NOTES
Descriptive statistics
A branch of statistics that contains ways of organising and summarising data so that the important characteristics can be described and communicated.
Inferential statistics
A branch of statistics that helps us determine whether sample data can be applied to the population from which they are taken.
Thematic analysis
A coherent way of organising the interview material and making sense of it in terms of the research question.
Timeline
A list, diary or calendar of activities linked to dates/specific times (i.e. deadlines).
Nominal scale/ level
A measurement scale in which a score/number is used for identification and does not indicate an amount.
Interval scale/ level
A measurement scale in which each score indicates an actual amount and there is an equal unit of measurement between two consecutive scores. A zero does not indicate zero.
Ratio scale/level
A measurement scale in which each score indicates an actual amount. There is an equal unit of measurement and there is a true zero or zero amount.
Ordinal scale/ level
A measurement scale in which scores/numbers indicate a ranking or rating.
Continuous variables
A measurement scale that allows for fractional amounts of the variable being measured.
Correlation coefficient
A number that represents the type (may be positive or negative) and strength of the relationship between two continuous variables.
Paradigm: definition
A paradigm is a model or system of practices and thinking that forms the basis of any research work.
Integrity
A quality involving honesty, principle and trust.
Back referencing
A starting point in information retrieval that involves obtaining the latest journal article on your particular topic. The list of references used in such article enables the researcher to identify a whole range of relevant literature.
Levels of measurement
A system of using numbers to measure variables.
Pearson r
A technique that tests the significance of the relationship between two continuous variables.
The object of your study is referred to as
the unit of analysis.
defines a research design as
"the plan or blueprint according to which data are to be collected to investigate the research hypothesis or question in the most economical manner"
So where do you start with developing your questionnaire?
(1) The first step is to define the aim or purpose of the questionnaire. (2) You then need to draft and refine your questions. (3) You then need to structure your questionnaire. (4) Finalise its format and layout. (5) Finally, you should pilot your questionnaire.
Non-probability sampling includes:
(a) Convenience sampling (b) Haphazard sampling (c) Purposive or judgement sampling (d) Quota sampling
The different probability sampling techniques are:
(a) Simple random sampling (b) Systematic sampling (c) Stratified sampling (d) Cluster sampling
Empirical evidence
(the record of one's direct observations or experiences) can be analysed quantitatively or qualitatively. Through quantifying the evidence or making sense of it in qualitative form, a researcher can answer research questions, which should be clearly defined and substantiated by the collected 'evidence' (usually called data). In some fields, quantitative research may begin with a research question, which is tested through experimentation in a laboratory or by means of a research instrument in the field. Usually, a researcher has a certain theory regarding the topic under investigation. Based on this theory some statements, or hypotheses, will be put forward as a possible explanation. From these hypotheses predictions about specific events can be formulated. These predictions can then be tested. Depending on the outcomes (research findings) the theory on which the hypotheses and predictions were based will be supported or opposed (disproved). Empirical evidence (as distinct from empirical research) refers to objective evidence that appears the same regardless of the observer. Accurate analysis of data using standardised statistical methods in scientific studies is critical to determining the validity of empirical research.
INTERVIEW SKILLS CHECKLIST: Analysing the interview information
- Identify main themes and group information accordingly. - Use direct quotes (interviewee's words) in the research report. - Reference (i.e. acknowledge source of) each piece of information (link to each interviewee). - List interviewees in the List of References section.
Unacceptable questions
- Misleading or ambiguous (What were your thoughts at the time of the use of force incident and the beating of the suspect?) - 'Catch-22' questions (e.g. When did you last beat your child?) - Overlapping (different questions which ask similar things). - Multiple possible answers (e.g. Tell me about your experiences of patrolling"). - Double-barrelled questions (i.e. ask more than one question at a time in the same question) e.g. Have you ever been accused and charged with the assault of a suspect? Yes/No. Whatever the response, it will be unclear as to which question the person has actually answered. - Leading questions, which make it very difficult for the respondent to give an honest answer (e.g. Don't you agree that he is responsible for assaulting the suspect? Yes/No). - Questions phrased in the negative (e.g. Police officers should not refrain from using force during arrests. Yes/No).
The sampling strategy to be followed should be explained in detail and substantiated from literature. Also, see 'unit of analysis' above.
- Who is your (target) population from which you are going to sample (characteristics of the required sample)? Why are they required? Where and when are you going to do this sampling? - How do you intend selecting this sample? - Why is this sampling technique appropriate to the study? - How will the sample size be determined? - How does it influence the way in which you will apply your findings?
INTERVIEW SKILLS CHECKLIST: The interview
1. Eye contact and body language 2. Probing questions 3. Note-taking 4. Follow-up interviews 5. Recorded interviews to be transcribed
INTERVIEW SKILLS CHECKLIST: Key actions
1. Formulate questionnaire/interview schedule of questions beforehand. 2. Obtain permission from relevant departmental head/supervisor to undertake interviews/observations. 3. Make contact with persons to be interviewed. 4. Arrange specific time/s and date/s for interview. 5. Take along a clipboard, paper and pen in order to write field notes. 6. If the interview is to be tape-recorded, ask interviewee for permission to do so.
GUIDELINE: Standard chapter layout:
1. Introduction/problem statement 2. Research methodology used 3. Literature review 4. Research findings (obviously if your research generates a great deal of information there can be a number of research findings chapters) 5. Recommendations and conclusions 6. List of references 7. Annexures
INTERVIEW SKILLS CHECKLIST: The interviewer
1. Respect for the respondent. 2. Ability to show an interest in his/her opinion or views. 3. Don't interrupt. 4. Pay attention. 5. Do not show disapproval of or openly disagree with what is being said (i.e. be non-evaluative; non-judgemental). 6. Do not be dismissive of the information given. 7. Be neutral at all times. 8. Be reflective. 9. Be aware of your own actions. 10. Be careful of the image you project. 11. Be careful how you ask questions (tone of voice/body language). 12. Avoid being patronising, arrogant, 'know-it-all' or smug.
Chi-square test
An inferential technique that tests whether there are significant differences in the frequencies of variable categories.
Structured interview
An interview that follows a certain predetermined order and that uses an instrument such as a questionnaire.
Profile
An outline or sketch of similar characteristics of a specific type of person.
Sample
This is the group of people/things that you choose to study. It is usually a small subset of a larger population.
Peer review
Articles submitted to journals for publication are thoroughly scrutinised for their academic and scientific merit by experts in the particular field. This process is known as a peer review.
Empirical research
Covers a range of research methods, such as participant observation, case studies, participatory action research, surveys, comparative cross-cultural and cross-natural studies, laboratory studies, field experimental designs, evaluation research, secondary-data analysis and content analysis (see Mouton 1996). It is a way of gaining knowledge by means of direct and indirect observations or experience. In scientific use the term empirical refers to the gathering of data using only evidence that is observable by the senses or in some cases using calibrated scientific instruments. What all empirical research has in common is the dependence on observable data to formulate and test theories and come to conclusions.
Randomness
In a random selection process, every element of the population has an equal chance of being selected for the sample.
Example 2: You are investigating the use of force by police officers in Gauteng Possible data sources in this example are:
Documentation on reported cases, accounts of victims
Haphazard sampling
Drawing a sample in a non-organised (haphazard) way, without any consistent guiding principle as a basis for selecting the elements of the sample.
Convenience sampling
Drawing a sample that is easily obtainable or accessible.
Systematic sampling
Drawing systematic samples can be easier and less time-consuming than drawing simple random samples. It is a quicker and easier and often a much more convenient way of sampling.
Probability sampling
Each element of the population has a known chance of being selected as an element of the sample
Simple random sampling
Each element of the population has the same chance as being chosen as any other. The elements in the population are thus selected individually, and because the process by which they are selected is a random one, each element has the same probability of being selected when the selection is made.
Research instrument
Forms such as questionnaires or techniques such as interviews that are used to collect information and data.
GUIDELINE: KEY THEORETICAL CONCEPTS
Full details and a discussion of concepts to be used in the research study. Do not merely provide a quote of a dictionary definition without any interpretation, explanation of how the key concepts will be used and applied in your study
Purposive or judgement sampling
In judgement sampling, 'experts' or expert judges are used to select the sample for a specific purpose. This form of sampling is normally used to select samples that are exceptional or populations that are difficult to reach. A flaw in purposive sampling is that different experts may have different views on who and who is not representative of the population.
Purposive/ Judgement sampling
In judgement sampling, 'experts' or expert judges are used to select the sample for your research.
We distinguish between one-stage and multi-stage sampling.
In one-stage sampling, a sample of the clusters is drawn and all the elements in the sampled clusters can be included in the sample. In multi-stage sampling, sub-samples can be drawn from each of the selected clusters. In other words, not all the elements in the selected clusters are included in the sample.
Research design
In the research context, a design refers to the specific way of conducting the study. What is the best way of addressing the research question?
Qualitative interview
Interview used in qualitative research.
MTech
Magister Technologicae degree awarded by South African universities (of technology) at masters' level.
Levels of measurement
Nominal Ordinal Interval
Demographics
People's characteristics (e.g. age, gender and occupation).
Study supervisor
Person appointed by the educational institution to provide you with advice and assistance on your postgraduate studies.
Example 2: You are investigating the use of force by police officers in Gauteng The unit of analysis in this example is:
Police officers
The three distinct paradigms are
Positivist Interpretive Constructionist
Background reading
Reading widely on the chosen research topic. The literature read should include a wide range of sources (e.g. journals, periodicals, newspaper reports, textbooks and magazines). This preparatory reading is known as background reading.
Plagiarising
Presenting the ideas of others as your own. This may involve copying ideas, word for word, from a resource or rewriting them as your own, without crediting the source of your information. Either way, this practice constitutes academic fraud and is to be avoided at all costs.
Processual development
Qualitative research is usually more flexible in that the design can change as the research progresses. It is useful to think of the qualitative research process in terms of phases of development rather than as steps to be adhered to rigidly.
The qualitative versus quantitative debate
Quantitative methodology is usually regarded as referring to the collection and analysis of numerical data. Qualitative methodology, generally associated with interpretative epistemology, tends to be used to refer to forms of data collection and analysis which rely on understanding, with an emphasis on meanings. The main schools of thought regarding the differences between qualitative and quantitative research approaches are as follows: = These two approaches come from completely divergent backgrounds (paradigms), ask different questions, are not compatible and therefore cannot be integrated. = The techniques used (interview, observation, questionnaire.) are independent of paradigm and should be appropriately used to answer the research question. = While these two approaches ask different questions, they have a shared commitment to the better understanding of issues and should be used to complement each other (i.e. a 'mixed-methods' approach).
Linear development
Quantitative research usually follows a series of prescribed steps, from 'formulation of the topic' to 'writing up the final report'. There is little or no deviation from these steps.
Open-ended questions
Questions to which there are no given responses, and which respondents can answer freely.
Coding
Refers to preparing the data for analysis. Codes contain information that is relevant to the research themes.
It is important to distinguish between your unit of analysis and your data source.
Remember that your unit of analysis is the object about which you want to draw conclusions, while your data source is the documents, records, people or other resources that provide your data.
Paraphrasing
Rewriting ideas that you have read in your own words. (Always remember to acknowledge the source of your information.)
Non-probability sampling: definition
Sampling procedures in which the likelihood or chance of selecting elements of the population is unknown. In other words, the sample is selected on the basis of accessibility or convenience.
Cluster sampling
Social research often means that the researcher has to draw samples from a population from which it is difficult to obtain or even compile the sampling frame. Examples of these are the population of a town, all police officers in the country, students at all schools in the country, medical personnel across the country. This sort of population compels the researcher to design a more complex sample. This normally involves initially drawing, from the population, certain groups of population elements. These groups are called 'clusters'. Elements are then drawn from these clusters, which then form the sample. In cluster sampling, the elements of the population are grouped together in non-overlapping clusters. The clusters do not have to be equal in size.
Scaled questions
Statements or questions followed by a rating scale. The respondent will indicate the extent to which he or she agrees or disagrees with the statement.
If you have a pile of questionnaires, you should follow three main steps:
Step 1: Organise and code the data. Step 2: Enter the data into the computer. Step 3: 'Clean' your data (correct errors).
Example 1: You are investigating students' performance in criminology Possible data sources in this example are
Student records, or the students' examination records for Criminology
Example 1: You are investigating students' performance in criminology The unit of analysis in this example is:
Students
Example 3: You are studying students' attitudes towards crime The unit of analysis in this example is:
Students
Piloting
Testing the data-collection instrument before it is finalised. Piloting is used to identify problems in the research or research instrument before the main study is conducted.
Data
Textual and numerical information generated by various research instruments and techniques.
Fieldwork
The actual data collection. This may take the form of an interview, questionnaire administration, observation, or the examination of records and documents.
Mean
The arithmetic average of a set of scores.
Standard deviation
The average by which a set of scores varies from the mean of that particular set.
Positivist paradigm
The belief that reality is objective and can be studied dispassionately.
Diagnostic test
The collection of information by means of questions to specifically assess existing skills levels.
Research: definition
The collection of information, facts and data through a process of investigation, examination, observation and testing.
Sampling error
The degree of error associated with using a sample to estimate a property of the population.
Constructionist paradigm
The idea that social and psychological phenomena are socially constructed (built, put together) and can be understood by looking at the elements that make up the phenomena.
Interpretive paradigm
The idea that the reality to be studied consists of people's subjective experiences.
Frequency
The number of times each score occurs within a data set.
Literature review
This is an integrated account of the literature available on the research topic. The discussion should include a critical examination of the research that has been conducted on the topic and the theoretical frameworks relevant to the study.
The respondent
The person providing you with information - usually by means of a questionnaire, interview schedule or focus group interview.
Transcribing
The process of transferring audio-recorded interviews on to paper.
Ratio level
This level of measurement is the most sophisticated (advanced) because it incorporates all the characteristics above. It classifies, ranks and has equal units of measurement. It also includes zero: in this instance, zero means that the characteristic does not exist. An example is the number of years a person has been in employment. Zero means the person has not worked. The difference between 3 and 7 years allows us to say that the latter has worked longer (rating) and the difference between 3 and 5 (2 years) is the same as that between 7 and 9.
Population
This refers to any group that is the subject of your study. It could be individuals, groups, events, organisations.
Conceptual definition
This refers to the description of a concept in order to convey its specific meaning.
Data analysis
This refers to the sense that a researcher makes of the data collected. It involves finding the meaning of the data with specific reference to the research question.
Motivation/rationale
This refers to the set of reasons for your research study. Why is the study important?
Nominal level
We use the this level of measurement when we use numbers. Example In a research study: 1 might represent a certain policing area; 2 another area; and 3 yet another area and so on.
Ordinal level
We use this level of measurement when we use numbers to rate or rank people, things or events. Example The number 1 could represent constables, 2 could denote sergeants and so on (in terms of seniority, the higher the number the higher the seniority).
Interval level
When a variable is measured using an interval scale, each score indicates an actual amount and there is an equal unit of measurement separating each score. Interval level is also referred to as an 'equal interval'. When a variable is measured using an interval scale, each score indicates an actual amount and there is an equal interval measurement. While interval scales include the number 0, it is not a 'true' zero. It does not mean zero. Example A scale on which the difference between 1 and 3 is the same as the difference between 4 and 6 (equal units separate the measurements). An example is the Celsius temperature scale. The interval between 3°C and 4°C is the same as that between 4°C and 5°C. There is no value that represents zero or nothing. 0°C is a specific temperature; it does not mean an absence of temperature.
To conduct stratified sampling, you need to
draw a random sub-sample from each of the strata. The combination of the sub-samples thus forms the total sample. sample size should be kept in mind.
Closed questions,
give respondents a number of limited options from which they have to choose. This then confines the responses to the options given.
Non-probability sampling
involves selecting your sample on the basis of accessibility or convenience. In the case of non-probability sampling, the principle of randomness does not apply. Everyone (the population) does not stand an equal chance of being selected into the sample. Non-probability samples thus do not allow generalisations outside the group of sample elements. Researchers themselves play a big role in non-probability sampling. It is essentially researchers who decide on who gets included in the sample or not and this can introduce its own bias Non-probability sampling methods are faster, more convenient and cheaper to apply than probability sampling methods.
Variables
is anything that changes in amount or type. For example, police violence, age and occupation are variables that could form part of a research study.
The actual list from which your sample will be drawn
is called your sampling frame.
the qualitative approach emphasises
meaning
quantitative approach emphasises
measurement
Stratified sampling
most often used when the members of the population are very different ('heterogeneous') in respect to a characteristic or variable that is being studied, and when the population can therefore be divided into sub-groups (or strata) according to that variable or characteristic. used when the researcher believes that he/she will get more reliable results by dividing the population into different, and yet more similar, ('homogeneous') strata. This means that you will have to know how the population is made up before you start your study. In a target population in which the researcher uses stratified sampling, elements in one sub-group will be more similar to other elements in that sub-group compared with elements in another sub-group. For example, the attitudes of a first-year student in a research study will probably be similar to the attitudes of other first-year students than, say, to the attitudes of fourth-year students.
There are two types of questions:
open-ended and closed questions.
saturation
you may start data collection without knowing beforehand the number of people you want to interview. You may continue interviewing the respondents until the same themes start emerging, and you have collected enough material. This leads to what is referred to as saturation. Saturation occurs when additional information will no longer add any new relevant information, and when the themes that have emerged are rich and thick with description.
Periodicals
These usually refer to non-academic publications covering topics of popular interest.
Information retrieval
This involves identifying and collecting relevant literature on the chosen research topic. It includes doing a subject search at the library, scanning research abstracts and dissertation abstracts, doing an internet search, talking to the experts and so forth.
Operational definition
This is a description of the way in which the concept/variable is to be measured. How is the variable to be measured?
Data {definition}
This is the information collected from the sample. It may take various forms: numbers, text, statistics, audiotaped or videotaped material.
Literature search
This is the search for anything published or written on a specific issue or topic. Such searches are usually conducted by using search facilities (e.g. abstracts database; catalogues; internet) at an academic library.
Plagiarism
This is the use, word for word, of other people's written work without acknowledging where you got the information from, literally 'stealing' other people's words.
Methodology
This refers to the techniques and methods used to collect information during the research process.
Quota sampling
The technique by which the population is divided into different groups based on certain characteristics (e.g. groups of males, females, old, young). Elements from these different groups are then proportionally selected to form part of the sample.
t-test for independent groups
The two groups are said to come from different 'populations' and are therefore independent. In order to use this technique, the data should be continuous and there should be at least 30 participants in each group.
Significance level
A value that helps decide on the probability of error in research conclusions and findings.
Variable definition
A variable is anything that changes in amount or type Variables are also characteristics that can take on different 'values', i.e. likely to change or vary; subject to variation; changeable; or something that varies or is prone to variation. In other words a quantity capable of assuming any of a set of values (in mathematical terms).
Code of Ethics
A written document outlining specific norms and behavioural rules that have to be adhered to and which govern the researcher's conduct.
Research proposal
A written outline of a proposed research project usually containing the title/topic, problem statement, research design and methods to be used for the collection of information and data, and a description of how the collected information/data will be analysed.
Simple random
Each element of the sampling frame has a known and equal probability of being selected in the sample.
Systematic sampling definition
Elements from a complete list of the population are drawn in a systematic or organised way for inclusion in the sample.
INTERVIEW SKILLS CHECKLIST: Drafting questions
- A good way to start is by simply listing all the problems you can think of. - Group similar questions together, revise your list and then make any necessary changes. - Make sure that the sequence of questions is logical. - Questions should be clear, simple and direct. - Questions should be asked in a way that is easy to understand. - No ambiguous questions. - Ask yourself whether the respondents will understand what is being asked. - Test (pilot) your questions with a colleague - he/she may add questions you did not think of.
Illustrate the steps in simple random sampling by using the example of the study of the use of force and violence amongst police officers. In that example, 500 officers are in Gauteng. Illustrate the steps you would take to do simple random sampling in this situation:
1 Develop a sampling frame >> In the example, you can develop a frame by listing all the names of the police officers from 1 to 500. 2 Decide how large the sample will be >> You should select a sampling ratio that is the same as in other fields of study. Because we have a small population size (500), we can use a sampling ratio of about 30% of the population (sampling ratios are explained later in this study unit). Our sampling size is thus 150 (n=150). 3 Select the 150 elements from the sample >> You can do this, for example, by using a table of random numbers and by selecting 150 numbers between 1 and 500. Another way of doing this would be to generate 150 random numbers on a computer. This ensures that each element in the population has an equal chance of being selected. 4 Refer back to the sampling frame and identify and list the elements that you have selected >> This list is your simple random sample.
GUIDELINE: VALIDITY, RELIABILITY AND ACCURACY OF COLLECTED INFORMATION
1 Ensuring validity 2 Ensuring reliability [Explain the criteria used to determine validity and reliability of the methods selected and the data collected.]
outline some of the preliminary steps (generally) that you would need to take when starting with a research project for an institution such as Unisa.
1 Select your topic 2 Writing your research proposal 3 Drawing up your timeline, budget and agreement with your supervisor
Once you have decided on the topic of your research, there are a number of steps you need to take to ensure that your research meets the ethical requirements referred to above. You should:
1. Ask permission from a person in authority to carry out the research. 2. Ask permission to use the research information collected. 3. Obtain the (willing/voluntary) consent of each respondent to be interviewed. If your potential respondent is not willing to provide you with information, do not force him/her to do so. Sometimes organisations like the police and correctional services require you to sign a letter of agreement adhering to these conditions. This is common practice. We also advise you, where possible, to get the respondents to sign a consent form giving you permission to interview them.
Below is a standard pro-forma structure - points that certainly need to be considered for inclusion in a good proposal:
1. TITLE: 2. INTRODUCTION - Literature review - Motivation - Research question 3. METHODOLOGY - Sample - Instruments/techniques - Procedure 4. PROPOSED ANALYSIS OF THE COLLECTED DATA/INFORMATION 5. TIMELINE AND BUDGET FOR THE RESEARCH 6. LIST OF REFERENCES 7. APPENDICES 8. CONCLUSION
There are a number of questions that can help you in developing your questionnaire.
1. What is the purpose of your study? 2. What type of information do you need? 3. What is your research question?
Use the following questions as guidelines when formulating your interview questions:
1. What type of information do you need to collect for your research report? 2. What question(s)/issues/problem/s do you want answered? * Remember that the aim of the questions would be to try to find out more about your research subject.
in terms of a literature review. Keep the following principles in mind:
= Concentrate on issues that are relevant to your research focus. = Aim to be concise. = Summarise the findings of past studies and integrate them into a meaningful description of the main trends. = Use the recommendations made by other researchers to back up your ideas. = Credit your sources of information. If you have taken information from an article, acknowledge the source of that information.
Some of the things to consider when you classify resources are the following:
= Is the area of research the same as yours? = Are the methods used in the research study appropriate for your study? = Does the study review other research that is directly relevant to your topic? = Does the reading identify gaps in the literature? = Can the recommendations made in the resource be linked to your research focus? = Does the article provide an argument for your study? = Are there other issues? (For example, the resource may contain different theoretical perspective of the research problem.)
The interview guide
= (schedule/list of interview questions) is a data-collection instrument used in qualitative interviews. = Qualitative researchers do not come to the interview with a set of open-ended questions. = the researcher has already drafted and listed - in sequential order - the questions he/she would like to ask the interviewee. = interview guide normally consists of only a core of primary questions or, in the case of some qualitative interviews = only one exploratory (exploring) question. = The formulation of the interview question(s) is therefore extremely important in qualitative research. = The formulation of the interview question(s) is therefore extremely important in qualitative research. = The aim of any question, in qualitative research, is to try to find out more about the experiences, perceptions, feelings, thoughts, and opinions of the interviewee. Questions should thus be formulated in a way that makes it easy for the interviewee to 'open up'. = The questions should be of an exploratory nature. =For example, "Would you like to tell me what you think/feel about ...........". You may then want to follow this up by asking more exploratory questions related to this topic.
Keep the research statement in mind at all times in terms of the following questions:
= What would you use as your main headings? = What subheadings would you consider? (Here you can use the keywords/themes that you have in mind.) = Is there a logical order in your outline? = What is the main function of your review? In other words, what specific function do you want it to serve? What message do you want to convey? = Does it have a common thread/theme running through it? = Is there enough information to give substance to the this outline?
Cluster sampling is frequently used for the following reasons:
= When no sampling frame for the elements of the population is available or does not exist (which means that a random sample cannot be drawn). = It can be used for economic and practical reasons (even when sampling frames are available).
GUIDELINE: Research design
= A 'research design' sets out the broad outline of how the research will be implemented and will typically include how (methods) the data will be collected, what research instruments will be utilised and the intended analysis of the collected data and research information. I = a systematic plan outlining the study, the researchers' methods of compilation, details on how the study will arrive at its conclusions and the limitations of the research. = may incorporate both quantitative and qualitative analysis. = incorporates the research methods most suitable to gather the required data and answer the original research question, = A research design varies by field, focus and topic and by the question being investigated. Many researchers combine qualitative and quantitative forms of analysis to better answer the research question/s. = The accountability of the research design is vital to the success and outcomes of the research project. = The detailed research design allows the reader to follow the methods and techniques for researching the research problem. = outline the reasons for selecting a particular research approach or method followed by an illustration and/or application thereof (how these scientific requirements were realised) within the context of the study. = It is a detailed description of how the research will be conducted in terms of: - Methods and instruments (techniques) used and reasons for choosing them.. - Design and/or use. - Is the study feasible? - Pilot study.
There are several ways of presenting a review:
= A brief introduction followed by two main sections: theoretical background and empirical research. = A description of the context (including policy or definition of terms, if appropriate) and theoretical background, followed by an analysis of the empirical research. Different variations on these combinations are also possible but all usually come down to either a sequential (connected) integrated approach or a categorised approach - where certain information is grouped and discussed in its entirety before moving onto the next category (issue). However, throughout there must be a logical flow of the information being presented.
GUIDELINE: DISSERTATION LAYOUT
= A broad plan or layout of chapters that demonstrates the key points in the dissertation. = Leads the reader through the successive stages, showing that the chapters are connected logically (what is required is is just the chapter heading of each proposed chapter with a short explanatory paragraph of each chapter's content and NOT just a heading/s framework).
Choosing a focus | Reading broadly on the topic is a good starting point, but the real challenge lies in producing the following:
= A clearly focused and specific question, which must be based on a sound argument. You should state the aim of the study. = A feasible research idea.
GUIDELINE: Data collection
= A detailed account must be given of how you intend collecting the data and what research instruments will be used. = Explain practical arrangements - access to sources, request/s to obtain permission to undertake the research (letters), relationship with participants, consent - Techniques and criteria selected for collecting information and data. - Are they appropriate for data collection in the context of your research - what is the 'degree of fit' between the aim of the research and the data-collection technique? - How does your intended data-collection technique help you to achieve your research aim?
What does a literature review involve?
= A literature review involves examining the literature available on the subject. = How much material/literature your subject search yields really does depend on your choice of research area = If you choose a more popular topic, or a topic that has aroused public interest, you might well find yourself inundated with information and literature on this subject = a more obscure topic may generate very little reading material for the literature review and such a search for information might require digging deeper and extending your search terms. that amount of readings. = A useful tip is to start filing your reading material according to themes. You could use the keywords submitted for your subject search as information categories for such filing.
chi-square test can be used only if the following holds true:
= A mutually exclusive assignment to categories. Each participant can only belong in one category. = The data are nominal/discrete. = There is a sizeable sample. You need to have a large sample so that a single cell (a block in which the data or frequency in this case, is entered) does not have fewer than five cases. (Otherwise, there is a high probability of error and the computer will not continue with the analysis.)
The following might be some of the factors that may impact on the availability of your unit of analysis:
= Access to the unit of analysis - You may have to write a letter requesting permission to gain access. = Time - When is the unit available? How much time do you have for your study? = Resources - In addition to the time factor, money, transport, and recording equipment may also impact on your study as far as the unit of analysis is concerned.
GUIDELINE: Unit of analysis
= All research questions (see below) are framed in terms of units of analysis; that is, what is it precisely that you want to research? What object, phenomenon, entity, process, event or cases? = The units of analysis (also) provide focus for the data analysis.
Some common interviewing errors:
= Asking too many questions = Asking leading questions = Asking closed questions = Asking sensitive questions too early in the interview = Being chatty and conversational = Not being respectful = Making negative judgements about people's views and behaviour = Rewarding people for their responses
To be a good researcher, you need certain skills:
= Being curious - being prepared to ask questions (an enquiring mind). = Interpersonal skills, i.e. being a 'people person', someone who can talk to people. = The ability to analyse and interpret information objectively and the ability to organise your thoughts logically. = The ability to express your ideas clearly in writing.
DEVELOPING THE RESEARCH PROBLEM
= Choosing a focus = Refining the focus
GUIDELINE: ETHICAL CONSIDERATIONS
= Clearly state the ethical considerations applicable to the research This also involves drafting a Consent Form]
Purpose of literature review:
= Communicates our knowledge of the research area - what research has and has not been done. = It gives us the opportunity to compare and contrast results of different research studies, thereby identifying common trends or any gaps in our understanding. = Sets the context of the study. = Provides the argument for the research focus.
HOW TO WRITE THE REVIEW: Some do's and don'ts
= DO NOT Reproduce chunks of information from articles. = Ethical research conduct includes, among other things, not plagiarising = you need to be selective about what you use in your write-up. = DO Include information that is relevant to your research focus. = Summarise what you have read, then write it in your own words (paraphrase it), = avoid misinterpreting the writer's meaning. = Acknowledge the source of this information = there are times when you may feel that the manner in which an idea is expressed is perfect for your purposes - to make a specific point. In that case, you may quote directly from the source, indicating this (using quotation marks and the author's name) clearly. = DO NOT Give mere descriptions of previous research = provide an evaluative summary of the material that relates to your focus. = Make a point of reading critically. = DO Be critical! If you do not agree with something, say so! Be careful of the tendency of some researchers to make sweeping generalisations based on flimsy evidence = Watch out for ethical violations, misinterpretations, weak methodology, poor designs and bad sampling. = Remember that the literature review is not simply a list of relevant resources; it is an analysis of related theory and research. = DO NOT Present summaries of research in list-like form without showing the links between the studies. = DO Summarise the main points of the articles, decide what is relevant to your research problem, and present it in an integrated way, showing how the findings of the different studies relate to each other and to your research focus. = present a coherent (consistent) argument based on the work that has and has not been done on your particular topic. Draw your own conclusions and link these to your research statement. = DO NOT Provide an overview of theory without reference to any empirical research and vice versa. = DO Give equal attention to the relevant theories that inform your research ideas and the reports of empirical studies that have been conducted on the topic. = Evaluate the different theories that have been put forward to explain findings. Which is most appropriate for your purposes and why? = DO NOT Rely entirely on dated (old) studies. = DO Use recent literature as far as possible. = DO NOT Depend on secondary sources = DO Make every effort to access the original article. = = =
FINALISING THE RESEARCH PLAN
= Describe the main reasons for the study (rationale or motivation); = Convert the research problem into a specific research question or statement. = Examine the best plan of action (research design) for answering the research question.
Your research proposal also needs to include the background to your study:
= Describe what you mean by the main terms of your research statement (outline of the 'problem' to be researched). = Describe the research that has already been done on this particular topic, and how it relates to your focus. = Explain the main reasons for your study (motivation and rationale). = Indicate what contribution your research is likely to make - practically, theoretically or both.
Before doing your fieldwork (collecting your information or data from your respondents), you need to do preparatory work to ensure that the data collection runs smoothly. Depending on the data-collection method, these preparations may vary. However, you do need to consider the following:
= Do you need to obtain permission from any person or organisation or community leader to conduct the interviews? = You need to make arrangements for setting up the interviews. This will involve setting the time for the interview, and arranging a venue = It is always a good idea to have a checklist for your fieldwork activities.
finalise and write up the research plan or proposal:
= Finalise the research question. = Describe the main reasons for the study. = Find the best research design for your research problem and think about appropriate research techniques. = Organise your time effectively and realistically (i.e. develop a time plan for research activities). = Estimate the costs involved.
A qualitative data analysis involves the following steps:
= First, you will break down your data into manageable segments. = Secondly, you will identify patterns or themes in the data. = Finally, you will interpret your data.
The following are some of the questions that you would need to ask to assess the viability of your selected research study. In each case, some of the answers that can be given are also provided.
= How many questions do you want to answer? What are these? = What are the specific issues, variables or concepts? = Describe the issues clearly. In other words, what exactly do you mean? - Can these issue be described in different ways? - What are these issues? - Why have you chosen the issues you have chosen? = Is the aim of the study clear? = Can a hypothesis be developed? = Are there reliable ways of measuring or examining the issues? What are these? = What methods/techniques would be appropriate? Why?=What method of analysis would you select? Why? = Does your research idea answer the questions: What?/ Why?/ Who?/ How?
In quantitative research, on the other hand, we tend to answer questions such as:
= How much? = How many? = How often? = Indicate on a scale what your feelings are about....? = Have you....? [Yes/No] Data generated by such questions is usually expressed in numbers, percentages, rates.
GUIDELINE: Data analysis and interpretation
= How will data be analysed - how to deal with the data once it is collected? = Specific processes must be described, and you need to refer to available techniques and literature consulted (e.g. code, cluster).
techniques to help make sense of information, remember important facts, and engage with the information in a goal-directed way.
= Identify the key points in a piece of information. = Break information down into manageable chunks. = Structure information and present the structure in a visual form (e.g. a diagram). =Identify key questions you have about information you read and try to find the answers. = Critically analyse information.
Qualitative data-collection methods
= Individual interviews / Focus group interviews = interviews use interview guides. = interview is structured in a way that it allows participants to express their feelings,/experiences/ views in own terms. The interview also allows participants to give their own meaning and emphasis to their experiences. = Participant observation/ Storytelling/ Role-playing
Units of analysis can be divided into four categories:
= Individuals - here you may be looking at categories of individuals such as adolescents, students, police officers. = Groups - examples of groups are sports teams, specific communities. = Organisations - examples include churches, police departments, government departments. = Social artefacts - this refers to both social actions and the products of human behaviour. Examples of social actions are marriage ceremonies, court hearings, traffic offences or elections. Examples of products of human behaviour could be cultural and symbolic objects such as books and paintings.
Qualitative data methods include:
= Interviewing = Observation = Storytelling = Drama
How to interpret the results of the t-test for independent groups
= computer printout will provide the t-value together with a significance level (p). = if the p value is less than 0,05, we can conclude that there are significant differences between the two groups. You will need to examine the group means to see which group obtained the higher mean (which indicates that the participants in this group have higher levels of job satisfaction).
The following are examples of different types of literature reviews:
= It may provide different understandings of a concept. = A literature review may also highlight and describe several different insights and views into the topic of your research. = Your research may require a different research approach - one that has not been used before.,In this case, the literature review will give prominence to the approaches used by other researchers, and explain why these approaches are ineffective or inappropriate for your purposes. = Let us assume that, to date, a topic has been researched using only quantitative approaches. In this case, the literature review will need to focus on the failure of past studies to realise the importance of including a qualitative approach to research, and the results of this failure. = The context of your particular focus may be different from other research investigations. In such a case, your literature review will need to include a clear description of the context of your research.
In any properly conducted qualitative interview, the following techniques will be used:
= Listening. = Clarification. = Reflection. = Probing. = Asking 'Why' questions. = Summarising. = Making use of silences.
Some of the disadvantages include:
= Low response rate, which may bias the sample. = Only people interested in the topic are likely to return the questionnaire - this, too, will create bias.
Hints for structured interviews
= Make sure that you have all the necessary items with you. This includes additional questionnaires, more than one pen, and a clipboard. = You should be thoroughly familiar with your questionnaire. If a respondent asks you to explain a question, you should be able to do so in a way that does not influence the respondent. It is always a good idea to practise beforehand. = You should be able to anticipate problem areas. = Be audible and make sure your questions are clear and straightforward. = Keep to the questions as asked in the questionnaire. = Be aware of your non-verbal behaviour and what this conveys to the respondent.
Kelly (1999:382) identifies some useful sampling strategies that could be utilised in planning interpretive research:
= Maximum variation sampling - this is used when the researcher seeks to obtain the broadest range of information and perspectives about the research topic. In this case, the researcher will probably find himself/herself interviewing participants who have different experiences or who think differently about the research topic. = Critical case sampling - data that are information-rich or enlightening are sought through what is known as 'critical case sampling'. This, in turn, may lead to a search for the exceptional case(s). = Confirming or disconfirming cases - data are collected that contribute to confirming or disconfirming the account ('story') that comes from the information collected. This will contribute to building a strong argument in cases where the account is being confirmed. However, data will also be sought to disconfirm the case. The case or account becomes rich and robust in the process. = Theoretical sampling - this refers to the process of collecting data for the purpose of theory building. This is a major element of Grounded Theory (Strauss & Corbin, 1990). In Grounded Theory, the researcher jointly collects data, codes and analyses the data and makes decisions about what data to collect next (on the basis of the researcher's needs).
Qualitative research
= Meaning model = Emphasis on words = Small samples = Researcher is involved = Authenticity - remain true to data = In-depth interviews/observation = Processual development (flexible, changing design) The term processual is used to indicate a flexible design, while the term linear refers to a series of steps followed in quantitative research designs.
Quantitative research
= Measurement model = Emphasis on numbers = Large samples = Researcher is neutral = Reliability/validity = Questionnaires, scales = Linear development (step-by-step design) The term processual is used to indicate a flexible design, while the term linear refers to a series of steps followed in quantitative research designs.
Quantitative research uses a more or less standard format for presenting research findings: tables, graphs and figures. Please note the following:
= Organise the presentation logically. Start by presenting your descriptive findings and then give the results of your advanced analysis. = Include a comment after each table or graph that highlights the main finding. = Means, standard deviations and frequencies are usually presented in tabular form. = Graphical displays (bar graphs, pie charts) are useful ways of illustrating frequencies. = t-values and correlation values, together with their significance levels, are presented in table form. = Remember that each table/graph has to be numbered and given a title.
To be a good researcher, you need to possess the following qualities:
= Patience and determined persistence. = The ability to pay careful attention to detail, = possession of organisational skills = Well-developed problem-solving skills. = Perseverance = A good, broad knowledge of and experience in your chosen field/focus area/topic of research.
Typical sources of information include:
= Periodicals, e.g. Policing Today = Journals, e.g. South African Journal of Human Rights; Acta Criminologica: Southern African Journal of Criminology or African Security Review = Books, e.g. Adams, RJ, McTernan, TM & Remsburg, C. 1980. Street survival: Tactics for armed encounters. Evanston, IL: Caliber Press. = Research reports e.g. Amnesty International (AI) 1996. AI Report: United States of America: Police brutality and excessive use of force in the New York City Police Department. Report - AMR 51/36/96 June. = The Internet, e.g. Hall, JD. 1994. FBI training on the new Federal Deadly Force Policy. Available at: www.fbi.gov/publications/-leb/1996/apr1966.txt (accessed on 20/07/2012). = Newspapers, e.g. Follain, J. 2001. General faces Genoa inquiry. The Sunday Times (UK), 29 July. = Magazines, e.g. Servamus; Hi-Tech Security Solutions; Security Focus.
Quantitative data-collection methods
= Personal interviews/ Telephone interviews = interviews are structured. = interview schedule or questionnaire used to collect data. = Quantitative indices (numbers) are used to represent the object or events being studied. = Self-administered questionnaire
When drawing up a budget you have to consider the resources needed, which may include the following:
= Time: How long will it take to complete the project? = Human resources: Do you need someone to assist you with the interviewing, data capturing or data analysis? = Money: How much will it cost to hire a tape recorder, print questionnaires, pay for an interviewer? What about possible travelling costs (in order to interview respondents)? What other consumables do you need (e.g. printing and binding of final report, books, literature searches and inter-library loans)? Do you need to include transport costs?
Inferential statistics rely on several requirements:
= Type of sampling procedure used. (Ideally, the sample should be randomly selected.) = The number of subjects used in the study and the number of related underlying assumptions. = The type of data collected. (Is it continuous or discrete?)
Refining the focus | Some of the common difficulties that lectures and supervisors face with students, especially with first-time researchers, are the following:
= Vague, fuzzy research questions = Unfocused research statements = The misconception that every study has to have an hypothesis (can be described as an 'educated guess') = Failure to consider the reasons for the study = A lack of concern about the feasibility of the study in terms of the availability of the sample and any specialised training required (i.e. for the use of certain data-collection techniques)
2. INTRODUCTION
= definitions of the key concepts; = the context in which the study is formulated; = reference to the main trends from past studies; = the main reasons for the importance of that particular area of research; = mention of the targeted sample.
Hints for qualitative interviews
= Practise beforehand and be prepared. = Practise using the tape recorder. = Ask participant's permission to record the interview. = Since you will be using a tape recorder to record the interview, make sure that the interviewing environment is reasonably quiet. = When conducting focus groups, be aware of group dynamics (interactions with each other). = Watch out for domineering persons in groups. In a polite and non-threatening manner, ask the domineering person to give quieter members of the group an opportunity to say something. = Encourage quiet persons to participate in the interview by making eye contact with them during the discussions.
When you identify your unit of analysis, you also have to identify the specific information about the unit of analysis on which you will focus. These include:
= Properties - this can be divided into conditions, orientations and actions. --Conditions refer to descriptions of individuals (e.g. age, gender), groups and organisations (e.g. size, structure, type) and social artefacts (e.g. size, appearance). -- Orientations refer to perspectives such as values, attitudes and organisational policy. -- Actions are behaviours such as violence and substance abuse. = Time - sometimes the properties of units of analysis are observed over time. In this case, the research is referred to as a longitudinal study. = Situations - research may involve the observation of units of analysis in different situations.
Advantages of the personal interview include:
= Respondents can ask for clarification if they do not understand some of the questions. = It has a high response rate.
Questions and/or responses need to meet the following requirements:
= Responses need to be mutually exclusive - this means that responses should not overlap. = Questions should be clear, simple and direct = Do not ask double-barrelled questions = Do not ask leading questions. = Ensure that the questions apply to all respondents.
reasons for doing background reading (one of the prerequisites for writing an effective literature review)
= Stimulates our thinking about potential research topics. = Introduces us to innovative ways of conducting research. = Alerts us to practical problems that may be encountered. = Informs us about appropriate theoretical models or relevant policy issues. = Clarifies our understanding of concepts. = Introduces us to new approaches in research.
In addition to the subject search, other ways of accessing material include the following:
= Talk to the experts. Consult your lecturers. They may either know of, or have in their possession, research reports or articles that you could use. = Scan the research conducted and completed (such as dissertations and theses) by other students at the university. = The National Research Foundation (NRF) also has a database of theses and dissertations. = Write to other institutions asking about their postgraduate research.
Some of the advantages of the postal interview include:
= The questionnaires are sent out at a low cost. = Since respondents can remain anonymous, their responses are likely to be more honest.
discrete variables.
= The term 'discrete' means "separate from each other". = Numbers that are discrete can only be expressed as whole numbers; = Nominal and ordinal variables are discrete. = when you are using a nominal level of measurement (e.g. assigning numbers to variables to identify them) and an ordinal level of measurement (assigning a rating or ranking),
GUIDELINE: TITLE
= The title will be your reference point and focus throughout your dissertation. = It is also a very brief abstract of the contents of the dissertation. = It must be concise, and yet clearly state the focus of the research.
Non-probability sampling is often used when:
= There is no sampling frame. = If the cost of probability sampling in terms of money and time is too high. You may need to do a rapid study of opinions and perceptions. = You are planning qualitative research. = You are doing exploratory research.
GUIDELINE: RESEARCH AIM/S, OBJECTIVES AND PURPOSE
= These precisely explain the focus of the research. = Specific research aims must be related to certain problems. = The ultimate research aim is to put forward evidence in support of an argument related to the topic of the research AND to add new knowledge/information to the existing body of knowledge. = The research aim is a general statement of what the research aims to accomplish. = The objectives of the research are smaller entities with specific outcomes as opposed to the more broader and general aim. = Therefore, the formulation of research questions 'inform' and guide the research - - they provide a focus and a framework to the research. (The research aim and objectives serve the same function as the delimitation (scope) of the study area and the assumptions on which the study is based).
Critical analysis
= Thinking carefully about the information. Do not accept any information at face value. You should examine the source of the information and weigh up its reliability. = Being aware of the different interpretations of an issue. {+} Evaluate it. {+} Give your own opinion. {+} Substantiate your opinion.
GUIDELINE: INTRODUCTION AND PROBLEM STATEMENT
= This orientates the reader in terms of the context of the research. = Introduce and briefly provide reasons for the research. = Research problem (problem formulation or research question) and rationale - identify and explain the nature of the research problem. = The rationale (or logic): -- Why is the research being conducted? -- What are the possible benefits of the research? = Explain how the research is part of a broader context of academic inquiry. = Preliminary literature review (background preparatory reading). Such preparatory reading should be done specifically to help you formulate your problem statement. It is NOT NECESSARY to insert a full 'literature review' into your research proposal (a full literature study/review will form a chapter in the final dissertation). Remember that your literature study is an ongoing activity throughout the entire research process.
PROPOSED ANALYSIS OF THE COLLECTED DATA/INFORMATION
= This section of the proposal should describe clearly the methods of analysis that are to be used to reduce the data into a meaningful form (i.e. analyse the data) and to make sense of what you have discovered (i.e. interpret the data).
First, let us see what we need to consider when drawing up a timeline. You already have an idea of how much time you have available to work on your project during a typical week. Consider the following:
= What are the university's requirements regarding the time period for completing a report/thesis/dissertation? = What goals would you like to set for yourself in terms of completing your research? = Are these goals realistic? = What short-term or long-term work commitments/family commitments do you envisage having an impact on your research?
The goal of qualitative research is to answer questions such as the following:
= What is going on here? = What does it feel like to experience a hijacking? = What is it like to be part of a particular group? Data generated from these questions is usually in the form of words - verbal descriptions, texts. A focus-group interview research technique is a good example of qualitative research.
DRAFTING THE RESEARCH PROPOSAL In your research proposal, you should try to answer three basic questions:
= What is the research about? = Why is it important to study this area of research? = How do you plan to answer the research question?
Why is background reading important?
= Without any background information to refer to, the researchers would not be able to convey the importance of the project and their knowledge of the topic. = The researchers' work will be unconvincing, simply because they would be unable to show the reader that they know about the topic they are researching. = A literature review helps to put a research project into context. = provides the background to the study by describing the context, current situation in the area of research selected to be undertaken, and what other researchers might have found (research findings) on the topic. = a literature review is based on background reading. = Unless the researcher does background reading, he/she may end up formulating a research question that has already been investigated. = Background reading puts the researcher in a better position to make recommendations relating to the field of research. = Good research and a comprehensive literature review will assist a researcher to demonstrate a thorough understanding of their research topic. = the research report should allow a researcher to, at the very least, make recommendations based on the research findings. = Without background research, a researcher may be oblivious to practical problems/obstacles that other researchers might have encountered. = For example, problems such as delays in getting ethical clearance, the time-consuming nature of file (docket) perusals and a research decision not to use certain information/data collection methods in one's research.
While doing your literature review you may encounter the following problems:
= You may have used incorrect or inappropriate keywords in the search. = You may be missing relevant information on the topic. = The research field may be too wide, which means that the amount of reading material retrieved will simply be unmanageable. = Difficulty in accessing printed material. An article may not be available locally or nationally. A book may be out of print. = You may have to pay for some information on the internet. = Some information on the internet may be difficult to download because a specific programme = You may experience delays in obtaining material through interlibrary loans. = The material may only be available in a foreign language.
when you undertake your research you might find that you encounter a number of delays and frustrations. We have already discussed some of these. Others might include the following:
= Your respondents do not meet you or fail to turn up at the appointed time. = Some might change their minds about participating or being interviewed. = You might experience delays in obtaining written information or books from the library. = You might even experience your computer crashing or being infected with a virus. (Get into the habit of making frequent document backups to an external hardrive!) = You may need to rethink or adapt your methodology after encountering unexpected obstacles in the field. = Your supervisor might leave the country or alternately go overseas to attend an international conference or take sabbatical leave.
Why do you think it is important for you to develop a timeline?
= Your supervisor may require you to submit your timeline to him/her so that he/she has an idea of when you will be finished with your literature search, or your data collection, or the whole research project. = drawing up a timeline will force you to make an open, public commitment to someone else, which might help to keep you motivated when things get tough. = Timelines are closely linked with schedules. = indication of how much time you realistically have available in your weekly schedule to work on your research project
Drawing up your timeline, budget and agreement with your supervisor
= Your timeline is an indication of how long it will take you to do your research and the deadlines for the various research activities. = look carefully at what available time = be realistic in your time allocation. = research not only takes time if you want to do it properly, but also tends to be costly. = need to draw up a budget of research costs and choose the methods you will use to collect the information. All these things require good time management. = Are you sufficiently self-disciplined to manage your time efficiently? = Is your timeline realistic? = Is it attainable? = give an undertaking and make a commitment to your designated study supervisor about beginning your studies and completing the work on schedule.
Research
= a series of activities (sometimes using specific research instruments and/or techniques/methods) = carried out in order to collect information systematically. = relevant information must be recorded (written down) and analysed using appropriate procedures and techniques. = results of your analysis must finally be presented in some form. For example, as a research article, dissertation (M-level), thesis (D-level) or research report.
t-test for correlated groups
= a t-test for correlated groups is used. The job satisfaction (PRE-TEST) scores obtained before the incentive programme are compared with the job satisfaction (POST-TEST) scores after the programme. = there is a significant (p-value less than 0,05) improvement in the post-test scores, we can conclude that the incentive programme was effective. (Note that the term 'correlated group' is just a synonym for 'related group' and should not be confused with the earlier discussion on correlations.)
Mind maps
= a way of putting down concepts, ideas or information in a diagram that follows the way the mind actually thinks. = In drawing a mind map, one would start by putting the main topic or key concept in the centre (usually in a circle), and then work outwards. = related ideas 'radiate' out from the centre, with additional lines splitting off (like the branches of a tree). = the simple technique of combining the 'mapping' of words with different colours improves memory = the brain 'tunes out' and becomes bored with reading a long list of ideas written in a single colour = Mind mapping is an extension of the keyword note-taking technique
An open-ended question allows respondents to answer a question in the way they want to;
= allows respondents to formulate their own responses. = may well elicit more than one response from a participant. = length of the questions may thus vary from a few words to a few lines and so on. = answer will also vary from respondent to respondent.
Are the resources peer-reviewed?
= any research that is regarded as scientific has to be evaluated by experts in the field so that an acceptable academic standard is maintained = If the resources are articles from accredited journals, reports from recognised institutions and books published by reputable firms, then it is probable that they have been refereed or reviewed prior to publication. = Research reports and conference proceedings are also usually scrutinised by fellow researchers.
Select your topic
= good rule to follow here is to choose a topic that interests you, and that is, ideally, in your line of work = additional factor that should influence your choice is a need for specific research on a topic = Once you have identified that interest you must read up on the topic. = request this information officer to do a subject search on your selected topic; will give you a good idea of the topics that have already been researched. = Once you have read widely on your topic, you will be in a position to think about the title of your research work.
How to interpret the results of chi-square test
= if the p-value is less than 0,05, this means that males and females differ in their response to the question. There are significant differences between males and females in terms of their willingness to change jobs given the opportunity. This technique is very useful when you have two discrete variables and where each has two or more categories.
The nature of the qualitative interview
= interview is a verbal technique to obtain information. = can be characterised as a face-to-face interview either with an individual (individual interview) or with a group (focus group or group discussion). = often described as non-directive, since it allows the interviewee to respond in a manner that he/she is comfortable with, using his/her own words and expressions. = manner that he/she is comfortable with, using his/her own words and expressions. The interviewer therefore needs to create a space where the interviewee will feel comfortable, respected and listened to. = non-threatening environment, enabling the interviewee to express him or herself in the manner the interviewee feels comfortable with, without the fear of being judged. = often criticised for being unstructured. = The trained qualitative interviewer makes use of a variety of techniques to provide structure to the interview.
PLANNING YOUR ACTIVITIES
= meticulous preparation and planning are crucial. = As a result, they want to get on with the research process - drawing a sample and collecting the data. = In fact, careful planning will mean that the final product (the research report) will be significantly better.
Interviewing is a very common method of collecting data in any form of research. There are different ways of conducting an interview. These include:
= telephone interviews = personal interviews = postal interviews
In conveying the importance or significance of the study, you need to give careful consideration to the following:
= the context (situational background/framework) of the topic to be researched = the urgency of the situation (whether or not the research is important/urgent; whether you need to collect certain, specific definitive information) = practical implications of the study in terms of recommendations/formulation of possible solutions
Sample section
= the group of people/things you decide to be participants in your research study. = sample could consist of a group of police officers, a set of documents (e.g. medical records), or a group of institutions (e.g. all South African universities). = For the research proposal, this section should contain a description of the intended sample - include only those characteristics that are relevant to the research problem. You should also explain your method/process of choosing the sample.
The factors that play a role in the choice of sample sizes to be drawn from the different strata include:
= the size of the strata = the survey cost per unit within the different strata = the variation in the relevant characteristic within the different strata Quite often, the size of the sample that is drawn from a particular stratum is in proportion to the size of that stratum. This is known as proportional stratified sampling.
The question that you need to ask yourself here is: Why do research?
= to make well-informed decisions you need to undertake research. = Information based on thorough research is usually regarded as valuable and credible. = Any crime prevention strategy that you would like to implement should be based on good research
There are a few basic rules essential for doing research with honesty and integrity.
= treat any respondent with respect and sensitivity. = once you have given an undertaking that you will not disclose the respondent's identity, you must at all times adhere to your undertaking; not only to your final research report, but also to your communications with other people. = acknowledge the source of your information. = If you fail to acknowledge the source of your information, this is generally considered to be plagiarism
continuous variable
= variables that are continuous can be of any value between two points. = An example of a continuous variable is that variable which you assign to the length of time a person has been working. This can be, for example, 6 or 7 years, but also any value in between, such as 6,5 or 6,7 years.
The following are some questions you could ask to determine whether the research idea would be feasible:
=Is it appropriate to my field of study? = Can it be done in the allocated time and with the available money, resources, skills? = Is the time period for my study realistic? Have I allowed for the possibility of delays? = Is the target population (sample) available? = Is it ethically acceptable? = Is it within my capability? Am I being realistic about what I can and cannot do? = Has this been done before? Is my specific research topic unique in any way? = What contribution will this research make to the field or existing body of knowledge? = Is there enough background information? = Am I truly interested in this topic enough to stay focused on it?
Piloting: Have a look at each completed questionnaire. You need to assess each and every aspect of it. This includes the introduction and all the questions. Ask yourself the following:
=Were the questions (or introduction) clearly understood? What questions did respondents misunderstand/require clarity on? List the problems and identify the questions that need refining. = Did the questions produce the types of responses you wanted? In other words, did the questionnaire measure what you intended it to measure? = Were the response categories inclusive? Put differently, do you need to add a wider choice to the responses of some questions? = How long did the interview take? Do you think that this was sufficient time? = In your opinion, how did participants respond to the length of the questionnaire? = Is there any information that needs to be included or removed from the questionnaire? = Do the questions address the research question at hand?
Schedule
A list, programme or plan of activities, tasks or events together with the times at which each thing should be done (similar to a timeline).
Discrete variables
A measurement scale that allows for measurement only in whole-number amounts.
Rating scale
A rating scale is a special type of questionnaire. It usually consists of a set of statements about an issue (e.g. attitudes toward capital punishment). Respondents may be expected to rate the extent of their agreement on a scale from one to five, with one indicating strong agreement and five indicating strong disagreement.
Sampling frame
A sampling frame is the list or tool from which the sample is selected.
Interlibrary loans
A service offered at most institutions of higher learning. Books and articles that are not available at your institution's library are obtained from other institutions at the researcher's request.
Qualitative research {definition}
A set of approaches that usually involve small sample sizes, with an in-depth examination of the experiences of the participants. The verbal accounts of the participants are examined using, for example, thematic analysis. An example is a case study of selected samples of police officers, using an in-depth interviewing technique.
Quantitative research {definition}
A set of approaches to research that usually involve large sample sizes, the use of numbers to measure variables, and statistical analysis to make sense of the numbers. An example is survey research.
Mind map (definition)
A technique that uses key words to construct a visual but structured form of taking notes that link ideas.
Paradigm
A typical example, pattern or model of something.
Example 3: You are studying students' attitudes towards crime Possible data sources in this example are:
The students themselves
Stratified sampling | definition
The technique by which the heterogeneous population is divided into homogenous sub-populations and random samples are selected from each sub-population.
Procedure
In this section you should describe how ethical issues are to be handled - how you are going to obtain permission from the authorities concerned and from the participants themselves. = In what setting do you plan to conduct the interviews? = How are the interviews to be carried out? = How many interviews do you plan to conduct? = Will it be necessary to do follow-up interviews? = Is it important to use a neutral setting? = If questionnaires are to be administered, is it possible to do this in a group setting?
Closed-ended questions
Questions with a number of fixed responses from which a respondent must choose only one response.
Sampling
The process used to select cases for inclusion in the research study.
Paradigm: Positivist
The researcher believes that he or she can adopt an objective stance towards the subject. The researcher also believes that the subject of study is both stable and unchanging. Methodologies used within this paradigm: Quantitative methodology
Paradigm: Constructionist
The researcher following this school of thought believes that reality consists of a fluid and changing set of social constructions. Methodologies used within this paradigm: Textual analysis and dismodule analysis.
Paradigm: Interpretive
The researcher recognises that the reality he/she intends to study consists of people's participant subjective experiences. Methodologies used within this paradigm: Quantitative methodology
Dichotomous questions
The responses to these types of questions are limited to Yes or No. The questions should thus be asked in a manner in which clear 'Yes' or 'No' answers can be given. These questions are more appropriate when dealing with facts.
Topic
The selected area or subject of your research.
Interview guide definition
The set of interview questions used to guide the researcher in the interview.
THE NATURE AND PURPOSE OF DATA: Qualitative analysis
Type of data usually involved in this analysis/// Data is usually textual. For example: = transcribed interviews Purpose of this type of analysis/// To gain a more in-depth understanding of the subject of your study.
THE NATURE AND PURPOSE OF DATA: Quantitative analysis
Type of data usually involved in this analysis/// Numerical data, that is, numbers. The numbers can be used in different ways. For example: = to count the number of times something happens (the frequency) = to separate things into groups (categorise) = to order (rate) things = to measure things Purpose of this type of analysis/// To describe the subject of your study in numerical terms and/or draw conclusions based on these numbers. Example of this type of analysis/// The quantitative component of the study on police misconduct involved an examination of the frequency of categories of violence as recorded in the discipline files.
Dismodule
Verbal communication, talk, conversation; a unit of text used for the analysis of certain linguistic patterns or trends that may range over more than one sentence.
drafting a research proposal. In order to achieve this aim, you were required to focus on three questions:
WHAT? = What is the research focus and, more specifically, what is the research question? = What is the background to the study? WHY? = Why is it important to study this area? What are the main reasons for the study? HOW? = How do you intend tackling the research question? = Who or what are you focusing on? = What techniques do you plan to use and why? = What are the steps in the process of collecting the data?
GUIDELINE: LIST OF REFERENCES
[Check your list of references against the referencing techniques prescribed by your academic department.] This is the list of sources consulted. Note that these sources should be up to date and relevant. (i.e. the latest edition of a particular publication).
GUIDELINE: ANNEXURE A: TIMELINE/ACTION PLAN
[Each activity in the research process in table form, with a column showing completion dates/deadlines.]
GUIDELINE: RESEARCH METHODS & DATA COLLECTION
[Methods/techniques and research instruments you will use to collect the information you need. In other words, how you are going to do the research.]
Timelines
a list of activities linked to dates or specific times (deadlines) by which a certain activity has to be completed. When embarking on research, it is essential that you develop a timeline to guide you and keep you on track throughout the process. In other words, like a study plan, you need to continually consult your timeline to check whether your research project is on schedule.
Journals
are discipline-specific, scientific and academic publications that contain reports on empirical research, research reviews and theoretical articles.
Scaled questions
consist of a statement or a question and a scale: respondents have to indicate the extent to which they agree or disagree with the question or statement. Scaled questions are most often used to measure attitudes or personality, because they give respondents the opportunity to indicate fine differences in meaning or in their views.
Hypothesis
is commonly referred to as an 'educated guess', and is a statement about how variables are related. Also termed a proposition, or set of propositions, (meanings of some information that support or deny a fact or group of facts) set forth as a suggested or tentative explanation for an observation or the occurrence of some specified group of facts (phenomena, events, incidents or trends), either put forward (claimed) merely as a provisional speculation, conjecture or assumption (guesswork) to guide research as a 'working hypothesis' or accepted as highly probable in the light of already established facts.
Reliability
is established when a technique or instrument proves to be consistent. If the technique is applied to the same object repeatedly and it gives the same results every time, it is reliable.
Phenomenology
is the philosophical study of the structures of subjective experience and consciousness. In phenomenology one must distinguish between objects as interpreted by human awareness (feelings), experiences and observations that generate understanding of the phenomena itself, and the objects (things) themselves which humans cannot directly experience. In its most basic form, phenomenology thus attempts to create conditions for the objective study of topics usually regarded as being subjective, i.e. consciousness and the content of conscious experiences such as judgements, perceptions, opinions and emotions of individuals. Although phenomenology seeks to be scientific by means of systematic reflection in order to determine the essential properties and structures of experience. Phenomenology believes that by analysing daily human behaviour can provide researchers with a greater understanding of its nature. This 'exploratory observation' of persons must be in the context (setting) of the society, community and place where they live since such examination can be better understood as it discloses an individual's 'lived experience'.
What kind of literature is regarded as academically suitable for preparatory reading and (especially) for the review?
make the distinction between scientific research & ordinary everyday research. You should apply the same standards to the literature you collect.
Saturation
occurs when additional information will no longer add any new relevant information, the trends in the collected information have been established and when the themes that have emerged are rich and thick with description. Further information will not add anything 'new'.
The various strategies used in research are commonly divided into two groups:
qualitative and quantitative approaches.
Epistemological
refers to a branch of philosophy that studies (investigates) the origin, nature, methods and limits of human knowledge. linked to the analysis of knowledge and understanding. In other words to discover the meaning of knowledge or the narrower meaning which is covered by the term 'theory of knowledge - literally the study of knowledge which emphasises the distinction between 'knowing how' (to do something) and' knowing that' (a fact, information, piece of knowledge exists), i.e. the difference between practical knowledge and theoretical knowledge.
Piloting
refers to the testing of your instrument. Before you undertake your fieldwork, you need to ensure the reliability and validity of your instrument. In other words, you want to ensure that you are measuring what you intend to measure (validity) and that the questionnaire will enable you to answer the research question (reliability). In order to pilot your instrument, you need to test the instrument with your target group. This means that you need to arrange interviews with your target group.
Illustrate the steps in Systematic sampling by using the example of the study of the use of force and violence amongst police officers. In that example, 500 officers are in Gauteng. Illustrate the steps you would take to do Systematic sampling in this situation:
select our sample by means of systematic sampling. If, in our sampling frame, the elements are all listed randomly, we do not first have to number all the elements in the frame to select our sample. What we will do, instead, is to select every third (this is calculated and called n) element from the frame. If we want to select 150 police officers from the sampling frame, we have to calculate the number (n) by dividing the population by the sample (500/150). The answer we get is approximately 3 and this is equal to n. We thus have to select every third case on the sampling frame. To do this, we have to select a random number between 1 and 3 (e.g. 2). We start by selecting case 2 on the sampling frame and then we select every third case on the sampling frame (e.g. 2, 5, 8, 11) until we have our 150 cases. We then have to go back to our sampling frame to identify our cases. In other words, we have to do the following: 1 Develop a sampling frame. All the elements will be listed randomly. It is not necessary to number the elements. 2 Decide how large the sample will be. As in the previous example, we want a sample size of 150. 3 Decide on what cases should be selected. This we can do by dividing the population size (500) by the sample size (150). The answer is 3. This means that we want to select every third case. 4 Select the cases. To find a point from which we can start selecting, we select a random number (e.g. 2). Starting at case 2, we select every third case from there.
Quota sampling
the researcher first identifies the sub-groups of individuals in the population and then selects the elements from each sub-group. The size of the sub-group is estimated from, for example, census data or other sources. The number of elements selected from each sub-group will be proportional to the number of elements in the population. By dividing the population into sub-groups, the researcher ensures that a wide range of groups is represented in the sample. Quota sampling is used frequently, where opinions need to be obtained quickly, but where the various opinions of the different target audiences need to be considered. Quota sampling thus ensures that the sample is selected in a manner where the key demographics that are relevant to the study are included in the sample selected. The problem with quota sampling is that it may be difficult to reach certain of these elements; also, it relies on the researcher's ability to be able to find the correct people
Convenience sampling
the sample is confined to parts of the population that are easily accessible. This could include a questionnaire in your newspaper or children in an aftercare centre.
There are two main types of two-sample t-tests:
the t-test for independent groups and the t-test for correlated groups.
