Upgrade to remove ads
Only $35.99/year

Research Methods: Sociology

Terms in this set (84)

The PERVERT model
P: Practical issues e.g. cost, time
E: Ethical issues e.g. consent
R: Reliability - Whether the method can be repeated and get similar results
V: Validity - The truthfulness of the research
E: Examples of method
R: Representativeness - Does the sample reflect the wider population?
T: Theory - Positivism or interpretivism
Primary methods
Methods that involve the researcher collecting the data themselves
Social Surveys
A social survey involves the collection of data from a fairly large number of people and this often means asking everybody the same set of questions. Questionnaires and interviews can be social surveys.
Questionnaires
Questionnaires are a list of standardised, Preset written questions that respondents are requested to complete by themselves. They are distributed by post, online or in person.
When writing questionnaires, what has to be considered,
Types of question
Design and layout
The imposition problem
Operationalising concepts
Questionnaires - Types of questions
Open questions: No pre-set choice of answers and require more than one word response.

Closed questions: Pre-set limited choice of answers
Questionnaires - Design and layout considerations
Questions should be understandable and explained.

Unnecessary information shouldn't be included.

It should be clear which answers belong to the questions.

Variety of answers should be included.

Instructions should be provided.

Text and font size should be clear.

Enough space should be left for people to answer questions.

Double-barrelled questions should be avoided.

The question should begin with the quickest and simplest questions.
The imposition problem
This is when the researcher imposes that own values, attitudes, or thoughts onto the participant. Clear and neutral language should be used and leading questions should be avoided.
Operationalising concepts
Questions and key terms need to be operationalised so they can be measured. Always ensure that key times are precise and the answers are put into measurable categories.
Questionnaires - Practical issues
Quick and cheap to reach a large sample.
Quick to analyse and present.
Easy to do a pilot study.

Postal questionnaires can have a poor or slow response rate.
Questionnaires - Ethical issues
Useful with consent.
Anonymity good for sensitive issues.
Participants can leave questions unanswered.

May be too impersonal for sensitive issues.
Questionnaires - Reliability
High reliability - Usually a large sample and standardised

Non-operationalised concepts
Questionnaires - Validity
No interviewer affect

Paul validity - Lacks detail and explanation, limited options to select l, dishonesty, imposition problem
Questionnaires - Examples
The census - A questionnaire that is sent to every household.

Callender and Jackson (2005)
- 2000 students in England
- Fear of debt arising from university fees
- Higher and lower social classes
Questionnaires - Representativeness
Representative and generalisable if there is a large sample.

Affected by a poor response rate.
Certain type of participant?
Questionnaires - Theory
Favoured by positivists - Quantitative, reliable, objective

Criticised by Interpretivists - Low validity
Interviews
Interviews involve a situation where the researcher asks a participant questions and they can be conducted face-to-face or via the telephone for example. Interviews involve social interaction between the interviewer and the interviewee.
Structured interviews
Structured interviews tend to be formal interviews.

An interview schedule is used which is a predetermined set of usually closed questions.

The interviewer is given strict instructions on how to ask the questions.

The interview is conducted in the same standardised way each time.
Unstructured interviews
Unstructured interviews tend to be informal and/or in-depth interviews.

Often there are basic areas for discussion planned but few if any pre-set questions.

The interviewer has complete freedom and can also encourage in-depth answers with open questions.
Semi-structured interviews
Semi-structured interviews are used when the interviewer has some pre-set questions but they also have the freedom to deviate away from the set list and can probe for more information.
Focus groups
The researcher sets a discussion topic for a small group of people and allows the group to tease out responses from one another.

The researcher acts as a facilitator helping to keep discussion on topic and to encourage members of the group to participate.
Structured interviews - PERVERT Evaluation
P: High response rate and quick to analyse.
Time and cost compare to questionnaires.

E: useful with consent.
Formal for sensitive issues and potential to feel less anonymous.

R: Higher (Large sample, Standardised)

V: Low, interviewer effect

E: Willmott and young, Dobash and Dobash

R: High (large sample)

T: Favoured by positivists (quantitative/reliable)
Unstructured interviews - PERVERT evaluation
P: Quicker to Design.
Time and cost compare to questionnaires, Harder to obtain a sample.

E: Useful with consent and for sensitive issues.
Potential to feel less anonymous than questionnaires.

R: Low (Time-consuming and not standardised)

V: Higher (open questions) but interviewer effect

E: Sharpe, Dobash and Dobash

R: Lower (smaller sample)

T: Favoured by interpretivists (qualitative/valid)
Ethnography - Observations
Ethnography is the study of simple small-scale societies by living with the people and observing their daily lives.

Sociologists use this term to describe modern day observational studies.

However observations do not have to be as extreme as living with the people you are studying.

Overall observations involve watching a group's behaviour in their usual setting.
What are the three key considerations researchers need to think about when completing observations?
1. Getting in, staying in and getting out of the group being studied.

2. How involved should the researcher be?

3. How much information should the researcher give to the participants?
Observations - Getting in
How easy is it to make initial contact and enter the group?

How easy is it to be accepted?
Observations - Staying in
How easy is it to maintain the group's acceptance?

How easy is it to avoid 'going native' (Becoming so involved in a group that all attachment is lost)
Observations - Getting out
How easy is it to leave the group behind without damaging relationships?

How easy is it to leave the group safely?

How easy is it to re-enter the 'normal' world?
Participant observation
The researcher joins a group and studies their behaviour.
Strengths of participant observation
Deeper understanding of the group.

More validity (Full experience, develop trust)

Useful for hard to access groups once you've identified a gatekeeper.
Non-Participant observation
The researcher observes the group but does not seek to join in their activities.
Strengths of non-participant observation
Reduced observer effect.
Objectivity and avoids 'going native'.

Easier to observe and record data.
Overt observation
The researcher is open about the research role.
Strengths of overt observation
No deception.
More ethical - Consent, Right to withdraw.
Easier to take notes.
Can conduct interviews.
Can help to avoid 'going native' and easier getting out.
Covert observation
The researcher does not admit to being a researcher.
Strengths of covert observation
No observer effect.

Valid - Research and observations are more truthful due to no observer affect.

Useful for hard to access groups.

Can gain acceptance more easily.
Ethnographic research - Practical issues
Good for hard to access groups.

Time-consuming and expensive.
Recording data.
Getting in/staying in/Getting out
Ethnographic research - Ethical issues
Ethically useful if overt

There can be deception, no consent, illegal activity, harm etc
Ethnographic research - Reliability
Low:
One precise instance.
Subjective interpretations.
Time-consuming.
Ethnographic research - Validity
High - In-depth, first hand experience, 'verstehen'.

Observer effect
Going native
Recording data later
Ethnographic research - Examples
Venkatesh (2008): Participant, overt

Richards (2012): Non-participant, overt

Patrick (1973): Participant, covert

Moore (2008): Non-participant, overt

Humphreys (1970): Participant, covert (mainly)
Ethnographic research: Representativeness
Low - Usually one instance
Ethnographic research - Theory
Favourited by interpretivist's: Valid, qualitative, Verstehen

Criticised by positivists: Not reliable or objective
Experiments
Experiments involve a highly controlled situation where the researcher tries to control variables in order to establish cause and effect relationships.

Experiment of the most extreme form of positivistic research and are mainly used in the sciences.

The researcher controls the variables to test the hypothesis.
The Hawthorne effect
Changes in the behaviour of participants due today awareness that they are taking part in an experiment.
How experiments work
Firstly the researcher will have a hypothesis to test. A hypothesis is a prediction which the researcher guesses might be true.

The researcher then takes two groups that are alike in every way and changes variables in one of the groups.
The control group
The group that stays the same in an experiment
The experimental group
The group that has variables changed in an experiment
The independent variable
Watch the researcher changes in an experiment
The dependent variable
What the researcher measures in an experiment
What are the two main types of experiment?
Laboratory experiments: Laboratory experiments are experiments involving artificial environments and high levels of control.

Field experiments: Field experiments are experiments involving normal social situations and some but less control variables.
Advantages of laboratory and field experiments
Advantages of laboratory experiments:
More control of variables.
More reliable.
If a field experiment is not possible.
More likely to have consent.

Advantages of field experiments:
More practical, Saves time and money.
More valid.
Reduced Hawthorne effect.
Laboratory experiments: PERVERT evaluation
P: More control of variables.
But time/cost/access to participants.

E: Immoral manipulating people, Deception and consent etc

R: Higher (Control of variables)

V: Low validity (Artificial and controlled) - Hawthorne effect

E: Milgram (1974): Lab, Bandura (1973) Lab

R: Low (small-scale)
Field experiments: PEVERT Evaluation
P: Easier to set up.
But less control of variables, Time/cost/access to participants.

E: Immoral manipulating people, more deception, no consent etc

R: Lower (Less control of variables)

V: More validity (real life setting), Hawthorne effect reduced still controlled, still exists if participants aware.

E: Elliot (1968) Field, Mayo (1925) Field, Rosenhan (1973) Field

R: Higher due to usual range of people but small-scale is still low
Case studies
A case study is the detailed study of a single example of whatever it is the sociologist wishes to investigate.

E.g. A single school, class, business group, individual or life history

Case Studies generate quantitative and qualitative data but mainly qualitative.
Case Studies - PERVERT Evalutation
P: Useful for new insights and information.
But time and cost to complete and analyse.

E: Ethical issues usually accounted for such as consent

R: Low - Level of detail and individuality of each case

V: High - Rich, detailed information
Can be affected by researcher effect and poor recall

E: Willis (1977): Learning to labour

R: Low - one instance

T: Favoured by interpretivists - Mainly qualitative and high validity
Longitudinal studies
Longitudinal studies take place over a period of time, sometimes years to evidence change over time (e.g. norms, values, attitudes).

This is opposite to a cross-sectional survey which provides a snapshot.
Longitudinal studies - PERVERT Evaluation
P: Analyse change over time and cause and effect relationships.
Time-consuming and expensive to complete and analyse.

E: Ethical issues usually accounted for

R: Low - Level of detail and time

V: High, overcomes poor recall BUT researcher effect

E: Douglas (1964-70)

R: Low overall - One instance BUT Improved by sample size

T: Favourite by interpretivists - Mainly qualitative and high validity
Secondary methods
Using material that already exists or has been collected by others.
Quantitative data
Statistical and numerical information
Official statistics
Official statistics are the main example as they are the most widely used a secondary sources sociological research.

They are collected by the national and local governments.

Statistics can be collected on education, health, crime, marriages, divorces, births and deaths etc

The government department responsible for collecting and publishing statistics about the UK is the office for National statistics (ONS)
The Census
The census is a questionnaire sent to every household every 10 years which must be completed by law. It collects information on the population and generates official statistics.
Advantages of statistics generated from the census
Large sample size
Representative/generalisable
Reliable and rigourous collection
Documents social change
Identify international comparisons
Influences policy
Easily accessible and easy to analyse
Saves time and money
Easily accessible sample
Disadvantages of statistics generated from the census
Statistics lack detail which could affect the validity.

Validity of statistics is affected if census incorrectly completed.

2001: 0.7% of the population class their religion as Jedi

Based on households - Not representative of the population

Takes time to analyse

Long time between results
Crime statistics
Official crime statistics have been collected since 1857.

Crime statistics are typically generated from recorded crimes by the police and they are published every six months by the Home Office.

However, crimes are not always reported and recorded.
Why might crimes not be reported?
People might be embarrassed or scared to report crime.

In cases of domestic abuse the victim I still love their partner.

In cases of child abuse children may be unable to report abuse.

Lack of evidence that a crime has been committed.

People may have empathy for the person that committed the crime.

People who are unsure how to report crime.

Small crimes may seem unimportant.

Lack of faith in the police.

People who choose to do with the crime themselves.

People may think reporting the crime is an inconvenience.

Victimless crime e.g. Corruption in businesses

Fear of consequences
Why might crimes not be recorded?
One in five crimes reported are not recorded.

Police discretion

Police targets

Paperwork

Insufficient evidence

Unreliable source or disbelief
The dark figure of crime
Official crime statistics cannot provide an accurate picture of crime rights as a dark figure of crime exists.

Some criminologists claim that only about 25 to 30% of all crimes appear in official crime statistics.

Victimisation surveys have been introduced to combat this problem.
The crime survey for England and Wales
This is a systematic victim survey carried out by the Home Office.

Collect statistics on the amount of crime in England and Wales by asking a large sample of people (approximately 50,000 households) About the crimes they have experienced in the last year, Including the crimes they didn't report.
Advantages of statistics generated from the crime survey
Includes those not reported or recorded.

Helps to estimate how large the dark figure is

Identifies patterns of victimisation

Anonymous

Reliable

75% response rate and the sample is representative of 98% of the population
Disadvantages of statistics generated from the crime survey
Faulty/biased memories

Certain crimes still under reported

Victimless crimes

Based on households

Under 16s not included
Hard and soft statistics
Hard statistics:
More valid in that they actually measure the thing they claim to measure

Soft statistics:
Give a much less valid picture and don't measure the thing they claim to measure in its in entirety
Qualitative data
Descriptive and in-depth data
The mass media
The mass media or a secondary source but some studies using the mass media produced primary data for example content analysis.

This is one documents and other sources are examined in detail to see what themes occur.

In particular semiotic content analysis involves examining themes and identifying underlying meanings.
Quantitative of secondary sources - PERVERT Evaluation
P: Cheap and quick to obtain and analyse. Influence policy.
Legal restrictions sometimes.

E: Issues already dealt with

R: High - Large sample and rigourous collection

V: Hard statistics But can be soft statistics

E: Census statistics, Crime statistics

R: High due to large sample

T: Favoured by positivists (Quantitative, reliable and objective
Qualitative secondary sources - PERVERT evaluation
P: Usually quick to obtain, Only source available.
Can be costly to obtain.

E: Issues already dealt with but need to ensure consent.

R: Low - One instance and based on interpretation.

V: High - Rich in detail but could be bias or not genuine

E: Personal letters/diaries/public newspaper/magazines. Aries: oil paintings

R: Low as one instance
Effectiveness of secondary sources: Scott (1990)
For secondary sources to be effective according to Scott (1990), the researcher needs to think about authenticity, credibility representativeness and meaning.
Authenticity
Is the source genuine or fake?
Credibility
Is the evidence believable, sincere and honest? Or does it contain bias, exaggerations and distortions?
Representativeness
Is the document typical of those appearing at the time? Is the evidence complete or are documents missing? What about those people in the past who couldn't read or write?
Meaning
What do documents mean? Do they have the same meaning now as they did at the time they were first produced? Secondary sources are always created for another purpose have you understood the meaning of this purpose?
Triangulation a methodological pluralism
Triangulation of methodological pluralism mean using two or more methods to study an issue.

Methodological pluralism also means combining both positivist and interpretivist approaches.
Why do you researchers use triangulation and methodological pluralism?
Generate both quantitative and qualitative data.

Improved breadth and depth of information.

Improve overall reliability and validity.

A larger sample improves representativeness and generalisability.

Strengthens research

To check findings
Examples of triangulation of methodological pluralism
Sharpe: Unstructured interviews and longitudinal study

Humphreys: Participant observation and structured Interviews

Dobash and Dobash: Structured and unstructured interviews

Willis: Case study - Unstructured interviews and participant observation
Other sets by this creator