Research designs & data collection methods
By Debbie Harris, Contributor

I trust that you have read all the information that I gave you last week.

Do not forget to keep it in a safe place so that during revision time you can find it easily. This week we will continue studying research by looking at sampling. In addition, we will analyse the notions of validity and reliability in both argument and research. But first, how did you manage to apply the appropriate type of research and data collection method(s) in last week's exercise? Here are the suggested answers. Let s see how you did!

QUESTION 1

Any TWO of the following are sources from which Kishauna could gather background information on the research topic:

* Previous similar researches
* Archives at television stations
* Newspaper clippings
* Magazines

QUESTION 2A

Any TWO of the following are methods of data collection which Kishauna could use in her research

* Opinionnaires
* Interviews
* Observation schedules

QUESTION 2B

If you chose the opinionnaire then any of the following strengths/weaknesses could be given:

Strengths
* It facilitates a survey of a large population
* It is easily administered
* It is easily and reliably scored
* It allows for anonymity

Weaknesses
* Limited scope for probing responses
* Inflexibility
* Responses can be misleading
* Limitations posed by the literacy of the respondents
* Respondents cooperation is imperative

If you chose the interview then any of the following strengths/weaknesses could be given:

STRENGTHS
* It yields in depth information
* It allows the researcher more flexibility
* It can yield a high return of data

WEAKNESSES
* It gathers a lot of unnecessary information
* It can be affected by researcher's biases
* It is not always easy to set up interview sessions
* The respondent s memory can be poor the mind selects only some aspects,
others may be lost.

If you chose observation schedule then any of the following strengths/weaknesses could be given:

STRENGTHS
* It allows for the collection of primary data
* It can yield a high return of data
* It allows for researcher flexibility
* It lessens respondents bias and possible interference, which could contaminate the data

WEAKNESSES
* It can be affected by researcher's bias
* Reactions of the respondents may be misinterpreted
* Important and relevant data may be missed as chosen times of observation may not be appropriate or even significant

QUESTION 3

Any TWO of the following are sources from which Kishauna could collect her research:

* Students
* Parents
* Any other family members
* Teachers

Well, if you had all the appropriate responses, CONGRATULATIONS! You would have been awarded the maximum number of marks in the examination.

SAMPLE SELECTION

Let us now turn our attention to the issue of selecting a sample in research. A researcher who undertakes a survey involving large numbers of people will find it impractical or even impossible to canvas the views of everyone in that population. He has to then ensure that he takes a sample of the total population. The sample is a portion or fraction or subset of the total population. It is assumed that the sample is homogeneous and that the characteristics found in the sample may reasonably be expected to be found in the whole. When generalisations are to be made about the total population this sample should accurately reflect the opinions, beliefs or choices of the population. Therefore it must satisfy two basic conditions:

* It must be numerically large enough to correctly reflect the views of the whole population ­ in other words, it must be sufficient or adequate.

* All segments of the population indicated by relevant variables such as sex, age, social class, occupation etc. are to be proportionately included in the sample in other words, it must be representative.

There are two types of sampling: probability and non-probability. Probability sampling, also known as random sampling, is one way of making sure that the sample is representative of the population and that the procedure used in your research has validity. It ensures that no member of the sample is likely to be preferred than another in the selection. The selection cannot be predicted. Here the researcher's bias is eliminated and the procedure becomes more consistent and efficient. Non-probability sampling, also referred to as convenience sampling is used when the total extent of the population is not known. The researcher selects the sample at his discretion not randomly. One cannot guarantee validity and the absence of bias in this kind of selection and the results gathered using this type of sampling may not be totally reliable. However it has some advantages; it is less expensive and uses less time and resources.

For more information on sampling read Introduction to Social Research, Canoe Press university of the West Indies, Mona, Jamaica (1997),Chapter 4 by Ian Boxill et al.

I used the terms VALIDITY and RELIABILITY in the preceding paragraph and they are commonly used in Module 1 of the syllabus. In this module reference is made to the methodology used by the researcher and in the evaluation of the argument they refer to truth and believability. It is important that you understand fully what each term means in both contexts.

Let us look at validity in research methodology. Validity, to the researcher, refers to the extent to which the data collection methods or the research instruments, such as the questionnaire, interview and observation, actually measure what they are supposed to measure. In the evaluation of the argument, validity of the information which you get from a source, refers to whether the information is based on good judgement, reasoning and evidence, as well as if it is sound, logical and incontestable. Validity of the information also refers to truth, that is whether each statement in an argument is factually accurate and leads logically to the conclusion.

SOME CHARACTERISTICS OF A VALID ARGUMENT/INFORMATION:

1. Information should be

* Logical in sequence and relevant; premise should logically lead to conclusion and both should relate to the same concept. There should be no mixture of fact and opinion.

* Sufficient and complete

* Consistent and agree with what is generally known about the subject matter.

2. There should be no shifts in the meaning of key terms and concepts throughout the argument.

RELIABILITY refers to how accurately and consistently data collection methods or research instruments perform by measuring the factors for which it was designed.

Now, we have learned much today. Let us see whether you can apply the appropriate information. Here is the continuation of last week's question. Read the information and answer the questions which follow.

Kishauna is conducting research into the kinds of television programmes students at her school prefer to watch.

1 (a) Suggest TWO errors that Kishauna might make that would affect the reliability of her research. (2 marks)

(b) Explain how EACH error identified in (a) above would affect the reliability of the research. (2 marks)

2 State TWO considerations that Kishauna should bear in mind as she selects a sample for her study. (2 marks)

Look out for the suggested answers next week! Continue working on the Expository section, do not wait until the last minute to do it because you cannot carry out all of the research in just one day. Until next week, walk good!