How sociologists investigate society: methods and sampling

Last time

Before we start: last time, in How sociologists approach the study of society, you learned these points. Here they are again on one card. Keep them in mind, because today builds on them.

Finding out

Suppose you want to know why some young people leave school early. You could ask them with a questionnaire, talk to them in depth, or watch them in school. Each method gives different kinds of evidence. This lesson covers the research process, sampling, and the main methods.

The research process

Research follows a series of steps: choose a topic, set a hypothesis or aim, do a pilot study, choose a sample, collect the data, analyse it, and report the findings. The pilot study is a small trial run that tests the method before the real research.

Sampling

The target population is the whole group being studied. A sampling frame is a list of everyone in it, such as a school register. A sample is the smaller group actually studied. It should be representative, so results can be generalised to the whole population. Sampling methods: random, systematic, stratified, quota, and snowball sampling.

Sampling methods

The main sampling methods, and when to use each.

Questionnaires

Questionnaires are lists of questions, often closed, with fixed answers, sent to many people. Strengths: quick, cheap, large samples, easy to compare, reliable. Positivists like them. Weaknesses: low response rates, people may lie or misunderstand, and closed questions give little depth.

Interviews

Structured interviews follow a fixed list of questions, so they are reliable and comparable. Unstructured interviews are like a guided conversation, giving rich, valid detail. Interpretivists prefer them. But interviews take time, and the interviewer's presence can affect answers: interviewer bias.

Observation

In participant observation, the researcher joins the group. In non-participant observation, they watch from outside. Covert observation is secret; overt observation is open. Covert gives natural behaviour but raises ethical problems. Observation gives high validity, but people may change their behaviour if they know they are watched: the Hawthorne effect.

Other methods

A case study looks at one group or event in depth. A longitudinal study follows the same people over many years. Experiments are rare, because it is hard and often unethical to control people's lives. Mixed methods, or triangulation, combine methods to check results and gain both reliability and validity.

Choosing a method

The topic: sensitive topics may need private methods, such as anonymous questionnaires. The researcher's approach: positivist or interpretivist. Practical issues: time, money and access to the group; and ethical issues, such as consent.

Mistakes that lose marks

Mistake one: describing a method without its strengths and weaknesses. Mistake two: forgetting to name a sampling method when designing research. Mistake three: confusing participant with covert observation. The fix: method, sample, strengths, weaknesses.

The full-marks tip

The full-marks tip: when designing research, name the method, the sample and how it is chosen, and why it suits the topic. Give each method one strength and one weakness, linked to reliability or validity.

On one card

So, on one card: research methods. Keep it for revision.