OCR Sociology: Research Methods

Validity

How TRUE a study is and how accurately is represents those being studied.

Underminded if:

- subjects lie

- subject changes behaviour due to anxiety

- researcher misinterpretation

Reliability

Ability of a test to yield very similar scores for the same individual over repeated testings

Representativeness

The idea that the sample mirrors characteristics of the greater population

Positivism

Belief that social behaviour and society are products of social forces beyond our control, originating in the structure of society

Positivism: Patterns and Trends

Research should be standardised, systematic and logical and all those who participate should be equally exposed to the same independent variable to uncover causes for behaviour

Value Freedom

The idea that all sources of bias have been eliminated

Interpretivism

Belief that human behaviour cannot be predicted by social patterns and forces over which people have no control. People are architects of society and make their own choices.

Verstehen

Empathy via a rapport, stresses validity over reliability

Researcher Imposition

Where the researcher decided what is and isn't important based on own experiences, evident in positivist research.

Subjectivity

The establishment of a rapport may create a subjective relationship, overcoming objectivity and introducing bias and selective interpretation

Factors influencing research

COST - research is expensive

FUNDING - hard to find and founders may dictate nature of research

TIME - longer studies allow for better rapport

NATURE OF SUBJECT MATTER - some subjects are sensitive and may rule out certain methods

CHARACTERISTICS OF RESEARCHER - white, middle class researcher unlikely to develop rapport with street gang

7 Stages of Research

1 - decide on a topic

2 - READ around the topic and others' findings

3 - Form a HYPOTHESIS based on knowledge

4 - decide whether to use PRIMARY or SECONDARY data

5 - OPERATIONALISE concepts so that they can be observed and measured

6 - chose data type (QUANTITATIVE or QUALITATIVE)

7 - PILOT studies

Pilot Studies

A small scale run through of an investigation to find any strengths and weaknesses. Determines if questions are clear, if they upset respondents, if right subjects have been identified, if interviewers are well trained and objective and if data produced is useful.

Longitudinal Problems

- respondents may drop out or die

- views of remaining mag not be representative

- researchers may become emotionally close, losing objectivity

Interpretation of Data

POSITIVISTS believe all data (even contradictory) should be included to reduce objectivity.

Quantitative data is more objective because qualitative is subjective

Sociology and Social Policy

Two views:

1 - sociologists should conduct research then hand it over to the policy maker to interpret, making them only OBJECTIVE PURSUERS OF FACT.

2 - sociologists should take responsibility for how their work is translated into social policy. Sociology is to OBSERVE AND CHANGE.

Random Sampling

Names picked out of a hat (sampling frame)

Systematic random sampling

The process of selecting a sample of subjects for a study by drawing every nth unit on a list

Stratified random sampling

Population divided into subgroups (strata) and random samples taken from each strata

Snowball sampling (non-random)

Getting directed between people

Volunteer sampling (non-random)

Involves advertising for people who fit certain criteria to get in touch (newspaper adverts etc.).

Participants may have their own agenda to skew results.

Opportunity sampling (non-random)

Involves the researcher asking people who happen to be about and 'fit' the criteria to take part

Purposive sampling (non-random)

Choosing specific people with certain characteristics

Quota sampling (non-random)

Targeting certain amounts of certain people who fit a quota

Research Ethics

- Informed consent

- privacy and confidentiality

- protection from harm

- legality and morality

Social desirability and demand characteristics

The subject says things based on what they think the researcher wants to hear

Impression management

Questions reflect what the sociologist thinks is important despite having not much contact with the group being studied