RESEARCH METHODS – new curriculum

Research is the process whereby data is systematically gathered, analysed and

interpreted in order to increase our understanding of a particular phenomenon.
Sociologists obtain their knowledge of human behaviour through research, which
results in a body of information which helps to move beyond guesswork and common
sense in understanding society.

 QUANTITATIVE AND QUALITATIVE RESEARCH

Research can either be quantitative or qualitative.

 Characteristics of quantitative research

 Quantitative research is generally a numerical approach to understanding
human behaviour. It produces quantitative data i.e. data in the form of numbers
and statistics, expresses information numerically and captures the ‘who, what, when
and where’ of behaviour in one of three ways: a raw number (e.g. total number of
people who live in a society); a percentage (e.g. the number of people per 100, in
a population) and a rate (e.g. the number of people per 1000).
 Quantitative methods include surveys, structured interviews, observations,
questionnaires and reviews of records, documents for numeric information.
 It is primarily a deductive process used to test pre-specified concepts,
constructs and hypothesis that make up a theory.
 It is more objective – provides observed effects of a program on a problem
or condition.
 Less in-depth but more coverage of information across a large number of
cases.
 Employs fixed response options e.g. closed ended questionnaires.
 Statistical tests are used for analysis. For example, the data can be analysed
using computer packages like the Statistical Package for the Social Sciences
(SPSS). It can also be put into a graph or chart.
 Can be valid and reliable – largely depends on the measurement device or
instrument used.
 More time used on planning while less time is used during analysis of results.
 More generalizable mainly because it is normally conducted using large numbers
of people e.g. surveys.
 Positivists (e.g. Comte) prefer quantitative to qualitative research.

Strengths

 The ability to express relationships statistically can be useful if the researcher does
not need to explore the reasons for people’s behaviour – e.g. if they simply need to
compare the number of murders committed each year in different societies.

1|Page
  Allows us to ‘summarise vast sources of information and make comparisons across
categories and over time’ – Kruger, 2003
 Quantitative research is more reliable because it is easier to replicate the study.
Standardized question do not change.
 Quantitative data also makes it easier for the researcher to remain objective. They
not need to have a close personal involvement with the subjects of the study, so
their personal biases are less likely to intrude into the data collection process.

Limitations

 Quantitative data only captures a relatively narrow range of information i.e. it only
captures the ‘who, what, when and where’ of people’s behaviour – day (1998). It’s
difficult to get the real meaning of an issue by looking at numbers.
 ‘Issues are only measured if they are known prior to the beginning of the research’
– McCullough 1988.
 ‘Quantophrenia’ – quantification for its own sake, regardless of whether it tells us
anything useful about the behaviour being quantified.

 Characteristic of qualitative research

 Qualitative research involves interpretive description (words) rather than
statistics (numbers) to analyse underlying meanings and patterns of social
relationships. It gives qualitative data i.e. data in the form of words.
 Qualitative methods include focus groups, participant observations, content
analysis, historical comparison, in-depth interviews and reviews of
documents for types of themes.
 Deals with data which aims to get personal views and opinions across.
 It looks into things in depth and quality is more important than quantity.
 Primarily inductive process is used to formulate theory or hypothesis.
 More subjective – describes a problem or condition from the point of view
of those experiencing it.
 Data that represent nominal scales such as gender, socioeconomic status and
religious preference are usually considered to be qualitative data.
 Makes use of unstructured or semi structured response options
 No statistical tests are used for analysis. The data cannot be easily turned into
a list of numbers or a graphs. Methods of analysing qualitative data include thematic
analysis, content analysis etc.
 Can be valid or reliable – largely depends on skill and rigor of the researcher.
 Less time is used on planning but more time is used on the data analysis
phase.
 Less generalizable since it is normally conducted on small groups.
 It gives a detailed picture of what people do, think, and feel i.e. it’s valid.
 It is subjective, involving opinions, meanings and interpretations.
 Interpretivists prefer qualitative data because it is richer than quantitative data.

2|Page
Strengths

 Allows the researcher to capture the complex reasons for behaviour.

Limitations

 Qualitative research focus on the intensive study of relatively small groups, which
may limit the opportunity for applying the data more widely – such data may not be
representative
 It is difficult to compare qualitative data across time and location because no two
groups will ever be qualitatively the same.
 The depth and detail of the data also makes such research difficult to replicate,
which means its reliability is generally lower than that of quantitative research.

 STEPS IN SCIENTIFIC RESEARCH

 The scientific method

It is a method of investigation in which a social problem is first identified and

observations, experiments, or other relevant data are then used to construct or test
hypothesis that purport to solve it.

 The steps that all scientists follow include:

 Identification of topic, phenomena or social problem to be studied

This involves identification of a social problem or phenomenon to be investigated -

anything in the world that catches our attention and which we want to describe, analyse and
/ or explain e.g. a desire to research the area of “crime”, “family life” or “education”.

 Literature review

It involves a preliminary investigation of studies that have already been done and
published by other researchers on the topic.

What are the advantages of conducting a literature review in sociological research?

-To find out what is already known about the topic and where more research is
needed (research gap)
-To familiarise the researcher with research instruments used in previous studies
as well as their strengths and weaknesses.
-To avoid duplication of research studies which would waste finances, effort and
time.
-Familiarise the researcher with conceptual and theoretical frameworks relevant to
the research topic.
-It shares with the reader the results of other studies that are closely related to the
study being reported (Fraenkel & Wallen, 1990).

3|Page
 - It relates a study to the larger, ongoing dialogue in the literature about a topic,
filling in gaps and extending prior studies (Marshall & Rossman, 1989).
- It provides a framework for establishing the importance of the study, as well as a
benchmark for comparing the results of a study with other findings.
- It “frames” the problem earlier identified.

 Observation and generation of ideas

This refers to the process of narrowing the research topic into something more specific and
manageable. These can take the form of a “research question or problem”. For example, why
do people steal? Does the availability of divorce lead to family breakdown? Etc.

It also involves formulating:

- Research questions
- Research objectives

 Development of testable hypothesis

A hypothesis can be defined as a general question or statement (educated

guess) that suggests a possible (and therefore testable) relationship between
two or more variables which then has to be tested by conducting empirical
research and collecting evidence to prove or disprove the orginal idea.
For example, a research question such as “Why do people steal?” It is too general for research
purposes because it doesn’t specify a relationship between “people” and “stealing” that can be
tested. There is need to create a hypothesis (or series of related hypotheses) that turn our
research problem into something we can test (by collecting and analysing evidence).

An example of a hypothesis
“Poverty makes people steal”.

 Design the research

Indicate the methodological steps you will take to answer every question or to test
every hypothesis illustrated in the Questions/Hypotheses section.
This serves as the roadmap for the entire study specifying

- The type of research design to be employed e.g. quantitative or qualitative

research.
- Participants
- How
- When and where data will be collected and
- The content of the program.
- Identify the population and draw a sample
A population is the group a researcher is interested in learning about. The study
cannot possibly include the entire population, so smaller sample group is used to
represent the population.
A sample is a subset of the population which should be representative of the
population in various social characteristics e.g. gender, age etc.

4|Page
 Random sampling techniques where every unit in the population has an equal chance
of being included in the sample have the best chance of reflecting the views and
behaviours of the entire population.
- Operationalise variables
This involves defining the elements/variables in the hypothesis carefully and unambiguously
e.g. “Using this definition people are considered poor when….” This means deciding
exactly how each variable of interest will be measured.
In survey research this means deciding on the exact wording of the question or
questions used to measure each variable, a listing of all possible response to closed
ended questions and a decision as to how to calculate variables using multiple
indicators.

 Selection and development of a suitable research instrument

There are a wide variety of quantitative and qualitative methods. Each method
comes with its own strengths and weaknesses e.g. interviews, participant
observations, experiments etc.
 Pilot study
It is a small scale study done before the actual study.
What are the merits and demerits of carrying out a pilot study?
- It gives an opportunity to try out the logic or designed of the planned study before
the main expenditure is incurred and it is too late to change.
- To test if the research instrument is feasible e.g. the instrument may be too long.
- To test how long the research will take
- To train research (novice) assistants
- To see if there are any ambiguous words that may confuse participants
- It is routine to pilot questionnaires with a small number of respondents who are
encouraged to explain any difficulties they have in interpreting or answering
specific questions.
However
- It wastes a lot of time and resources.

 The careful/systematic observation and data collection.

Data collection must be systematic (planned and organised carefully).

 Data analysis and interpretation

Once data has been collected it needs to be analysed and interpreted in relation to
original hypothesis. This involves checking to ensure that sufficient data has
been collected, checking that the sample used has remained representative,
making decisions about whether or not to include or discard data that appears
irrelevant etc.
 Testing the hypothesis

5|Page
Once the data has been analysed a decision can then be made about whether
or not the tested hypothesis has either been:

- Shown to be untrue (falsified) or

- Shown to be true (confirmed).

 Discussion of results

 Formulating a theory

A theory consists of a hypothesis that has already been tested against the evidence
and shown to be confirmed.

 Reporting the results/findings and conclusions

The researcher reports the findings and conclusions to the larger scientific
community, in the form of reports, presentations and publications in peer reviewed
journals, which must be open to further retesting by others who may be interested.

 RESEARCH TOOLS

 Questionnaires

Questionnaires are sets of written questions handed out or mailed (electronically or

postal) to respondents, who are then required to fill in the answers and send them
back.

Questionnaires can be used to obtain both qualitative and quantitative data, depending
on the type of questions.

Types of questions

 Closed ended questions

Standardised multiple choice answers which obtain quantitative data. For example:

How long have you lived in your current house?

A. Under 1 year D. 11 – 20 years

B. 1 – 5 years E. 21 – 30 years
C. 6 – 10 years F. 31 years or more

 Open ended questions

These give insight into meanings and motives and give qualitative data.

Therefore, the reliability and validity of a questionnaire depends on how it is

designed.

Designing a good questionnaire

Questionnaires should:

6|Page
 Use clear simple questions which are easy to understand (avoid use of jargon)
 Give clear instructions and make it easy for the respondent.
 Have a nice clear layout that does not intimidate people.
 Give a range of options on multiple choice, closed ended questions.
 Measure what you want to measure (must have high levels of validity).
 Not be biased – the respondent is not led to a particular viewpoint or answer.
 Be concise – as short as possible, since people usually cannot be bothered to spend
a long time completing questionnaires.
 Have indicators which are correct – it must ask the right questions, which unearth
exactly the information wanted, - in sociological terms ‘ the concepts must be well
operationalised’

Questionnaires should not:

 Ask embarrassing, threatening or complex questions.

 Ask two questions instead of one.
 Be too long.
 Use sociological terms that no one understands.
 Lead the respondent to answer a question in a particular way.

Topics studied

Questionnaires can be used to investigate the Knowledge, Attitudes and Practices

(KAP) on topics such as:

 TV viewing habits
 Purchasing habits
 Voting behaviour
 Experiences of crime

Strengths of questionnaires

 Positivists claim that this method allows them to remain objective since they have
almost no contact with respondents and thus cannot influence the research process.
 They are used to reach a large number of people (survey research) because the
forms can just be handed out. Widely used in survey work. If the sampling has been
correct then the questionnaire should produce questions that are generalizable to the
whole population.
 Less time consuming and less expensive – quick and cheap to conduct compared to
other methods as large amounts of information can be collected from a large number
of people
 Can be analysed more ‘scientifically’ and objectively than other forms of research.
 High reliability since the questions are standardised
 Easy to analyse with computer programs e.g. Statistical Packages for Social Sciences
(SPSS) and graphs and charts.
 A practical method of gathering data
 No interviewer to affect people’s responses.

7|Page
 The researcher can spot patterns in the answers and make comparisons between
different groups

Limitations of questionnaires

 Analysing the results on a large scale might be very time consuming and could be
out of date when the results are ready
 Anti-positivists claim questionnaires lacks validity - Statistics give no indication of
how the person feels and could not be accurate because it could not be how the
person truly feels.
 There is a level of researcher imposition – when developing the questionnaire, the
researcher is making his/her own decisions and assumptions as to what is and is not
important, therefore they may be missing something that is of importance.
 Respondents may not tell the truth.
 Questions may be misleading or mean different things to different people – they may
not measure what you actually want to measure (less valid).
 Postal questionnaires have low response rate and if it’s too low it will not give a
representative sample.
 Open ended questions make it hard to quantify the data into nice neat numbers.
 No one can explain the questions if the respondent does not understand them.
 Questionnaires do not allow the researcher to have a deep understanding of the
people he/she is studying.
 They are not suitable for sophisticated detail e.g. a person’s experiences.

 INTERVIEWS

It is a conversation between the researcher and an interviewee where the researcher

asks a set of questions and records the answers. Sociologists generally use interviews
if the subject of inquiry is complex, where a questionnaire would not allow the
researcher to explore the issues adequately.

An interviewer should first create a friendly atmosphere (good rapport).

Types of interviews

a) Structured interviews

At their most structured, they can be tightly organised, with the interviewer simply
reading out questions from a prepared questionnaire. The interview consists of
predetermined, pre-set question which are administered to respondents, although the
researcher may be allowed to deviate minimally.

Advantages – (are similar to questionnaires)

Disadvantages – (are similar to questionnaires)

 Interviewer cannot build up a rapport – a close and harmonious relationship in which

there is common understanding - with the person being interviewed because the
whole process is mechanical and scientific.

8|Page
Structured interviews vs questionnaires

Structured interviews:

 Can be more reliable than questionnaires.

 Often used in large scale social surveys e.g. Norris etal (1999) used the method as
part of a survey of 3615 people’s responses to political campaigns in the 1997
general election.
 Have a better response rate than questionnaires especially postal and website
questionnaires.
 An interviewer can clarify ambiguous questions, making misinterpretations less
likely.

However

 They are not perfect e.g. suffer from ‘interviewer effect.’

b) Unstructured interviews

It is a free flow conversation. The interviewer needs to build a relationship of trust –

rapport – with the interviewee.

They are popular with interpretivists e.g. Charlesworth (2000) used unstructured
interviews to create a valid description of working class experiences.

Advantages

 Produce qualitative data that gives an insight into the interviewee’s emotional
reactions.
 The person interviewed is more likely to be honest if they feel a rapport with the
interviewer (i.e. response tend to be more valid)
 Interviewer is able to follow up any confusing or interesting responses.

Disadvantages

 They are expensive, time consuming and it’s hard to estimate how long each
interview will take in advance.
 It is not easy to make comparisons between one respondent and another since it
produces qualitative data.
 A bad interviewer can produce poor results, especially if the interview goes off –
subject.
 They are not reliable, if a researcher repeats the study it will probably produce
different results.
 The sample usually has to be small because they tend to produce long, unwieldy
responses i.e. difficult to manage because of its size.
 It is difficult to generalise the results because answers tend to be specific to
individual subjects.
 Bias can get in the way at each stage of the process. An ‘interviewer effect’ is when
the response given is not what the interviewee really thinks. This can be caused by

9|Page
 the gender, class, age or personality of the researcher. The opinions of the
researcher and interviewer can influence the interviewee’s responses.

c) Semi structured interviews

 It is an interview in which a few key questions are used to give shape to the
conversation i.e. the interviewee can answer in any way they like and the interviewer
can ask as many follow up questions as they feel are necessary.
 It is less restricted than structured interviews, but more likely to produce data that
can be compared with other respondents than unstructured interviews.
 For example, Muggleton (2000) used semi - structured interviews to research the
link between fashions and subcultures in post-modern society.

d) In-depth interviews

These are mostly useful when investigating sensitive issues e.g. sexual abuse.

 OBSERVATION TECHNIQUES
Observation is watching behaviour in real situations. To understand the way of life
of a social group, a researcher may decide to observe that group in its natural
setting, either by becoming a member of the group, participating in all its activities
(participant observation) or just taking notes without participating in what the other
members are doing(non-participant observation – ‘fly on the wall approach). An
example of P.O is Paul Willis’ study on counter school culture and Malinowski on
totemism.

Types of P.O

The researcher doing observation decides well in advance whether to make his
identity known to the group or to keep it secret.

a) Covert observation

The researcher does not tell the group that they are being observed (keeping one’s
identity secret), although it is not recommended e.g. Fielding (1993) used covert
observation researching the National Front (a political party) because he believed he
would encounter hostility if they knew he was a Sociologist.

b) Overt observation (direct observation)

This is when the group is aware of the research and they know who the researcher
is (researcher reveals identity to the group being studied) e.g. Skeggs (1991) used
overt observation studying female sexuality among students at a college.

Situations where P.O can be used

 In ethnographic studies - PO is used to study complex phenomenon like culture which

cannot be studied from a far distance but needs one to be involved to gather the
correct reasons given to certain cultural activities.

10 | P a g e
 In leadership, politics, social interaction and taboos as these areas need social
interactions with the participants.
 When the researcher is the source of questions to be addressed with participants.

Strengths of participant observation

 Validity

Personal involvement and experience with the subjects of a study means the
researcher gains valuable insights into the meanings, motivations and relationships
within a group that, in turn, may explain why people behave in certain ways.

 Access

Gets the researcher right to where the action is, so they can check the dynamics of
the group from close up. Useful for studying subcultures e.g. Thornton (1995)
investigated clubbing subculture (i.e. he dressed like a clubber and attended raves)
and lifestyles e.g. Abrams (2002) took a series of minimum wage jobs to explore life
as experienced by the poorest workers in Britain.

Allows the researcher to research the workings of deviant groups e.g. prostitutes.

 Primary data

The researcher gets first hand insight of people in natural –life settings.

If it is covert people cannot mislead the researcher.

In this regard participant observation studies may be said to be high in validity.

 Qualitative data

Intrerpretivists prefer observations because the researcher can get to the action and
tends to produce qualitative data that is more valid.

Limitations of participant observation

 Poor reliability especially covert research.

 Representativeness is generally poor although interpretivists argue that it is worth
losing this.
 The researcher may get too involved and find it hard to stand back and objectively
observe the group.
 Observer effect
The observer’s presence changes the way the people behave - and it’s probably true
that, at least initially, Venkatesh’s presence had some (unknown and unquantifiable)
effect on people’s behaviour. Overt research may influence the behaviour of the
group.
 In covert observation the researcher may join in with illegal acts if they are in a
deviant group.
 Replicability

11 | P a g e
 Other researchers cannot repeat the research – i.e. it lacks reliability.
 Recording
A covert observer may find it difficult to remember all the events and accurately
record them.
 Ethical problems
The researcher may encounter ethical problems in getting in, staying in and getting
out of the group. The British Sociological Association set guidelines for research
involving participant observation:
- Subjects must give informed consent – should fully understand what the research
is about, how data is collected and what the data will be used for.
- Avoid doing any physical, psychological or emotional harm to their subjects during
research.
- Special care should be taken when researching vulnerable people e.g. children,
elderly.
- Anonymity (unknown identity) and privacy of research subjects must be protected.
 Sample size
It involves a small sample, so it is not representative of the population i.e. it is difficult
to generalise the results to the whole population.
 Practical problems
It involves hard work, and is time consuming and expensive (practical problems)

Non participant observation

Strengths

 Researcher is not drawn into the group so they can be more objective about the
group behaviour.
 If one wants to study deviant groups (s) he has to be inconspicuous (not clearly
visible or attracting attention).

Limitations

 Observing from the outside stops the researcher from getting to where the action is.
 Overt research may influence the behaviour of the group.

Conclusion

In conclusion the usefulness of any research method depends on a range of factors.

Positivists reject the method because it is unscented, it lacks objectivity and it is
impossible to achieve the large samples necessary to find correlations and make
generalisations. However, interpretivists are concerned with validity and gaining an
emphatic understanding, then PO is the ideal method to use.

 EXPERIMENTS

Experiments are a form of primary research method used in natural sciences mainly
to test a hypothesis by controlling a variable where the researcher manipulates one
variable (i.e. Independent variable) and observes its effects on the dependent variable.

12 | P a g e
To discover which factors are likely to produce an effect, external factors have to be
removed and this is why they are often carried out in a lab.

Experiments let the researcher establish the cause and effect relationships between
two variables.

Types of variables

All variables are kept constant, apart from the one you are interested in - the
INDEPENDENT VARIABLE i.e. scientists change/manipulate/controlled the I.V and
observe its effects on the DEPENDENT VARIABLE.

Example: testing the effects of alcohol on memory, alcohol be the I.V you control,
memory would be the D.V which you measure.

Subjects are typically divided into two groups: the experimental and control group.

a) Experimental group

Experimental group contains the subjects who are exposed to an independent variable
to study its effect on them.

b) Control group

The control group contains the subjects who are not exposed to the IV. It is a group
left alone to see what happens if the researcher does not do anything to the
Independent Variable.

The experimental and control group are then compared to see if they differ in relation
to the DV.

Types of experiments:

 Lab experiments

These are done in a controlled environment. Researcher changes the Independent

Variable and observe effects on the Dependent Variable. Laboratory experiments are
mostly used by psychologists, natural scientists and positivists.

 Field experiments

These take place outside the lab in real social settings. Those involved are often
unaware that they are being researched. Used by interpretivist sociologists.

 Natural experiments

These are not set up artificially. For example: twin studies

Examples of experiments carried out in social research

 The Hawthorne experiment (Mayo)

It falls under the human relations school, which reflects the influence of sociologists
and psychologists on the development of management theory. Business organisations

13 | P a g e
are seen as social systems in which psychological and emotional factors have a
significant influence on productivity.

The first significant research was Mayo’s experiment at the Hawthorne works of the
western electric company in Chicago (1923 to early 1930s). Mayo altered the working
environment e.g. by adjusting the lighting, to study the impact on the productivity of
workers. He discovered that whatever he did was followed by a rise in productivity.

From these experiments he concluded that employees worked harder if they believed
that management was concerned about their welfare. If work appears to have meaning
and if people feel valued, they will be more content in their work and more productive.

This relationship between attention paid to workers and the subsequent rise in
productivity has become known as the “Hawthorne effect” i.e. when people know they
are being studied they sometimes act differently or modify an aspect of their behaviour
being experiementally measured simply in response to the fact that they know they
are being studied, not in response to any particular experimental manipulation i.e. data
from experiments may not be valid.

 Bobo doll experiment (Bandura social learning theory)

Refer to notes on nature – nurture debate or mass media.

 Twin studies

One studies genetically identical twins in different situations, to see whether

something is caused by genetics or if it’s caused by socialisation.

 Milgram’s Electric Shock experiment.

In 1963 Stanley Milgram carried out a laboratory experiment. He told participants that
the experiment was to study how punishment affected learning.

The research group was divided into ‘teachers’ and ‘learners’. The ‘teachers’ saw the
‘learners’ sit in a chair that looked like an electric chair. The ‘teacher’ was then asked
to test the ‘learner’ and if they made a mistake give them an electric shock.

The equipment used to give the electric shock was labelled 15 volts to 450 volts. The
450 volts was clearly labelled ‘Danger: severe shock’. The ‘teacher’ was told the shock
would be painful but it would not hurt the ‘learner’. The equipment was really a fake.
The ‘learners’ were not really in danger but as the voltage went up the ‘teachers’
heard cries of pain. At 330 volts the ‘learners’ went silent.

In the experiment none of the ‘teachers’ questioned the activity until the ‘learners’ fell
silent. This showed that people will usually follow the instructions of an authority
figure without question.

Strengths of experiments

 They produce data which are reliable

 Natural social settings (field and natural experiemts) are more like real life, hence
it is most likely to give valid data.

14 | P a g e
  Show hidden meanings of everyday social interaction.
 Experiments lets the researcher find cause and effect relationships as well as make
correlations.
 The researcher has control over the experiment, especially lab experiments.
 Obtains data which is easily quantifiable i.e. quantitative data.
 Other researchers can replicate (repeat) the research, especially lab experiments.

Limitations experiments

 In field experiments the researcher cannot control variables like in lab experiments.
 People may change their behaviour if they know they are being studied ie
‘Hawthorne effect’
 Some field experiments are associated with many ethical problems, for example at
times the subjects are not aware that they are taking part in an experiment.
 It is hard to reproduce real social situations (lab experiments are artificial).
 Difficult to isolate single variables. Social behaviour is influenced by many factors.
 There are often moral and ethical issues in lab experiment.
 People may feel intimidated or act differently in the lab.

 LONGITUDINAL STUDIES

These are studies of the same person over a long period of time to find out how much
change is taking place in a society or smaller group.

Longitudinal studies are mainly used by positivists, since they bring quantitative data.

Examples of longitudinal studies

 ‘Seven up’

‘Seven up’ was a documentary that asked fourteen (14) kids what they thought about
life, and what they wanted to be when they grew up. The programme makers came
back to interview the children every seven (7) years. The latest instalment was ‘forty
nine up’.

Strengths

 Researchers can analyse changes and make comparisons over time.

 It provides more valid data than other types of research methods.
 It is more reliable to follow a particular sample over a long period of time.
 They often examine a large number of variables.
 It has the ability to pick up changes and one can make comparisons over time.

Limitations

 It is hard to recruit a committed sample who will stay with the study.
 It is hard to keep in contact with the sample, which may make the study less valid.
 The researcher needs long term funding and keep the research team together.

15 | P a g e
 Longitudinal studies rely on interviews and questionnaires which might not be valid
or reliable.
 They are time consuming.

 DOCUMENTS

A document is a written text. These are a source of secondary data.

Types of documents

a) Personal documents

Letters, diaries, autobiographies, memoirs and suicide note.

Strengths of personal documents.

 Used to supplement quantitative secondary data like official statistics e.g. a patient’s
diary can be used alongside statistics about the treatment of mental health.
 They give a rich, detailed and valid insight into everyday health experiences and
practices in their natural environment.
 They are also used where no other source of data exists.
 They are often the only insight that sociologists have into the past.

Weaknesses of personal documents.

 They tend to be biased and therefore invalid because the person might want to justify
their actions and not present an objective view of the situation.
 They are not reliable because it cannot be checked for accuracy by the sociologists.
 They are not representative and typical i.e. not all people are used to diaries – diaries
etc. tend to be mainly the product of middle class professionals.
 There are doubts about their authenticity and therefore may be invalid.
 Sociologists may interpret what the writer is saying in a different way to what the
writer intended.
 Personal documents are not easily accessible.
b) Official/Formal documents

For example school records, health records, church records, social work records etc.

c) Expressive documents

These are documents which have more to do with meanings, e.g. suicide note, are
mostly preferred by interpretivists because they give qualitative data.

d) Public and historical documents

These are published documents e.g. gvt reports, reports of companies etc. they can
be contemporary or historic.

For example it is used by Weber studying how religious beliefs of Calvinism brought
about social change.

Strengths of public and historical docs

16 | P a g e
 They are often the only way of finding out how things were done in the past.
 Allow comparisons over time in order to identify trends in mortality, morbidity etc.
 Can help show the effectiveness of social policy.
 Data is readily available , hence less time consuming

Weaknesses

 There are doubts about their authenticity that undermine their validity.
 Their validity and objectiveness may be questioned.
 They are not reliable.
 May be high levels misinterpretation

 OFFICIAL STATISTICS

They are a source of secondary data. They are produced by local governments, central
governments and gvt agencies and can be:

Types of O.S

a) Hard statistics

Hard statistics are objective e.g. statistics on births and marriages. Politicians cannot
fiddle with them.

b) Soft statistics

These are more subjective since politicians can fiddle with them e.g. statistics on
crime, poverty, unemployment etc.

c) Census

A survey of every household every 10 years (in Zimbabwe).

d) Social trends

A collection of regular gvt surveys published every year.

Strengths of official Statistics

 Easy and cheap to access – involve little effort on behalf of the sociologist.
 Usually extreme contemporary/current e.g. the census.
 Gathered by surveys which involve large representative samples and their findings
can usually be generalised to similar populations.
 Positivists see official statistics as hard reliable facts because they have been
collected in a standardised, systematic and scientific fashion.
 Statistical relationships can be identified by comparing official statistics from
regularly conducted surveys e.g. census.
 Trends over time can easily be seen.

Weaknesses

17 | P a g e
 They may not represent a complete picture e.g. illness statistics only cover reported
illness to doctors and hospitals i.e. there is a ‘clinical iceberg’ of unreported self-
medicated illness.
 Open to political abuse – can be manipulated by governments for political advantage,
especially soft statistics.
 May be socially constructed – they are the end result of someone making a decision
that a particular set of activities needs recording.
 Statistics tell us very little about the human stories or interpretations that underpin
them e.g. morbidity statistics tell us little about the social experience of illness.
 Official statistics may be based on operational definitions that sociologists would not
agree with e.g. gvt works out the amount of illness in a society by counting visits to
doctors and hospitals whereas sociologists may prefer to uncover the amount of
illness by asking people about their subjective experiences of being well, or unwell
which may uncover a dark figure of unreported illness.

 FOCUS GROUPS

It is a group which consists of a very small sample, perhaps fewer than ten people.

The sample is put in a room together, and asked to talk about a particular issue or try
to answer a specific set of questions, the discussion is observed by the researcher.

For example, Chahal and Julienne (1999) researched the effects of racism, with
respondents discussing their experiences of racism etc. using focus groups.

 CASE STUDIES

These are sometimes known as in-depth studies where a small group of people, or a
particular event (e.g. strike) are examined in detail. The results obtained are then
generalised to all peephole.

These are an analysis of persons, events, decisions, periods, projects, policies,

institutions or other systems that are studied holistically by one or more methods.

Case study methods

a) Prospective

An individual or group of people is observed in order to determine outcomes e.g. a

group of individuals might be watched over an extended period of time to observe
the progression of a particular disease.

b) Retrospective

involves looking at historical information e.g. researcher starts with an outcome

such as disease and then backwards at information about the individual’s life to
determine risk factors that may have contributed to the onset of the illness

Limitations

18 | P a g e
  It is difficult to generalise the results.

 CONTENT ANALYSIS

It involves the analysis of patterns or messages in the mass media e.g. newspapers,
magazines and TV programmes.

Content analysis is a method of systematically analysing a communication e.g. a

speech, letter, film and mass media to understand its meanings.

Can generate both quantitative data (involving a counting approach and normally uses
a content analysis schedule that records the patterns and frequency of certain images
or themes) and qualitative data (known as semiology, involving analysing language or
images, to work out how they might symbolise a particular political or cultural position.

The Aim is to identify how particular social groups are portrayed in the product being
analysed.

Strengths of media content analysis

 Cheap, all the sociologists need to do is to buy the newspaper or magazine or watch
the TV programme.
 Allows sociologists to compare depictions and representations over a period of time
if a longitudinal version of content analysis is used e.g. anti-smoking television
adverts could be analysed from different time periods to work out their
effectiveness.
 Quantitative content analysis is reliable because other sociologists can cross check
the results by looking at the media using the same categories.

Weaknesses of media content analysis

 Time consuming method because media products would need to be checked over a
fairly long period of time.
 It is a subjective method e.g. semiotics largely depends upon the interpretation of
the researcher. Different researcher may interpret the same images differently.
 Sociologists who have used this method have been accused of taking the meaning
of an image or set of words out of context.
 The use of content analysis of media products often assumes that these products
have an effect upon their audience which may not be the case. Media products may
only tell us about the personal and political beliefs of journalist and broadcasters
who produced them.

 SEMIOLOGY (also known as semiotics)

It is the study of symbols and signs and their use or interpretation.

 ETHNOGRAPHY

19 | P a g e
Studies the way of life of a group, the scientific description of a specific culture by
someone with first-hand experience of observing that culture.

 TRIANGULATION

All research methods have strengths and weaknesses. Some sociologists combine
different methods – they want to combine the strengths of all the different methods to
overcome their weaknesses. This is known as triangulation.

Feyerbend (1975) was one of the first people to reconcile the positivist and
interpretivist approaches, arguing that sociologists should combine quantitative and
qualitative methods as necessary, and even change methods during the process of
research if the study calls for it. His ideas contributed to the development of
methodological pluralism i.e. studying society using a mixture of methods instead of
having to declare one in advance.

Triangulation is when sociologists combine different methods or data to get the best
out of them all e.g. combining questionnaires (quantitative method) and unstructured
interviews (qualitative method).

Methodological pluralism is the rejection of the idea that any one theoretical approach
to research is superior, and acceptance that sociological studies may be enhanced by
the use of different techniques for analysis and collecting data.

Methodological pluralism became popular in the 1970s.

Three kinds of triangulation (according to Harvey and MacDonald):

 One researcher uses two or more methods. So the results from one method can be
checked against the results from another.
 Two or more researchers use the same method. So the results obtained by one
researcher can be checked against the results obtained by the other.
 Two or more researchers use two or more methods. So researchers and methods
can be checked against each other.

Examples of triangulation studies

a) Barker (1984) studied the Moonies, a religious organisation using in-depth

interviews, overt participant observation and questionnaires.
b) Ghaill (1994) studied masculinity using overt participant observation and interviews
with young straight men and young gay men. He discussed ideas about what made
a man masculine.
c) Dodson etal (2001) studied childhood disability and family life using a mixture of
diaries, questionnaires and focus groups. However, when they organised a focus
group on categories of disability, they found that the various children had such
complicated medical histories that they could not be categorised. So the focus group
turned out to be not fit for purpose, but Dodson etal still managed to complete their
research because that focus group was just one of several methods they were using
to obtain data.

20 | P a g e
Benefits of triangulation

 Gives a more detailed picture than when a researcher uses only one method, so it’s
more valid.
 The researcher can check different sets of data against each other, so it is more
reliable. A researcher can even triangulate between data produced by different
researchers.
 It combines the strengths and the weaknesses of different types of data.
 Mixing methods makes it more likely that some are “fit for purpose” – is capable of
giving useful results e.g. if you were studying people’s experiences of crime, a
questionnaire given to a large sample should produce useful results. So, for that study
a questionnaire is fit for purpose. However, it is not always obvious which methods
are going to be fit for the purpose before a study begins. Using mixed methods can
be an advantage – if one of your methods turns out to be not fit for purpose during
the study, then you could still be able to collect useful results from the other methods.

Weaknesses

 Triangulation can be very expensive.

 It is time consuming to do the same research with lots of methods.
 Sometimes it is not possible to use triangulation e.g. you might only be able to use
one method, or there might not be very many different kinds of data for what you
are studying.
 Makes it harder to analyse data – trying to find patterns is more difficult when you
have to simultaneously analyse lots of different kinds of data etc.

 FACTORS INFLUENCING CHOICE OF RESEARCH METHOD

The type of research method that sociologists use is undoubtedly guided by their
values and by the theoretical perspective that most influences their work.

 Theoretical factors

 The theoretical background of the researcher

Positivists use reliable methods that give quantitative data, for example
experiments, questionnaires, official statistics etc. Interpretivists use valid methods
that give qualitative data, for example observations, unstructured interviews etc.

 Topic to be studied

Some topics lend themselves more readily to the use of quantitative methods e.g.
voting patterns involves large numbers of people which require quantitative
methods like questionnaires.

 Reliability

21 | P a g e
 Reliability is the extent to which a technique for collecting or analysing information
may be used repeatedly and produces the same results. It is when the researcher
decides to replicate a research, using the same questions, the same group, with the
same method, with the same sample and get the same results.

Qualitative methods are often criticised for failing to meet the standards of reliability.

 Validity

Data are valid if they provide a true picture of what is being studied. Supporters of
qualitative methods argue that quantitative methods lack validity.

 Practical factors

 Time available

Quantitative methods are generally less time consuming and require less personal
commitment as compared to qualitative methods e.g. participant observation studies
take a long time.

 The amount of financial resources

Financial resources are needed to pay the researcher, transportation to interviews,

pay for resources like computers etc. Large scale surveys are expensive e.g. census

 Source of funding for a proposed project

Many funding bodies support the use of quantitative methods.

 Access and opportunity (the research population)

The researcher may not have access to certain groups to carry out interviews, then
they turn to secondary sources.

 Characteristics and skills of the researcher

It will be difficult for a female researcher to be involved in a participant observation

of monks in a monastery.

 Ethical factors

As in the choice of research topic, ethics can also affect choice of research method.
Ethics are moral guidelines that sociologists follow when undertaking and writing their
research. Ethical procedure was outlined by the British sociological association as
follows:

 Informed Consent
 Avoidance of Harm
 Informed consent
 Limited deception
 Privacy and confidentiality of data
 Anonymity

22 | P a g e
  Debriefing

Ethical concerns pose great difficulties for sociologists engaged in certain research
methods especially covert participant observation.

Conclusion

Methodology is guided as much by practicalities as it is by theories, ethics and values.

 IMPORTANCE OF RESEARCH IN ZIMBABWE

 SUBJECTIVE AND OBJECTIVE RESEARCH IN SOCIOLOGY

One of the most bitterly contested concepts in sociology has been over the question
of the place of personal and political values in theory and research. Distinct positions
can be identified on this issue:

a) Value freedom (objectivity)

Sociology has to be free of personal and political biases. This is known as value
freedom or objectivity.

Sociology should seek to copy the methodology of the physical sciences i.e. the
importance of objectivity. Theories must be constructed and tested.

Social facts (statistics obtained by surveys and official publications) should be a

perfect reflection of the subject under study

Personal opinions and political biases of researchers are irrelevant as long as the
research is well designed and there is no attempt to distort the findings.

b) Value laden (subjectivity)

Ideally our personal values should not intrude into our sociological studies, but in
practice it is impossible to keep them out. This is known as value laden sociology.
Evidence to support this claim include:

 Historical context
 Paying for research
 Career trajectories
 Personal beliefs and interests
 The domain of sociology
 The postmodern critique
 Foucault
 Committed sociology
Any one doing sociology must surely use their studies to improve the condition of
those most oppressed in society. Sociology is therefore more a tool that helps bring

23 | P a g e
 about social change – committed sociology – than just an academic subject studying
society.
 Marxist view
 Feminist view
 Left realism

END…………………. ©

24 | P a g e