choosing a research method

26
CHOOSING A RESEARCH METHOD An Assignment Report Based on Paper-I (Research Methodology) [As per requirement for pursuing M.Phil. under The Global Open University, Nagaland] Submitted by Shakil Ahmed Shakil A student of M.Phil.

Upload: mckohima

Post on 27-Apr-2015

1.544 views

Category:

Documents


2 download

DESCRIPTION

In this assignment report, I have discussed briefly about various research methods, their advantages and disadvantages as well as few design issues what I came to learn from various sources.

TRANSCRIPT

Page 1: Choosing a Research Method

CHOOSING

A

RESEARCH METHOD

An Assignment Report

Based on Paper-I (Research Methodology)

[As per requirement for pursuing M.Phil. under The Global Open University, Nagaland]

Submitted by

Shakil Ahmed Shakil

A student of M.Phil.

Page 2: Choosing a Research Method

Abstract

Research is a scientific and systematic search for pertinent information on a

specific topic. Different people carry on research works to serve different purpose.

Whatever the purpose may be, a researcher should always aim at a successful

completion of his work. A successful and good research relies on the accuracy of

data collected which in turn relies upon the method used. So it is very important

for a researcher to know very clearly about the various data collection methods

and tools. The researcher then has to choose the method suitable for the specific

purpose or research. In this assignment report, I have discussed briefly about

various research methods, their advantages and disadvantages as well as few

design issues what I came to learn from various sources.

Page 3: Choosing a Research Method

Table of Contents

1. Abstract ___ page 1

2. Choice of Research Method ___ page 5

3. Questionnaires ___ page 7

4. Interviews ___ page 11

5. Focused Group Discussions ___ page 17

6. Observation ___ page 21

7. Documentary Analysis ___ page 25

8. Experiments ___ page 27

9. Bibliography ___ page 31

Page 4: Choosing a Research Method

Choice of Research Method

Generally, research is understood to follow a certain structural process.

Though step order may vary depending on the subject matter and researcher,

choosing a suitable method for collecting the relevant data is of utmost

importance in order to accomplish a research successfully.

In practice six main research methods tend to be used:

1. Questionnaires

2. Interviews

3. Focused group discussions

4. Observation

5. Documentary analysis

6. Experiments

The following discussion looks at the main pros and cons of using these

methods, some of the main design features of each and some issues relating to

their administration.

Page 5: Choosing a Research Method

Questionnaires

A questionnaire is a research instrument consisting of a series of questions

and other prompts for the purpose of gathering information from respondents.

Although they are often designed for statistical analysis of the responses, this is

not always the case. The questionnaire was invented by Sir Francis Galton.

These are sets of questions and attitudinal statements administered to

individuals for self-completion. They can be posted to respondents, sent by email

or filled in by the researcher during face-to-face enquiries. They can be:

wholly closed-ended, with every question having a fixed range of

alternative responses, or

open-ended, with very broad questions designed to elicit the sample's own

views rather than their responses to a re-specified range of answers, or

a mixture of the two.

Benefits and strengths

1. Questionnaires are a useful means of getting data from a relatively large

number of people or from a representative sample of that population.

2. Therefore they are very efficient in terms of the researcher's use of time

and effort. That is, you can distribute hundreds of them to students,

colleagues or clients and get them completed and returned in the time it

takes to set up and do half a dozen interviews. Questionnaires are also,

usually, quicker to code and analyze that semi-structured or unstructured

interviews.

3. Questionnaires are a means of standardizing the data collection process,

i.e. everyone who completes it is being asked exactly the same questions in

Page 6: Choosing a Research Method

Choosing a Research Method

Questionnaires

8

the same sequence [but of course this does not mean that they will choose

to answer every question].

4. The self-completion questionnaire may provide the respondent with time

to reflect on the questions before answering them [especially if they can

complete them in their own time].

5. Respondents may feel that they can say what they really think if the

questionnaire can be completed in privacy and anonymously [especially if

the researcher is known to them or might be thought to have a vested

interest in their answers].

Problems and criticisms

1. There are hidden costs in the use of questionnaires. It may be a quicker

method of collecting data and the format may facilitate data analyzed but

the design of a good questionnaire with clear instructions and

unambiguous questions can take a long time. Semi-structured and

unstructured interview schedules can be developed more quickly because

the researcher gets immediate feedback from the respondents if they do

not understand the questions.

2. You may not always know that your carefully constructed questionnaire is

not asking the 'right' questions until you start analyzing the data, i.e. when

it is too late to do anything about it. If they have sufficient time researchers

try to get round this problem by piloting the questionnaire with a small

group drawn from the population they are interested in before they make

the final adjustments to it. But this is time consuming and practitioners

doing research on a part-time basis do not always feel that they can spare

the time to do this. Pilot your questionnaire if you can. At the very least

send it to some colleagues and your tutor for comment.

Page 7: Choosing a Research Method

Choosing a Research Method

Questionnaires

9

3. Once you have sent out the questionnaire you have little control over the

situation. Suppose only a few completed ones are returned, i.e. what

researchers refer to as a low response rate. What can you do? You can try

several things to increase the likelihood of a good response rate:

i. Get a 'captive' population or sample to complete it, i.e. ask students to

complete it during a seminar or clients to complete it in your

presence;

ii. Explain at the beginning of the questionnaire (or in an accompanying

letter if it is a postal questionnaire) why their responses are important

and what use will be made of the data.

iii. Make the questionnaire as user friendly and attractive as possible by

using different font sizes, color printing, attractive layout and provide

as many opportunities as possible to tick boxes or ring code numbers

so that it can be filled in as quickly as possible.

iv. Distribute reminders perhaps with a second copy of the questionnaire.

Avoid a lot of open-ended questions. As we have seen, one of the

advantages of a self-completion questionnaire is that it pre-codes people's

responses into a standardized set of answers to facilitate data analysis. Why lose

that advantage? If you feel the need to have a lot of open-ended questions

perhaps you have opted for the wrong method.

Design Issues

Some of the key issues when designing a questionnaire are as given below:

1. Avoid imprecise and ambiguous questions.

The ambiguity and imprecision will still be there when you try to analyze the

data.

Page 8: Choosing a Research Method

Choosing a Research Method

Questionnaires

10

2. Avoid trying to get too much information from one question

Two points here. First, if you have several sentences in your question simplify

it. Second, avoid trying to get information on more than one dimension in the

same question. For example, if you want to ask teaches (a) what they do in a

particular classroom situation and (b) how the students react, then it might be

possible to do this in one question but it would be better to do it in two separate

questions and cross-reference (or cross-tabulate) their answers.

3. Avoid leading questions, e.g. ‘do you agree that…?’ Or even ‘Most people

agree that…?’

Research shows that many people are inclined to agree with leading questions

regardless of the content (or their actual views). Its known as ‘acquiescence

response set’.

4. If you are using questions which include a list of statements on which the

respondent is asked to agree or disagree then make sure that there is a

good mix of positive and negative statements.

This also reduces the risk of acquiescence response set.

5. Do not ask too many open-ended questions.

They take too much time to answer and to analyze. Open questions are more

difficult to code, analyze and interpret. However, one or two of these can be a

useful means of allowing the respondent to express their views in their own words

and to provide you with more detail and depth, but as noted earlier, if you want to

ask a lot of open-ended questions consider using an interview schedule instead.

Use open-ended questions when not enough is known in advance to allow

you to decide on a list of response categories for a closed-ended question. But,

better still, pilot these open-ended questions with a small sample and use the

results to develop your response categories before you do the survey.

Page 9: Choosing a Research Method

Interviews

An interview is a conversation between two or more people (the

interviewer and the interviewee) where questions are asked by the interviewer to

obtain information from the interviewee.

The degree to which it is structured can vary greatly. It can be highly

structured (and then usually it looks like a questionnaire or checklist but

administered by the researcher rather than self-completed). It can be semi-

structured, mainly comprising a set of open-ended questions but often with

follow-up probes and prompts; or it can be relatively unstructured - a list of

themes or topics or headings which can be adjusted to each individual

interviewee.

Benefits and strengths

1. Provides an opportunity for the interviewee to give a more detailed

response than in a questionnaire.

2. The data will usually be richer and more full of contextual information than

the data provided by a questionnaire.

3. The data will be collected in a natural setting. For example, interviewing a

lecturer immediately after a seminar will probably generate more data than

a questionnaire which could be completed at any time. There is scope for

the interviewer to cross-reference to the seminar, particularly if he or she

also observed it [an example of triangulation at work].

4. An interview is a particularly useful tool if you are trying to understand the

experiences and actions of each individual respondent. The more unique

each respondent's experiences are the more important it is to use a

research technique which acknowledges and elicits that uniqueness.

Page 10: Choosing a Research Method

Choosing a Research Method

Interviews 12

5. An interview is also a useful method if the respondents are not very literate

or not used to expressing their views on paper.

6. It provides an opportunity to probe respondents' views in ways that might

be difficult to plan for in advance.

7. It provides the researcher with an opportunity to also observe and record

the non-verbal behavior of the respondent as well.

Problems and criticisms

1. The interview is not necessarily a cost effective method for a researcher

who does not have any help in collecting data from a relatively large

sample or population of respondents.

2. The full transcription of interviews takes a lot of time. A one-hour tape

recorded interview takes about 8-10 hours to transcribe.

3. If you want to it is possible to quantify some of the answers (e.g. the

number of respondents who answered this question positively or

negatively) but other methods might be more appropriate for this purpose.

4. Good interviewing requires expertise and experience. It is not an easy

option for the inexperienced researcher. It requires, for example, good

listening skills; body language that encourages the interviewee to relax and

talk; a capacity to ask questions, perhaps take notes and yet maintain eye

contact; an ability to prompt people who are not very responsive; knowing

just how long to allow a silence to continue before intervening; an ability to

probe sensitive areas and issues; being able to 'think on your feet' and be

flexible in your questioning e.g. to recognize that someone has just

answered 3 of your questions in one response and therefore being able to

instantly adjust your schedule of questions accordingly to being able to

decide on-the-spot whether an interviewee is leading you down a blind

alley and needs to be brought back to the sequence of questions on your

Page 11: Choosing a Research Method

Choosing a Research Method

Interviews 13

schedule or is raising interesting issues and should be encouraged to

continue. In other words, interviewing is not a soft option. It requires

planning (including planning for unexpected contingencies) and it requires

skill.

5. There is always a high risk with interviews that the researcher will unduly

influence the responses of the interviewee: through the way the questions

are asked, the researcher's body language and manner, the interviewer's

position of authority in the interviewee's institution or vice versa (a

potential problem in practitioner research).

Design Issues

Following is a brief summary of some of the key issues when designing an

interview schedule.

1. Tape record it or take notes

Using a tape recorder may mean that you can concentrate on the interview

process and maintain eye contact with the interviewee. Plus you will have a

complete record of the interview. But some interviewees get nervous when being

recorded and some think you are not interested in what they have to say if you

don’t take notes.

Some researchers do both. But remember that taped interviews take a long time

to transcribe, your notes may be quicker (depends on the quality of your writing).

2. If you have sensitive questions that might affect the rapport between you

and the interviewee leave them until late in the interview.

Some researchers, for example, ask personal questions at the end of the

interview for the same reason.

3. Always start by explaining the purpose of the interview and what you will

do with the information.

Page 12: Choosing a Research Method

Choosing a Research Method

Interviews 14

You should use this introduction to also reassure the interviewee about

confidentiality and ask them if they mind being recorded, etc.

4. Avoid very long questions.

You cannot expect the interviewee to remember what you asked them if you

speak in paragraphs.

5. Avoid multiple questions, e.g. How many students are taking the course

this year, are they mostly school leavers or have you also got some

mature students and how do the two groups compare in performance?

Again, interviewees will not usually remember all of the elements of the

multiple questions. Ask separate shorter questions.

6. Plan some prompts in case the interviewee looks blank when you ask a

question or asks you to clarify the question or just says ‘What do you

mean?’

There is an art to designing and asking good prompts; they help the

respondent to understand the question without ‘feeding’ him or her the answers

you want. Knowing when to ask a prompt is critical. How long can you cope with

silence before jumping in with a prompt. Jump in too quickly and you annoy the

person who was thinking about their answer. Allow the silence to go on too long

and the respondent begins to lose confidence.

7. Plan some probes.

The probe is designed to get the interviewee to expand on his or her initial

response to a question. Some experienced interviewers rely on a raised eyebrow

or encouraging noises such as ‘mmmhm?’ or ‘uh huh?’ But most of us usually ask

the respondents if they want to add anything or ‘Could you tell me a bit more

about …….?’ Or ‘I’m interested in what you said about…..’; ‘Why/what do you think

that……?’

Page 13: Choosing a Research Method

Choosing a Research Method

Interviews 15

Administrative issues

1. Once you have explained what the interview is all about and why you are

interviewing the respondent try one or two 'warm-up' questions to

establish a good rapport.

2. Listen, listen, listen! You can lose rapport if you ask someone a question

which they have already answered when replying to an earlier question.

Also you may want to cross-reference the answer to one question again the

answers given to earlier questions, e.g. 'Earlier in the interview you said

that...'

3. Do not be afraid to leave out a question if you think it is not relevant to a

particular interviewee.

4. If your interviewee is talkative and giving you a lot of useful information do

not hesitate to change the sequence of your questions if you think it would

be appropriate (e.g. if the respondent has mentioned something that you

were not originally intending to ask about until later in the interview) but

remember to go back to the questions you have temporarily skipped when

the time is right.

5. Think about how you are going to end the interview. Don't let it simply 'trail

off' because you have both run out of time or things to say.

6. Even when you have switched off the tape recorder or put away your

notebook you need to stay 'switched on'. A common phenomenon in

interviewing is that people often give you very interesting and useful

information once the formal interview is over. Write it down as soon as you

can but don't foreclose the situation by either asking if you can switch on

your recorder again or by saying 'I'm afraid that I'm in a bit of a hurry!'

Page 14: Choosing a Research Method

Focused group discussions

A focus group is a form of qualitative research in which a group of people

are asked about their attitude towards a product, service, concept, advertisement,

idea, or packaging. Questions are asked in an interactive group setting where

participants are free to talk with other group members.

These are widely used in market research. The method is also sometimes

used in health promotion research and social work research. As an approach it

tends to be used in combination with other research techniques.

Benefits and strengths

1. Group discussion produces data and insights that would be less accessible

without interaction found in a group setting -- listening to others’

verbalized experiences stimulates memories, ideas, and experiences in

participants.

2. A good medium for exploring a group of people's views without imposing

your own agenda on them too strongly.

3. Provides opportunities to explore the thinking behind the kinds of

responses which might have been given to a questionnaire but without

opting for the more expensive option of follow-up one-to-one interviews.

4. In the early stages of a project it can be a useful means of identifying issues

or areas of interest that could be followed-up using more quantitative

research methods.

5. It provides the researcher with an opportunity to cross-check responses,

e.g. 'Does everyone agree with that?'...Does that reflect your experience as

well? ..etc.

Page 15: Choosing a Research Method

Choosing a Research Method

Focused Group Discussions 18

Problems and criticisms

1. The researcher has less control over a group than a one-on-one interview,

and thus time can be lost on issues irrelevant to the topic.

2. Difficult to follow-up the views of individuals during the group discussion,

especially on topics which may be sensitive.

3. A fundamental difficulty with focus groups (and other forms of qualitative

research) is the issue of observer dependency: the results obtained are

influenced by the researcher, raising questions of validity.

4. Moreover, the number of members of a focus group is not large enough to

be a representative sample of a population; thus, the data obtained from

the groups is not necessarily representative of the whole population, unlike

in opinion polls.

5. This method can be heavily affected by the dynamics of the group,

especially if some members of the group are senior to others. One or two

people can easily dominate the proceedings if they have clear views and

are articulate. It is possible as a facilitator to counter this tendency but it

takes experience and self-confidence.

Design Issues

1. Keep the size of the group fairly small (8-12).

2. An hour is probably the minimum time to allow for the discussion to

develop. It takes a while for the group to settle and for the right

atmosphere to develop.

3. Draw up a list of topics or themes to be covered. These should loosely

guide the discussion. You want a balance between allowing the discussion

to develop its own momentum and making sure your issues are addressed.

Page 16: Choosing a Research Method

Choosing a Research Method

Focused Group Discussions 19

4. Introduce yourself (if they do not already know you) and explain the

purpose of the session, how long it will take and what you will do with the

information.

5. Get them to briefly introduce themselves (if they do not know each other).

6. Decide how you will record the discussion. Will you tape it? Take notes?

Get a colleague to take notes?

Good chairing skills are essential (a) to ensure that the discussion is dominated

by one or two and (b) to find out if there is a consensus within the group on a

particular issue or theme and (c) to judge when to move on to the next theme or

issue.

Page 17: Choosing a Research Method

Observation

Observation is either an activity of a living being (such as a human),

consisting of receiving knowledge of the outside world through the senses, or the

recording of data using scientific instruments.

This tends to be classified along three dimensions. Firstly, whether the

observer is a participant in what he or she is observing or is detached. Secondly,

on whether or not those who are being observed are aware of this. Thirdly, the

extent to which the observation is structured. The degree of structure can range

from the observer taking notes or keeping a diary at one end of the spectrum to

using coding schemes, checklists and category systems at the other.

The scientific method requires observations of nature to formulate and test

hypotheses. It consists of these steps:

1. Asking a question about a natural phenomenon

2. Making observations of the phenomenon

3. Hypothesizing an explanation for the phenomenon

4. Predicting a logical consequence of the hypothesis

5. Testing the prediction in a controlled experiment, a natural experiment, an

observational study, or a field experiment

6. Creating a conclusion with data gathered in the experiment

Observation plays a role in the second and fifth steps. However the need

for reproducibility requires that observations by different observers be

comparable.

Page 18: Choosing a Research Method

Choosing a Research Method

Observation 22

Benefits and strengths

1. The biggest advantage of collecting data through direct observation is that

it is one of the most direct research techniques. You are not asking people

what they would do or think you are watching what they do and listening to

what they say.

2. Used in combination with questionnaires or interviews, observation can

therefore provide useful insights into the extent to which there is a

correspondence or discrepancy between what people say and what they

actually do. As such observation makes a useful contribution to the process

of triangulating evidence.

3. The researcher observing his or her own practice has an advantage because

the situation to be observed is familiar (which in turn reduces preparation

time).

4. It allows the researcher to describe the full complexity of the situation,

including the dynamics of the group being observed.

Problems and criticisms

1. It is very time consuming. How many times do you need to observe a

situation or a group before you can be sure that you are describing with

confidence what is really happening.

2. When the observer is also a practitioner observing his or her own practice

then there is a potential problem of how to ensure that the observer can

detach himself or herself sufficiently.

3. It is difficult to estimate the effect of the observer on the situation being

observed.

4. The analysis of observational data is time consuming and often open to

very different interpretations.

Page 19: Choosing a Research Method

Choosing a Research Method

Observation 23

Design Issues

Here is a brief summary of some of the key issues when designing an

observation schedule:

1. The first key question is: 'Are you going to be a participant or detached

observer? What are the advantages to either role in this particular setting?

2. Linked to that question is: 'Are you going to observe your own practice or

the practice of your colleagues and are you going to observe your students

or a colleague's?'

3. Are you going to tell the group that you will be observing them?

4. Do you want a narrative description of what is happening or do you want to

design a specific observation schedule. The former approach is often

described as 'naturalistic' and used by social anthropologists and some

sociologists. Some form of observation schedule is more often used by

psychologists and educational researchers. Look at the different kinds of

schedule in the recommended reading before making up your mind on this.

5. What specifically do you want to observe:

a. Do you need to describe the setting? [the layout of the room, where

people are sitting, the equipment used, etc)

b. Do you need to describe the people involved?

c. Do you need to describe every event and activity over a specific time

period or just a sample of events, or select the ones you are

interested in?

d. Are you just interested in the interactions between the tutor and the

students or are you also interested in the interactions between the

students?

e. Will it be useful to time each event or activity? What will this tell

you? How will you use this data?

Page 20: Choosing a Research Method

Choosing a Research Method

Observation 24

6. When will you record your observations: at the time or immediately

afterwards or much later? [This can be a real problem when you are a

participant observer].

7. How will you record your observations? Pen and paper, video recording, audio

recording?

8. If you are using an observation schedule can you use a coding system to make

the data collection process easier?

9. Is observation the only source of data for your research or will it be

complemented by interviews, questionnaires or other methods? If the latter

how will you relate one kind of data to another?

Page 21: Choosing a Research Method

Documentary Analysis

Sources of documentary research include historical documents such as

laws, declarations, statutes and people’s accounts of events and periods. Also,

reports based on official statistics would be covered, as well as governmental

records, mass media, novels, plays, drawings, and personal documents such as

dairies and biographies.

Documentary research method has had little attention compared to other

methods because of the dominance of positivism and empiricism so that statistics

and quantification are popular forms of data collection and analysis.

Benefits and strengths

1. Unlike most research instruments and data sources, documents are not so

prone to being influenced by the researcher.

2. Documents enable the researcher to investigate the background and

context of the situation and the specific problem they are interested in.

3. A useful means of evaluating the extent to which the rhetoric (or the policy)

is actually put into practice [but of course you also need other means and

other sources for evaluating practice].

4. Documentary analysis is a useful means of analyzing the 'official' view and

accessing the 'official' record of events, decisions and plans.

5. Some documents [e.g. student assessments] can provide a measure of the

impact of changes introduced during the Action Research process.

Page 22: Choosing a Research Method

Choosing a Research Method

Documentary Analysis 26

Problems and criticisms

1. Documentary research is regarded as being not clear-cut, not having a

method and nothing on how a researcher uses it. Yet these criticisms are

misplaced

2. Research bias may be restricted in one sense because the researcher has

not produced the documents [in the same way as he or she has devised the

questionnaire or interview schedule] but bias can still be present in the

selection of the documents.

3. Documentary analysis - if it is to be systematic - is time consuming. Analysis

can range from reading and commenting on it to undertaking a full-scale

content analysis or from quoting a mission statement to undertaking a

statistical analysis of student assessment results for the last five years.

4. There is little guidance available from experienced researchers on how to

analyze some kinds of documents [e.g. students' diaries or reflective

journals].

5. Documents require a critical reading similar to the skills employed by the

historian when analyzing primary sources. Each document was written for a

particular purpose and within a specific context which may not necessarily

be obvious from the document itself. Documents have to be interpreted as

well as read and this calls for expertise and experience.

Design Issues

Historians, political scientists and economists may design their research

around the analysis of documents, records and statistics but for most action

researchers investigating their own and others' practice documentary analysis is

mainly a supplementary method employed for collecting background information

and for triangulating with the main methods used.

Page 23: Choosing a Research Method

Experiments

The key feature of any experiment is that the researcher deliberately

controls and manipulates the conditions under which the effects of a change or

intervention can be measured. Cohen, Manion & Morrison (2000) provide a useful

brief description of experimental research in the natural and physical sciences:

Imagine that we have been transported to a laboratory to investigate the

properties of a new wonder fertilizer that farmers could use on their cereal crops,

let us say wheat. The scientist would take the bag of wheat seed and randomly

split it into two equal parts. One part would be the grain under normal existing

conditions - controlled and measured amount s of soil, warmth, water and light

and no other factors. This would be called the control group. The other part would

be grown under the same conditions - the same controlled and measured

amounts of soil, warmth and light as the control group but, additionally, the new

wonder fertilizer. Then, four months later, the two groups are examined and their

growth measured. The control group has grown half a meter and each ear of

wheat is in place but the seeds are small. The experimental group, by contrast has

grown half a meter as well but has significantly more seeds on each ear, the seeds

are larger, fuller and more robust."

The key features of the experiment are:

An experimental group and a control group;

A random allocation to each group to eliminate the possibility that any

variables not thought to be crucial to the experiment might have any

unintended effects;

The identification of key variables that will have some effect;

Page 24: Choosing a Research Method

Choosing a Research Method

Experiments 28

The control of these key variables

The application of the special treatment to the experimental group but not

the control group;

Measurement of the effect of the treatment and comparison of the

outcomes for the two groups.

Will this classical experimental design still work when the subjects of the

experiment are people rather than wheat seeds?

The experiment is still the norm in medical research and is widely used in all

forms of psychology, research into health care and, to a lesser extent, research

into social care. One particular form of the experimental design, the randomized

controlled trial (through which, for instance, new drugs and forms of medical

treatment are tested) is still generally regarded in those disciplines as the 'gold

standard' of research. It is used to test new drugs and forms of medical treatment

and to evaluate the effectiveness of new interventions in health and social care

[Gomm & Davies (2000) p.xi].

In each case the experiment is designed in such a way that it reduces the

likelihood that the prior knowledge of the subjects, the practitioners and the

researchers taking part in the trial might unduly influence the results of the

experiment.

However, in many social situations it is simply not practical (or sensible) to try

to control all of the possible variables that might influence the outcomes of a

specific change or intervention. Indeed, in some instances it would also be

unethical to use a controlled experiment if, for instance, the subjects were not in a

position to give their informed consent to participation in the experiment or if

participation meant that they might suffer or be treated unfairly or if the

Page 25: Choosing a Research Method

Choosing a Research Method

Experiments 29

experiment required them to do something illegal or immoral. Also, in the real

world, it may not be possible to assign people randomly to either the

experimental or the control group.

In such circumstances some researchers have introduced the idea of a quasi-

experiment. Perhaps the most common kinds of quasi-experiment employed in

social research are: Where the researchers collect data that enable them to

compare the same subjects before-and-after an intervention or change has been

introduced.

Where the researchers establish a pilot or study group (the subjects of their

research) who experience the change or intervention and a reference or

comparator group from whom data are collected using the same research

instruments even though they are not experiencing the change or intervention.

Therefore the quasi-experiment retains the element of comparison which is so

central to the experimental research design but subjects are seldom allocated to

their groups and, if they are, this is rarely done at ransom. In practice, in most

quasi-experiments the researcher does not have any control at all over the so-

called 'control group' (or reference or comparator group).

Ultimately the central question for any researcher opting for a quasi-

experimental design will be: 'Am I comparing like with like?' If the answer is: 'I

believe so' then the follow-up question will inevitably be: 'How do I know?’

There are a variety of different kinds of quasi-experimental research design

and each has its own advantages and disadvantages. Rather than outline all of

them here it is recommended that you read the relevant section in Colin Robson's

book, Real World Research (1993 Oxford) Chapter Four: Experimental Design

Outside the Laboratory, especially pp.96-108.

Page 26: Choosing a Research Method

Bibliography

BLAXTER, L., HUGHES, C. & TIGHT, M. (1996), How to Research,

Buckingham, Open University Press, pp. 153-156

BURGESS, R. G. (1984), In the field: an introduction to field research,

London, Allen & Unwin (see especially the chapter on 'Interviews as

Conversations'.

COHEN, L., MANION, L. & MORRISON, K (2002) Research Methods in

Education, London, RoutledgeFalmer

HAKIM, C. (1982), Secondary analysis in Social Research: a guide to data

sources and methods with examples, London, Allen & Unwin

HOINVILLE, G. et al (1977), Survey Research Practice, London, Heinemann

LINDLOF, T. R., & TAYLOR, B. C. (2002). Qualitative Communication

Research Methods, 2nd Edition. Thousand Oaks, CA: Sage.

LOFLAND, J. & LOFLAND, L. (1984), Analysing Social Settings: A Guide to

Qualitative Observation and Analysis, Belmont, Wadsworth.

MARSHALL, CATHERINE AND GRETCHEN B. ROSSMAN. 1999. Designing

Qualitative Research. 3rd Ed. London: Sage Publications, p. 115

POWNEY, J. & WATTS, M. (1987), Interviewing in Educational Research,

London, Routledge & Kegan Paul.

ROBSON, C. (1993), Real World Research, Oxford, Blackwell

WEBB, E.J. et al (1966), Unobtrusive Measures: non-reactive research in the

social sciences, Chicago, Rand McNally.