responses of all the group members. Focus groups can be used to find information you would not otherwise be able to access. They create a space where people can collectively create meaning around a specific subject. Observations Simple observations

Researcher remains an outside observer

Participant observations

Researcher is a member of the group he/she is studying and also the researcher doing the study (this can become a dilemma) Overt vs. covert research Role of the observer? Relations to subject

What can we observe?  exterior physical signs  expressive movement  physical location  language behaviour 42

 time duration Personal documentation What type of documentation can be used?  Autobiography  diaries & diary interviews  Letters - Ceremonial/ information/ sentimental/ literary/ business What can we use personal documentation for?  Evaluation of theories  Subjective provision of data on institutional experiences  new exploration of apparently resolved issues  Establishing links between subjective and objective data  Sensitizing concepts  complementing objectivist accounts of social life

Qualitative data analysis There are various qualitative data analysis techniques. Here we will discuss some of these techniques to give you an indication as to how you could about analysing your data. Content analysis This method can be used when we want to analyse personal documentation, unstructured interviews and open-ended questions in a questionnaire. Content analysis can be seen as a four step process: STEP 1: Define the phenomena being analysed. STEP 2: Define the universe of appropriate media/ interviews and choose a sampling method. STEP 3: Give a description of the way in which the unit of analysis should be coded. This could mean that we count the number of times (frequency) a word or sentence appears in the text that are seen indicative of a certain construct. STEP 4: We must train the coders properly

43

Discourse Analysis This is a more complex process and fairly difficult to learn. The essence of discourse analysis lies in making explicit the unspoken and lived notions. Seven criteria for distinguishing discourse: 1. A discourse is realised in text 2. A discourse is about objects 3. A discourse contains subjects 4. A discourse is a coherent set of meanings 5. A discourse refers to other discourse 6. A discourse reflects its own way of speaking 7. A discourse is historically located Grounded theory The discovery of regularities as the identification of categories of elements and the establishment of their connections. Grounded theory begins with coding. Normally three types of coding can be identified: 1. open coding- the creation of certain categories pertaining to certain segments of text 2. axial coding- set of procedures where data are put back together in a new way after open coding. This is done by means of making connections between categories. The focus is on specifying a category. These categories can denote condition, strategy and consequence. 3. selective coding- process of selecting a core category, systematically relating it to other categories, validating those relationships and filling in categories that need further refinements and development. The key is to find the main storyline.

COMPUTER AIDED QUALITATIVE DATA ANALYSIS SOFTWARE (CAQDAS) There are a number of qualitative data analysis packages you could have a look at: Atlas.ti- http://www.atlasti.de Nudist 4- http://www.kerlins.net/bobbi/research/nudist Winmax- http://www.winmax.de

44

It is important to remember that when we are doing qualitative data analysis it starts at the collection process. The researcher will constantly reflect on the relationships and connections he or she sees while collecting the data. You will start the analysis by dividing the data into smaller more meaningful units. These units or segments as their called will be organised into a system based on the data, which implies that the analysis is inductive. You will use comparisons to build and refine categories, to define conceptual similarities and to discover patterns. Remember that categories are flexible and may be modified during the analysis. The analysis should always truly reflect the respondents’ perceptions. The results of an analysis can be seen as synthesis in the form of a descriptive picture, pattern or theme, or emerging or even substantive theory. Notes: __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________

REPORT WRITING GENERAL GUIDELINES: 

Choose the approach that best suits your audience, topic and purpose



State the purpose and focus clearly



Organise ideas and arguments logically



Make it easy to find information



Use visual organisation and formatting

Ways to organise your report: 

Use structured writing, including: Blocks

Side labels

45

Overview 

Integrated graphics

Use margin caption format: Left column: Headings and labels Right column: Text and graphics





Visual organisation: White/open spaces

Headings

Ordered and bulleted lists

Italics and bolded words

Use tools of emphasis: CAPITILISE underline bold significant words

The LANGUAGE you use in your report is also important and will depend on the audience. Ensure that you: 

use plain language,



Use scientific writing styles,



Use active rather that passive voice Example: Security checks all bags brought into the building. All bags brought into the building will be checked by security.



Use performance orientation – “what to do”



Clearly define actions and assign responsibility



Ensure correctness in grammar, spelling and punctuation

E.g. Punctuation:

A woman without her man is nothing.

SENTENCES AND PARAGRAPHS 

Keep to simple sentences of no more than 25 words



Use short, common, concrete words (avoid jargon, complex words) 46

-

Omit unnecessary words – repetitions, needless words, long-winded expressions facilitate

help/assist

concept

idea

indicate

say/show



Arguments: Put the important things up front.



Make the transition from one step of an argument to the next clearly – use headings

GUIDELINES FOR REPORTING ANALYSIS Data analysis should contain maximum detail without being cluttered. With quantitative data you need to ensure that the data can be recomputed. Ensure that you provide enough detail on the analysis in order for a secondary analyst to be able to replicate the analysis from the same data set. With qualitative analysis there should be sufficient detail to give the reader the sense of having been there with the researcher. You cannot only present information that supports your interpretation of the data. You must present the reader with alterative viewpoints. You must be able to demonstrate the reliability of the measuring instrument and the data. Tables, figures & graphs should be integrated into the text and not discussed in annexure or separate sections. 1. Describe the purpose of presenting the table 2. Present the table 3. Review and interpret it You need to draw explicit conclusions Adapted from Babbie & Mouton (2001:568-569) Notes: __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ 47

GENERIC STRUCTURE FOR A RESEARCH REPORT/ARTICLE Two main questions you want answers to: What was the research problem? How was the problem investigated? Two additional questions you need answered: What has been found? What are the implications & meaning of findings for the original problem posed? Main sections of the report  Introduction (the research problem)  Theoretical background & the research hypothesis/ question arising from it  Procedures and methods used to investigate the research problem  Results  Interpretation and discussion of results  Conclusions  Recommendations Title page Should contain the following information:  The report title which clearly states the purpose of the report. It should be concisely yet unambiguously reflect the exact topic of the project. It should include the important variables and the study population.  full details of the person(s) for whom the report was prepared  full details of the person(s) who prepared the report  the date of the presentation of the report Table of content This is a list of the headings and appendices of the report. Depending on the complexity and length of the report, you could list tables, figures and appendices separately. Make sure the correct page numbers are shown opposite the contents. Up-to-date word processing packages can generate a table of contents for you. Abbreviations and/or glossary of terms (not included in an article) If necessary, you should provide an alphabetical list of the abbreviations you have used in the report, especially if they may not be familiar to all readers of the report. If you have used a lot of technical terms, you should also provide a glossary (an alphabetical list of the terms, with brief explanations of their meanings). Abstract 48

The abstract is a brief summary usually no more than 200 words that appears after the tile and the name of the author. The following needs to be dealt with in the abstract:  What was the research problem?  How was the problem investigated?  What has been found?  What are the meaning & implications of the findings? Introduction The purpose of this section is to describe the objective of the particular research project and its importance. You will start with a wide, general description of the problem and move from there to formulating the specific problem statement (funnel approach). Body of the report The content of the body depends on the purpose of the report, and whether it is a report of primary or secondary research. A report of primary research (based on data you collected) would include: 

Literature review (what other people have written about this topic).



Problem statement & research question/hypotheses The literature review leads up to this section and is of the titled the statement of the problem. You can decide to state the research problem in terms of a question or in terms of some sort of logical relationship between the variables (hypotheses). 

Methods and Procedures (summarises what you did and why). Use the past tense. The following information is contained in this section  Unit of analysis - who or what of your study  Research design & methodology (data collection procedures)  Measuring instruments



Findings or results (describes what you discovered, observed, etc, in your empirical research). Use the past tense. Here you present the results with discussing or interpreting them, in some cases it might be more appropriate to combine them, depending on the design you choose. In this section you will make no reference to other publication. Presenting your findings in tables and graphs will make a good impression. Presenting results: Numbers Write out numbers smaller than 10 as words unless they represent a percentage, decimal number, sample size, ages and the numbers of tables, graphs and groups in which case figures

49

should be used. The same apply for fractions except for the common ones such as: half, quarter, one-third etc. If sentence starts with a number rather write it out as a word. Leave a space between a number and the measurement unit it refers to (example 8 m & 20 kg).In the case of percentages & degrees these is no spaces (example 20% & 5°). 

Discussion (discusses and explains your findings and relates them to previous research). Use the present tense to make generalisations. In this section the results presented in the previous section are interpreted in terms of the research problem introduced at the beginning of the report. You need to explain the meaning & implications of the results in light of the purpose for which the research was undertaken Discuss whether you agree or disagree with the findings of previous project, based on the literature review & discuss possible alternative interpretations of the findings. Try and point out the practical implications of the research. Remember that negative results do not mean that the study was useless.



Conclusion The purpose of the conclusion is to sum up the main points of the report. The conclusion should clearly relate to the objectives of your report. Don’t include new information here. Never draw conclusion you cannot support with facts from your study & discuss the limitations of your study. The report can be concluded with some suggestion as to further research that could be carried out on the topic. Recommendations, (if appropriate, can be made. These are suggestions for future action. They must be logically derived from the body of your report.

A report of secondary research (based on a literature review) would include: 

Information organised under appropriate topics with sub-headings. It is unlikely that your report will discuss each source separately. You need to synthesise material from different sources under topic headings.



Analysis/discussion of the sources you are reporting.

List of references This should be a list of sources referred to in your report Purpose- gives the reader the opportunity to consult these sources DO NOT put in any sources that you have not consulted and referred to directly in your report. Please use the Harvard or APA system of referencing Endnotes or footnotes should only be used to explanatory purposes. Appendices or Annexure (not included in an article) An appendix contains material which is too detailed, technical, or complex to include in the body of the report (for example, specifications, a questionnaire, or a long complex table of figures), but which

50

is referred to in the report. This is material that is important for the reader to fully understand your report & evaluate the report properly. Appendices are put at the very end of the report, after everything else. Each appendix should contain different material. Number each appendix clearly. CHECKLIST FOR EVALUATE YOUR OWN RESEARCH REPORT PROPOSED QUESTIONS The title: Is it a true reflection of the content of the article Is the title informative and appropriate Is the title descriptive, but not too long Does the title contain the important variables The statement of the research problem: Is the formulation clear and understandable Does it relate to something of theoretical and scientific value or practical significance Does it clearly set out the different points views and assumptions Does it clearly describe the theory or practical problem it logically flows from Does it culminate into a research hypotheses or research question which are formulated in terms of the relationship between the different variables The literature review: Is related to the aim and problem statement of your study Is the literature relevant to the topic Is comprehensive and uses a variety of essential information sources (is material suitable and persuasive? Does it adequately support the topic? Is an logical integrated summary of the information in the researchers own words Has the main disciplinary approach to research in this field been outlined

51

The research design: Did you choose the most appropriated design to investigate the research question Is it clear in terms of the sampling procedures, selection of respondents and measuring instruments used Did you check for internal & external validity Measuring instruments: (if any) Are the content are described briefly Did you describe the administering and data gathering procedures Did you discuss the reliability and validity of the instrument The analytical/statistical techniques: Are the techniques you selected appropriate for the given problem The data that are presented: Are credible evidence used? Is the supporting material suitable and persuasive? Does it adequately support the topic? Are visual materials – tables, figures, charts, maps, and the like – introduced before they appear in the text? The writing style: Is the organization logical? If headings and subheadings are used, do they consistently follow an accepted format? Are sentences varied in length and structure? Are tone, voice and diction consistent and appropriate? Are transitions smooth from sentence to sentence, paragraph to paragraph, section to section? Technical requirements of the article: Is the article between Are long quotations set off from the text? Is proper credit given to sources throughout? Does the paper consistently adhere to the documentation style specifications in format and documentation, both within the text and at the end of the text? 52

Did you include a list of references

LIST OF POSSIBLE SOURCES The following list of references can be used as a starting point for researching and writing your report/article. CORPORATE GOVERNANCE Blair, M M. 1995. Ownership and Control: Rethinking Corporate Governance for the twenty-first century. Brookings Institute: Washington DC. Gordon, J N. & Roe, M J. 2004. Convergence and persistence in corporate governance. Cambridge University Press: United Kingdom. Du Plessis, J. J., McConvill, J. & Bagaric, M. 2005. Principles of contemporary corporate governance. Cambridge University Press: New York. Khoza, R.J. & Adam, M. 2005. The power of governance: enhancing the performance of state owned enterprises. Pan Macmillan: Johannesburg. Munshi, S & Abraham, B.P. (eds.) 2004. Good governance, democratic societies and globalisation. Sage Publications: London. Farazmand, A. (ed.) 2004. Sound Governance: Policy and Administrative Innovations. Praeger: London. Bardach, E. 2005. A practical guide for policy analysis: The eightfold path to more effective problem solving. CQ Press: Washington DC. Weimer, D.L. & Vining, A.R. 2005. Policy Analysis: concepts and practice. Prentice Hall: New Jersey. Charan, R. 2005. Boards that deliver: advancing corporate governance from compliance to competitive advantage. Jossey-Bass: San Francisco, CA. Chew, D H & Gillan, S L. (eds.) 2005. Corporate governance at the crossroads: a book of readings. McGraw-Hill: Boston. Monks, R A G. & Minow, N. 2004. Corporate governance. Blackwell Publishers: Malden, Massachuchetts. Wixley, T. & Everingham, G. 2005. Corporate governance. Siber Ink: Cape Town. King Committee on Corporate Governance: 2002. King report on corporate governance for South Africa 2002. Institute of Directors: South Africa. FINANCIAL MANAGEMENT IN PUBLIC SECTOR Gildenhuys, J. S. H. 1997. Introduction to the management of public finance: a South African perspective. Van Schaik: Pretoria.

53

Schwella, E. (et al). 1996. Public resource management. Juta: Cape Town. Reed, B.J. & Swain, J. W. 1997. Public finance administration. Sage: Thousand Oaks. Visser, C.B. & Erasmus, P.W. 2000. The management of public finance: a practical guide. Oxford University Press: Cape Town. Connolly, S. 1999. Economics of the public sector. Prentice Hall: London. Gruber, J. 2004. Public finance and public policy. Worth Publishers: New York. Rosen, H.S. 2002. Public finance. McGraw-Hill: Boston, Massachusetts. Steiss, A.W., Nwagwu, E.O.C. 2001. Financial planning and management in public organizations. HUMAN RESOURCE AND PERFORMANCE MANAGEMENT Daley, D.M. 2002. Strategic human resource management: people and performance management in the public sector. Prentice Hall: New Jersey. Armstrong, M. & Baron, A. 2005. Managing performance: performance management in action. Chartered Institute of Personnel and Development: London. Kenny, G. 2005. Strategic planning and performance management: develop and measure winning strategy. Elsevier/Butterworth-Heinemann: London. Williams, R.S. 2002. Managing employee performance: design and implementation in organizations. Thomson Learning: London. INFORMATION AND COMMUNICATION TECHNOLOGY Snellen, T.M., Van de Donk, W.B.H.J. 1998. Public administration in an information age: a handbook. IOS Press: Amsterdam. Kahn, R. & Blair, B.T. 2004. Information nation: seven keys to information management compliance. AIIM International: Silver Spring. Cairncross, F. 2003. The company of the future: meeting the management challenges of the communications revolution. Profile: London. Axford, B. & Huggins, R. 2001. New media and politics. Sage: London. LEADERSHIP AND CHANGE MANAGEMENT IN PUBLIC SECTOR Randall, J. 2004. Managing change/changing managers. Routledge: London. Donahue, J.D. & Nye, J.S. (eds.). 2003. For the people: can we fix public service? Institution Press: Washington D.C. Storey, J. (ed.) 2004. Leadership in organizations: current issues and key trends. Routledge: London.

54

Avery, G.C. 2005. Leadership for sustainable futures: achieving success in a competitive world. Edward Elgar: United Kingdom. Milner, E.M. & Joyce, P. 2005. Lessons in leadership: meeting the challenges of public services management. Routledge: London. Cooper, T. L. & Wright, N.D. (eds.). 1992. Exemplary public administrators: character and leadership in government. Jossey-Bass: San Francisco, California. Terry, R.W. 1993. Authentic leadership: courage in action. Jossey-Bass: San Francisco, California. Joyce, P. 2000. Strategy in the public sector: a guide to effective change management. Wiley: Chichester: West Sussex. PROJECT MANAGEMENT Burke, R. 1999. Project management: planning & control techniques. Wiley: Chichester: West Sussex. Mantel, S. J. (et al). 2005. Core concepts of project management in practice. John Wiley & Sons: New Jersey. Harrison, F.L. 2004. Advanced project management: a structure approach. Gower: Alderhot, Hants, England. Rakos, J.J. 2005. The practical guide to project management documentation. John Wiley: New Jersey. Frame, J. D. 2003. Managing projects in organizations: how to make the best use of time, techniques, and people. Jossey-Bass: San Francisco. Knipe, A. (et al). 2002. Project management for success. Heinemann: Sandown. POLICY ANALYSIS and PUBLIC POLICY MANAGEMENT Cloete, F., Wissink, H. 2000. Improving public policy. Pretoria: Van Schaik. Radin, B. A. 2000. Beyond Machiavelli: Policy analysis comes of age. Georgetown University Press: Washington DC. Geva-May, I. (ed.). 2005. Thinking like a Policy Analyst: Policy Analysis as an clinical profession. Palgrave Macmillan: New York. Weimer, D.L. & Vining, A. R. 2005. Policy analysis: concepts and practice. Pearson Prentice Hall: New Jersey. Birkland, T.A. 2005. An introduction to the policy process: theories, concepts, and models of public policy making. M.E. Sharpe: New York. Van Niekerk, D., Van der Waldt, G., & Jonker, A. 2001. Governance, politics and policy in South Africa. Oxford University Press: Oxford. Rose, R. 2005. Learning from comparative public policy: a practical guide. Routledge: Abingdon, Oxon. 55

Weimer, D.L. 2005. Policy analysis: concepts and practice. Pearson Prentice Hall: New Jersey. Mintrom, M. 2003. People skills for policy analysts. Georgetown University Press: Washington D.C. Turner, M.M. & Hulme, D. 2003. Governance, management and development: making the state work. Palgrave Macmillan: Basingstroke. Meiring, M.H. 2001. Fundamental public administration: a perspective on development. School of Public Administration: Port Elizabeth. Brynard, P. A., Botes, P.S. & Fourie, D.J. 1997. Critical issues in public management and administration in South Africa. Kagiso Tertiary: Pretoria. Hanekom, S.X. 1996. Public policy: framework and instrument for action. International Thomsom Publishing: Johannesburg. SUSTAINABLE DEVELOPMENT Lafferty, W. M. & Meadowcroft, J. 2003. Implementing Sustainable Development: Strategies and initiatives in high consumption societies. Oxford University Press: New York Dresner, S. 2002. The Principles of Sustainability. Earthscan: London. Agyeman, J, Bullard, R., & Evans, B. 2003. Just sustainabilities: development in an unequal world. Earthscan: London. Kirbby, J. & O’Keefe, P & Timberlake, L. 1995. The Earthscan reader in sustainable development. Earthscan: London. Carley, M. & Christie, I. 1992. Managing sustainable development. Earthscan: London. Dalal-Clayton, B. & Bass, S. 2002. Sustainable development strategies : A resource book. Earthscan: London. Neefjes, K. 2000. Environments and Livelihoods: Strategies for sustainability. Oxfam: United Kingdom.

56

GLOSSARY Action Research: A type of applied research designed to find the most effective way to bring about a desired social change or solve a practical problem; it is usually conducted in collaboration with the subjects of the research. Basic Research: Research undertaken with the primary goal of advancing knowledge and theoretical understanding rather than solving practical problems. (Sometimes referred to as pure research). Census: A complete count of an entire population or collection of data from all members of a population. Code: Rules specifying how data are to be represented Codebook: A list of variables and how they have been coded so that they can be read and manipulated by a computer Coding: “translating” data form one language format into another, often to make it possible for a computer to operate on the data this coded. In qualitative research coding is often done after the data are collected, the codes, called inductive codes, are generated by the researcher during data analysis. These codes focus on single categories rather than variables. Empirical: said of data based on observations or experience and of findings that can be verified by observations or experience. Inductive reasoning: is a form of reasoning where genuine supporting evidence (such as empirical data) can at best lead to highly probable conclusions. Two form of induction are typically distinguished: inductive generalizations (when we draw generalizations from samples to a greater target population for example) and retroduction (when we generate explanations or hypothesis to account for observable patterns in data). Research design: is the plan according to which we obtain research participants (subjects) and collect information from them. In it we describe what we are going to do with the participants, with a view to reaching conclusions about the research problem. Research methodology: Are the methods, techniques and procedures that are employed in the process of implementing the research design or research plan, as well as the underlying principles and assumptions that underlie their use. Sample: a group of subjects or cases selected from a larger group in the hope that studying this smaller group (the sample) will reveal important information about the larger group (the population). Unit of analysis: is the “what: or “whom” being studied. In social sciences research the most typical unit of analysis is individual people.



Mostly out of Babbie & Mouton (2001) with slight adaptations & Vogt & Johnson (2011).

57

Variables: is a characteristic or an attribute of the study object (unit of analysis). The variable gender is made up of the attributes male and female. Independent variable: Factor which the researcher selects and manipulate in order to determine its effect on the observed phenomena (the problem that is being investigated) Dependant variable: The factor which the researcher observes and measure to determine what effect the independent variable has on it. Factors which appear disappear or vary as the researcher introduces, removes or varies the levels of the independent variable. Validity: A descriptive term used of a measure that accurately reflects the concept that is intended to measure (reflects real meaning of concept) Reliability: The quality of measurement method that suggest that the same data would have been collected each time in repeated observations of the same phenomena (same result each time) Questionnaire: A document containing questions and other types of items designed to solicit information appropriate to analysis.

58

WORKSHEETS WORKSHEET 1-2: DAY 1 WORKSHEET 3-5: DAY 2 WORKHEET 6-8: DAY 3 WORKHEET 9-11: DAY 4 WORKSHEET 12: POST COURSE ASSIGNMENT

DAY 1 WORKSHEET 1 REFINING YOUR RESEARCH TOPIC 1.

Write down your broad topic _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________

2.

Write down two or three key words _____________________________________________________________________ _____________________________________________________________________

3.

Adapt this broad topic to link to something specific you are interested in or something specific like a policy aspect operational difficulty or an innovation etc. Write down another two or three keywords capturing this specific focus. _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________

4.

Take the keyword you have written down and write them down in the blocks provided. Identify the links between the key concepts and connect them with pencil to capture a logical thread.

5.

Linking the keywords together in a logical order formulate a provisional topic. _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________

6.

What problem/tension/gap is your topic addressing? Formulate your provisional topic into a problem question/statement. _____________________________________________________________________ _____________________________________________________________________ ____________

7.

Identify about four purposive aims your research will try and pursue linked to the above question. (For example, start with: explore, investigate, identify, determine, interpret, distinguish, analyse, measure, compare, contrast, understand, access, etc.) 1

_____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ 8.

Now write down the provisional title of your research project _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ [Adapted form: Nelleke Bak (2004) Completing your thesis, a practical guide. Van Schaik Publisher: Pretoria.]

WORKSHEET 2 RESEARCH DESIGN Read the following problem statement and explain whether this study would be a crosssectional or a longitudinal design? A Machine operator thinks that fumes emitted in the workshop play a role in the low efficiency of the operators. The operator would like to prove this to the supervisor through a scientific research study. Answer: ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________

2

DAY 2 WORKSHEET 3 REFERENCING Please look at the following extract from a book by John Gall (1975). Systematics: How systems work and especially how they fail. New York: Quadrangle, and then identify which of the citations provided are acceptable or unacceptable in term of academic writing practices? The original work: Laws of growth Systems are like babies: once you get one, you have it. They don’t go away. On the contrary they display the most remarkable persistence. They not only persist: they grow. And as they grow, they encroach. The growth potential of systems was explored in a tentative, preliminary way by Parkinson, who concluded that administrative systems maintain an average growth of 5-6 per cent per annum regardless of the work to be done. Parkinson was right so far as he goes, and we must give him full honours for initiating the serious study of this important topic. But what Parkinson failed to perceive, we now enunciate- the general systems analogue of Parkinson’s Law. The System Itself Tend to Grow At 5 To 6 Per cent Per Annum Again this Law is but the preliminary to the most general formulation, the Big-Bang Theorem of Systems Cosmology. Systems Tend to Expand to Fill The Known Universe Citations 1: In this paper, I want to look at some of the characteristics of the social system we create in our organisations. First, systems are like babies: once you get one, you have it. They don’t go away. On the contrary they display the most remarkable persistence. They not only persist; they grow. Citation 2: John Gall warns that systems are like babies. Create a system and it sticks around. Worse yet, Gall notes, systems keep growing larger and larger.1 Citation 3: I want to look at some of the characteristics of the social systems we create in our organisations. First systems are a lot like children: once you get them, it’s yours. They don’t go away; they persist. They not only persist, in fact: they grow. Citation 4: It has also been suggested that systems have a natural tendency to persist, even grow and encroach (Gall, 1975:12).

John Gall (1975). Systematics: How systems work and especially how they fail. New York: Quadrangle, 12-14 1

3

WORKSHEET 4 UNIT OF ANALYSIS Look at the following research topics and identify what the unit of analysis are in each? Topic 1: Conventional wisdom has it that women and children in women-led households are more economically disadvantaged and more vulnerable to impoverishment, than their counterparts in other households. Research in the township of Bathurst in the Eastern Cape challenges this I a number of respects. Women here took the view that they had a better chance of domestic stability and economic security in the long term if they set up their own homes and entered into co-operative housekeeping and childrearing partnerships with other women. (Jones, 1999) Answer: ___________________________________________________________________________ ___________________________________________________________________________ Topic 2: NGO’s have different traditions and social contexts from which they have emerged and within they operate. On the one hand, there are those that had a closer relationship with the government in the apartheid era. On the other hand, a substantial NGO sector existed which had its roots in the anti-apartheid movement. This study addressed the impact of the social changes in South Africa on NGOs and their ability to meet the challenges of service delivery (Patel, 1998). Answer: ___________________________________________________________________________ ___________________________________________________________________________

WORKSHEET 5 SAMPLING READ THE CASE STUDY and explain what type of sampling is used in this research and whether it is the most appropriate strategy? ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________

1

DAY 3 WORKSHEET 6 QUESTIONAIRE CONSTRUCTION READ THE CASE STUDY and answer the following questions. What is the advantageous and disadvantageous of using the type of questionnaire that was used in this case study? ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ Develop graphic examples (items) of any two attitude questions that you think was used in the questionnaire in this article. ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ Develop one graphical evaluation scale (item/question) that could have been used in this case study. ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________

WORKSHEET 7 CODING Draw up a codebook – with variables and code assignments- for the following questions in a questionnaire and critique the questions: a.

Have you ever attended any training course? Yes, if yes go to b No 2

b.

What training did you attend? Short course Formal academic training Workshops Other: ______________________________________

c.

What do you feel is the most important challenges you as a manager are faced within your organisation?

d.

In the space provided below, please indicate the three implementation challenges/problems that most concern you by putting a 1 beside the one that most concern you, a 2 beside your second choice and a 3 beside your third choice.

_______Insufficient funding _______Lack of capacity _______Lack of commitment _______Political interference _______Institutionalisation _______Policy design flaws Code Book: Question a: ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ Question b: ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ Question c: ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ Question d: ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________

3

WORKSHEET 8 QUALITATIVE DESIGN AND METHODOLOGY READ THE CASE STUDY and describe what type of qualitative design the researcher used and what additional data collection strategies could have been used? ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________

DAY 4 WORKSHEET 9 LEVELS OF MEASUREMENT Name two variables that would be naturally considered for nominal scales. Set up mutually exclusive and exhaustive categories for each of the variables mentioned above. ___________________________________________________________________________ ___________________________________________________________________________ Develop an ordinal scale, with instructions, for consumer preference for three types of training programmes. ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ Name three variables that could be measured on an interval scale ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ Read the following questions on achievement motivation and develop an interval scale to measure these questions. You may re-phrase the questions, if you wish, without changing their meaning. QUESTIONS To what extend would you prefer a job that is difficult but challenging to one that is easy and routine? 4

___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ To what extend would it frustrate you if people did not give feedback on how you were progressing in your job? ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________

WORKSHEET 10 QUALITATIVE ANALYSIS The following are responses on an open-ended question to people who recently moved into a RDP house and represent a type of qualitative data. Describe how you would decide on how to code the responses into more manageable categories (Ignore those with $ signs): What was the best thing of moving here?

Valid

Frequency 218 80

Percent 56.8 20.8

Valid Percent 56.8 20.8

To have my own property & not to live in a shack\hokkie

15

3.9

3.9

Closer to work opportunities To have my own property & safer environment Safer environment

12 7

3.1 1.8

3.1 1.8

5

1.3

1.3

4

1.0

1.0

4

1.0

1.0

4 3

1.0 .8

1.0 .8

2 2 2 2 2 2 2

.5 .5 .5 .5 .5 .5 .5

.5 .5 .5 .5 .5 .5 .5

1 1 1

.3 .3 .3

.3 .3 .3

1 1 1 1

.3 .3 .3 .3

.3 .3 .3 .3

1 1

.3 .3

.3 .3

To have my own property Not to live in a shack\hokkie [to have a formal house]

To have my own property, to have own electricity connection & to have own toilet To have my own water tap, to have my own electricity connection & to have my own toilet Closer to school User missing\Not applicable Other, specify Nothing was good $2$9 To have own water tap To have own electricity connection To have own toilet $1$2$3$4$5$9 $1$2$5$3 To have my own property, not to live in a shack\hokkie & safer environment $1$3$4$5 $1$3$5 $1$3$5$9 To have my own property & to have own toilet $1$8 $10$13

5

$2$3

1

.3

.3

$2$3$5

1

.3

.3

$2$5

1

.3

.3

$2$6 $3$4 $3$4$5$9

1 1 1

.3 .3 .3

.3 .3 .3

To have my own water tap & to have my own toilet $4$4

1

.3

.3

1

.3

.3

1 384

.3 100.0

.3 100.0

$4$5 Total

WORKSHEET 11 READ THE CASE STUDY and answer the following questions and complete all the activities: What is the measurement level according to which the data for the case study was captured? ___________________________________________________________________________ ___________________________________________________________________________ ___________________________________________________________________________ Choose an appropriate graphical technique to represent the results in the case study according to theme, age group and theme by age. (Bar diagram, pie chart or histogram) Use the spread sheet in Table G1 for this purpose. ___________________________________________________________________________ ___________________________________________________________________________ Present your results according to your chosen technique in (1) graphically.

6

WORKSHEET 12 POST COURSE ASSIGNMENT Using the knowledge and skills you acquired during the three days, please identify a practical research problem within your workplace and write a short proposal as to how you will research the problem. Please base your short proposal on the following generic structure: Title page Background/rationale Literature review Research problem and objectives Research design and methodology Data analysis Time-frame Budget (optional) References This short proposal should not be longer than 10 pages.

7

SLIDES

CASE STUDY