Share on Facebook
Share on Twitter
Share on Google Plus
  • Join over 1.2 million students every month
  • Accelerate your learning by 29%
  • Unlimited access from just £6.99 per month

Main methods

Questionaires, interviews, secondary data. Learn about the different methods and gain a deeper insight with our explanations.


A questionnaire is simply a list of questions. There are two basic types of question: open questions – where the respondent can express their answer in their own words; and closed questions – where the question has a fixed set of alternatives. Typically questionnaires use large samples and generate quantitative data.

Questionnaires can be completed using two different approaches. Self-completion questionnaires are filled in by the respondent themselves, either on paper, online or using a laptop or tablet provided by the researcher. Alternatively structured interviews can be used where an interviewer asks the questions.

In structured interviews the interviewer is able to clarify questions and note how the respondent responds to the questions. However, there is the possibility of interviewer bias (see below). If the issue is sensitive then self-completion may encourage the respondent to be more truthful. In the more sensitive parts of the Crime Survey for England and Wales, the interviewer will give a laptop to the respondent to use so they do not see the answer.


An interview is a conversation with a purpose – for the interviewer to find out something from the respondent. Interviews vary is style. Structured interviews use a questionnaire with mainly closed questions and produce quantitative data from large samples. This approach is favoured by positivist sociologists who argue that sociology needs to take a more scientific approach.

Semi-structured interviews are more like guided conversations. The interviewer may have a list of topics to cover and some idea of specific questions but they have the flexibility to make each interview unique. This type of interview can be time consuming so samples are often quite small. Semi-structured interviews produce qualitative data and are favoured by interpretive sociologists who argue that research should uncover the meanings people give to their own and other’s actions.

Unstructured interviews are a more extreme version of semi-structured interviews. The respondent sets the agenda and can talk about any issue that is important to them (although usually the interview will have a broad theme).

Interviews are social situations – they involve more than one person interacting. This means that the parties involved in the interview may influence each other at any point. For example, the interviewer’s appearance and language may affect the respondent’s answers – they may not want to appear ignorant, intolerant or selfish for example. This interviewer bias can affect the validity of data.

Group interviews or focus groups involve several people discussing an issue together, usually with a researcher chairing the discussion. The idea is that these discussions more accurately reflect real-life conversations and that there is less chance of interviewer bias. However, the discussions can be affected by the various personalities in the group so not everyone’s opinions may be heard and some may be influenced strongly by the contributions of others.

In the case of semi- or unstructured interviews the interaction between the interviewer and respondent is used positively to improve the validity of interview data. The interviewer prompts and probes for more detail and changes their questions in the light of responses. Respondents talk at length and in the way that makes sense to them.

In contrast, the role of the interviewer in structured interviews is to be a ‘data-collecting machine’. To ensure reliability every question is asked in the same way to every respondent and every interview follows the same format, often using closed questions.


Watching the behaviour of groups of people is perhaps the oldest form of social research. Observation can be quantitative with an observer adopting a ‘fly on the wall’ role, keeping their distance from the situation and noting the number of times certain events or behaviours occur. However, the most well-known type of observation in Sociology is participant observation. In this method the researcher joins in with the group or situation they are studying and experiences it as a participant. Advocates of this method argue that the key to understanding social life is to understand how it actually feels to be part of a group or situation. The German sociologist Max Weber (1864-1920) called this kind of empathic understanding ‘ verstehen’.

An essential part of participant observation is to gain access to the group or situation. Sometimes this is achieved by the researcher building a relationship with a ‘key informant’ – an existing member of the group who can provide the researcher with credibility and contacts. Once in the group the researcher needs to adopt and maintain a role throughout the period of research, which can last for years.

Participant observation can produce vivid, rich and unique qualitative insights. However, it is almost impossible to check the data collected so there are question marks about the reliability of data collected in this way. There are also ethical issues around deception if the researcher remains covert (hidden) and personal bias can also develop as the researcher gets more integrated into the group. If the researcher is overt (open) then there is the question of validity – has the presence of a researcher affected behaviour?

Secondary data

Secondary data is the term given to describe existing information that a sociologist might use as evidence. Secondary data can take a wide variety of forms, both quantitative and qualitative. Official statistics are quantitative data produced by local and central government for their own purposes. They include data derived from the Census that provides information about every household in England and Wales. Official statistics cover much larger samples than is normally possible in sociological research but are collected by the state for its own purposes so are not always in the most useful form for researchers. Some data, such as crime statistics need to be treated with particular care as they rely on public reporting and police judgements.

Qualitative forms of secondary data include personal and historical documents such as letters and diaries. These can provide a vivid picture of an individual’s experiences and feelings but they need to be treated with caution as they are subject to personal biases and cannot usually be checked for authenticity and accuracy.

Content analysis

Content analysis refers to a range of techniques for analysing the content of secondary data such as mass media texts. These can be analysed quantitatively by counting up the number of times something is mentioned or particular words are used; or qualitatively by analysing the way language and visual images are utilised to communicate meaning. Content analysis is often used to analyse bias, distortion and stereotyping in the media.

Mixed methods

Many pieces of research use a mixture of methods. In the approach known as methodological pluralism methods are combined to balance out the type of data collected – a mixture of quantitative and qualitative methods may be used to provide a balanced picture of the issue under investigation.

In order to check the validity and reliability of data an approach known as triangulation is often used. Here the data produced by one method is checked against data collected by a different method. For example the findings of a participant observation study might be checked by conducting a survey to see if the patterns identified in the observation are apparent when a wider sample is used.