Definitions of core concepts covered as part of the research methods component of AS and A Level Sociology. Organised in alphabetical order – so effectively this is a research methods A-Z. If this is too much for you, then have a look at my ‘top ten research methods concepts‘ first!
Anthropology – the study of humans, past and present. Historically, anthropologists mostly studied traditional (e.g. tribal) cultures using participant observation as its main method, however, more recently anthropologists have increasingly focused much a greater array of aspects of culture within modern and post-modern societies using a more diverse range of methods. One of the key aims of anthropology is to explore and explain the enormous diversity as well as the commonalities within and between human cultures.
Attrition rate – the percentage of respondents who drop out of a research study during the course of that study. This can often be a problem with longitudinal research.
Bias – where someone’s personal, subjective feelings or thoughts affect one’s judgement.
Case study – researching a single case or example of something using multiple methods, for example researching one school or factory.
Confidentiality – the idea that the information respondents give to the researcher in the research process is kept private. This is usually achieved through anonymity.
Covert research – where the researcher is undercover and respondents do not know they are part of a research study. The opposite of covert research is overt research – where respondents know they are part of a research study.
Dependent and independent variables – a dependent variable is the object under study in an experiment, the independent variables are what the researcher varies to see how they effect the dependent variable.
For example, if you grow tomato plants as a hobby and wanted to find out the effect which the amount of water, the temperature, and the amount of light has on the amount of tomatoes each plant produces you could design a series of experiments in which you varied the amount of light etc. and then measure the effects on the amount of fruit produced. In this example, the amount of tomatoes produced is the dependent variable and the water, the temperature and the amount of light are the independent variables.
Ethnography – an in-depth study of the way of life of a group of people in their natural setting. Ethnographies are typically long-term studies (over several months or even years) and aim for a full (or ‘thick’), multi-layered account of the culture of a group of people. Participant observation is typically the main method used, but researchers will use all other methods available to get even richer data – such as interviews and analysis of any documents associated with that culture.
Ethics/ ethical factors – ethics means taking into consideration how the research impacts on those involved with the research process. Ethical research should gain informed consent, ensure confidentiality, be legal and ensure that respondents and those related to them are not subjected to harm. Ultimately research should aim to do more good than harm to society.
Experiments – experiments aim to measure the effect which one or more independent variables has on a dependent variable. Experiments typically start off with a hypothesis, and a good experiment will be designed in such a way that objective cause and effect relationships can be established between variables, so that the original hypothesis can verified, or rejected and modified.
Extraneous variables – undesirable variables which are not of interest to the researcher but might interfere with the results of the experiment.
Field diary – A notebook in which a researcher records observation during the research process. One of the key tools of Participant Observation.
Field experiments – experiments which take place in a real-life setting such as a classroom, the work place or even the high street. See experiments and related terms for a fuller definition.
Focus groups – a type of group interview in which respondents are asked to discuss certain topics.
Formal content analysis – a quantitative approach to analysing mass media content which involves developing a system of classification to analyse the key features of media sources and then simply counting how many times these features occur in a given text.
Going native – where a researcher becomes biased or sympathetic towards the group he is studying, such that he or she loses their objectivity.
Group interviews – where an interviewer interviews two or more respondents at a time.
Hawthorne effect – where respondents alter their behaviour because they know they are being observed. This is one of the biggest disadvantages of overt laboratory and field experiments.
Hypothesis – a theory or explanation made on the basis of limited evidence as a starting point for further investigation. A hypothesis will typically take the form of a testable statement about the effect which one or more independent variables will have on the dependent variable.
Imposition problem – the imposition problem limits the validity of social surveys. It is where respondents may not be able to express their true feelings about the topic under investigation because the questions (and the range of possible responses) which have been pre-chosen by the researcher limits what they are able to say, and may not reflect the issues that respondents themselves feel are important.
Independent variable – see dependent variable.
Informed consent – where the respondent agrees to take part in a research study with full awareness that research is taking place, what the purpose of the research is and what the researcher intends to do with the results.
Interpretivism – an approach to social research which tries to understand human action through the eyes of those acting. Interpretivists want to know the meanings actors give to their own actions, what their own interpretation of their action is. They thus emphasise respondent-led qualitative methods to achieve insight, in-depth explanations and empathy, in order to realise a humanistic, empathetic understanding from the respondents’ point of view.
Interviews – a method of gathering information by asking questions orally, either face to face or by telephone. Interviews can be individual or group and there are three main types of interview – structured, unstructured and semi-structured.
Interviewer bias – where the values and beliefs of the researcher influence the responses of the interviewee. If an interviewer feels strongly about a subject, then he or she might ask leading questions, or even omit certain questions in order to encourage particular responses from a respondent.
Interview schedule – A list of questions or topic areas the interviewer wishes to ask or cover in the course of an interview.
The more structured the interview, the more rigid the interview schedule will be. Before conducting an interview it is usual for the researcher to know something about the topic area and the respondents themselves, and so they will have at least some idea of the questions they are likely to ask: even if they are doing ‘unstructured interviews’ an interviewer will have some kind of interview schedule, even if it is just a list of broad topic areas to discuss, or an opening question.
Laboratory experiments – experiments which take place in an artificial, controlled environment, such as a laboratory. See experiments and related terms for a fuller definition.
Leading questions – questions which subtly prompt a respondent to provide a particular answer when interviewed. Leading questions are one way in which interviewer bias can influence the research process, reducing the validity of data collected.
Life documents – written or audio-visual sources created by individuals which record details of that person’s experiences and social actions. They are predominantly qualitative and may offer insights into people’s subjective states. They can be historical or contemporary and can take a wide variety of forms.
Longitudinal studies – a study of a sample of people in which information is collected from the same people at intervals over a long period of time. For example, a researcher might start off in 2015 by getting a sample of 1000 people to fill in a questionnaire, and then go back to the same people in 2020, and again in 2025 to collect further information.
Likert scale – used to measure strength of opinion or feeling about a statement in social surveys. For example respondents might be asked whether they strongly agree, agree, disagree, or strongly disagree with a particular statement.
Multistage sampling – with multistage sampling, a researcher selects a sample by using combinations of different sampling methods. For example, in Stage one, a researcher might use systematic sampling, and in Stage two, he might use random sampling to select a subset for the final sample.
Non-participant observation – where the researcher observes a group without taking part with that group. This method can either be overt or covert, and data may be recorded quantitatively or qualitatively. Probably the most commonly experienced example of non-participant observation is the OFSTED inspection.
Objective knowledge – knowledge which is free of the biases, opinions and values of the researcher, it reflects what is really ‘out there’ in the social world.
While most sociologists believe that we should strive to make our data collection as objective as possible, there are some sociologists (known as phenomenologists) who argue that it is not actually possible to collect data which is purely objective – the researcher’s opinions always get in the way of what data is collected and filtered for publication.
Official statistics – numerical information collected and used by the government and its agencies to make decisions about society and the economy. Examples include the UK National Census, police recorded crime and data on educational achievement.
Open-ended question – questions for which there are no set answers. Open questions allow individuals to write their own answers or dictate them to interviewers. For example ‘have you enjoyed studying Sociology this year?’
Operationalising concepts – the process of defining a concept precisely so that it can be easily understood by respondents and measured by the researcher. The term may also be applied to the process of determining variables in experiments.
For example, rather than ask a respondent ‘are you religious’, which is a vague question with many interpretations, a researcher might operationalise the concept of religion by using a range of more precise questions such as ‘do you believe in God’, ‘do you believe in the idea of heaven and hell’, ‘ how often do you pray’, and so on.
Overt research – see covert research.
Participant observation – involves the researcher joining a group of people, and taking an active part in their day to day lives as a member of that group and making in-depth recordings of what she sees.
Participant Observation may be overt, in which case the respondents know that researcher is conducing sociological research, or covert (undercover) where the respondents are deceived into thinking the researcher is ‘one of them’ and do not know the researcher is conducting research.
Personal documents – first-hand accounts of social events and personal experiences, which generally include the writer’s feelings and attitudes about the events they think are personally significant. Examples of personal documents are letters, diaries, photo albums and autobiographies.
Pilot study – a test study carried out before the main research study and on a smaller scale, to uncover and iron potential problems which may occur in the main programme of research.
Positivism – an approach to social research which aims to be as close to the natural sciences as possible. Positivists emphasise the use of quantitative data in order to remain detached from the research process and to uncover social trends and correlations which are generaliseable to society as a whole. Their ultimate aim is to uncover the objective social laws which govern human action.
Practical factors – include such things as the amount of time the research will take, how much it will cost, whether you can achieve funding, opportunities for research including ease of access to respondents, and the personal skills and characteristics of the researcher.
Pre-coded, or closed questions – questions where the respondent has to choose from a limited range of responses. Two of the most common types of closed question are the simply yes/no questionnaire and the Likehert Scale (a strength of feeling scale).
Primary data – data collected first hand by the researcher herself. If a sociologist is conducting her own unique sociological research, she will normally have specific research questions she wants answered and thus tailor her research methods to get the data she wants. The main methods sociologists use to generate primary data include social surveys (normally using questionnaire), interviews, experiments and observations.
Public documents – are produced by organisations such as government departments and their agencies as well as businesses and charities and include OFSTED and other official government enquiries. These reports are a matter of public record and should be available for anyone who wishes to see them.
Qualitative data – refers to information that appears in written, visual or audio form, such as transcripts of interviews, newspapers and web sites. (It is possible to analyse qualitative data and display features of it numerically).
Quantitative data – refers to information that appears in numerical form, or in the form of statistics.
Quota sampling – In this method researchers will be told to ensure the sample fits with certain quotas, for example they might be told to find 90 participants, with 30 of them being unemployed. The researcher might then find these 30 by going to a job centre. The problem of representativeness is again a problem with the quota sampling method.
Random sampling – in random sampling everyone in the population has the same chance of getting chosen. A simple example of random sampling would be picking names out of a hat.
Rapport – a close and harmonious relationship between researcher and respondents, such that both parties understand each other’s feelings and communicate well.
Reliability – if research is reliable, it means if someone else repeats the same research with the same population then they should achieve the same results.
In order to be reliable, research needs to be easily repeatable. Self-completion questionnaires have high reliability because it is easy for another researcher to administer the questionnaire again. More in depth methods such as participant observation, where the researcher can spend several months or even years with a small group of respondents are not very reliable as it is impossible to replicate the exact procedures of the original research. More qualitative methods also open up the possibility for the researcher to get more involved with the research process, with further detracts from the reliability.
Representativeness – research is representative if the research sample reflects the characteristics of the wider target population that is being studied.
Representativeness thus depends on who is being studied. If one’s research aim is to look at the experiences of all white male AS Sociology students studying sociology, then one’s sample should consist of all white, male sociology students. If one wishes to study sociology students in general, one will need to have a proportionate amount of AS/ A2 students as well as a range of genders and ethnicities in order to reflect the wider student body.
Research sample – the actual population selected for the research – also known as the respondents.
Sampling – the process of selection a section of the population to take part in social research.
Sampling frame – a list from which a sample will be drawn.
Secondary data – data that has been collected by previous researchers or organisations such as the government. Quantitative sources of secondary data include official government statistics and qualitative sources are very numerous including government reports, newspapers, personal documents such as diaries as well as the staggering amount of audio-visual content available online.
Semi-structured interviews – those in which researchers have a pre-determined list of questions to ask respondents, but are free to ask further, differentiated questions based on the responses given.
Snowball sampling – with this method, researchers might find a few participants, and then ask them to find participants themselves and so on.
Social surveys – typically questionnaires designed to collect information from large numbers of people in standardised form.
Social surveys are written in advance by the researcher and tend to to be pre-coded and have a limited number of closed-questions and they tend to focus on relatively simple topics. A good example is the UK National Census. Social surveys can be administered (carried out) in a number of different ways – they might be self-completion (completed by the respondents themselves) or they might take the form of a structured interview on the high street, as is the case with some market research.
Socially constructed – Interpretivists argue that official statistics are socially constructed – that is they are the result of the subjective decisions made by the people who collect them rather than reflecting the objective underlying reality of social life. For example Crime Statistics do not reflect the actual crime rate, only those activities which are defined as crimes by the people who notice them and who then go on to report those activities to the police.
Stratified sampling – this method attempts to make the sample as representative as possible, avoiding the problems that could be caused by using a completely random sample. To do this the sample frame will be divided into a number of smaller groups, such as social class, age, gender, ethnicity etc. Individuals are then drawn at random from these groups. If you are observing doctors and you had split the sample frame into ethnic groups you would draw 8% of the participants from the Asian group, as you know that 8% of doctors in Britain are Asian.
Structured or formal interviews – those in which the interviewer asks the interviewee the same questions in the same way to different respondents. This will typically involve reading out questions from a pre-written and pre-coded structured questionnaire.
Subjective knowledge – knowledge based purely on the opinions of the individual, reflecting their values and biases, their point of view. See also ‘objective knowledge’.
Systematic sampling – an example of a systematic sample would be picking every 10th person on a list or register. This carries a similar risk of being unrepresentative as random sampling as, for example, every 10th person could be a girl.
Target population – all people who could potentially be studied as part of the research.
Textual analysis – involves examining how different words are linked together in order to encourage readers to adopt a particular view of what is being reported.
Textual analysis also involves the use of semiology – which is the analysis of signs and symbols.
Thematic analysis – involves trying to understand the intentions which lie behind the production of mass media documents by subjecting a particular area of reportage to detailed investigation.
Theoretical factors – validity, reliability, representativeness and whether research is being carried out from a Positivist or Interpretivist point of view.
Positivists prefer quantitative research methods and are generally more concerned with reliability and representativeness. Interpretivists prefer qualitative research methods and are prepared to sacrifice reliability and representativeness to gain deeper insight which should provide higher validity.
Transcription – the process of writing down (or typing up) what respondents say in an interview. In order to be able to transcribe effectively interviews will need to be recorded.
Triangulation – the use of more than one method in social research. For example a researcher might combine structured questionnaires with more in-depth interviews. Triangulation is often used to verify the validity of other data sources and is a good way of improving the reliability of research.
Unstructured interviews – also known as informal interviews, are more like a guided conversation, and typically involve the researcher asking open-questions which generate qualitative data. The researcher will start with a general research topic in and ask questions in response to the various and differentiated responses the respondents give. Unstructured Interviews are thus a flexible, respondent-led research method.
Validity – research is valid if it provides a true picture of what is really ‘out there’ in the world.
Generally speaking, the more in depth the research, the fuller picture we get of the thoughts and feelings of the individuals acting, so the more valid the data; and the more the researcher stands back and allows the respondents to ‘speak for themselves’ the more valid the data. In more quantitative research, such as social surveys, validity may be lacking because the researcher has decided on what questions should be answered by respondents, rather than letting the respondents decide on what they want to say for themselves, as is typically the case with more qualitative methods.
Value Freedom – where a researcher’s personal opinions, beliefs and feelings are kept out the research process so that data collected is not influenced by the personal biases of the researcher.
Verstehen – a German word meaning to ‘understand in a deep way’ – in order to achieve ‘Verstehen’ a researcher aims to understand another person’s experience by putting themselves in the other person’s shoes.
Interpretivists argue that to achieve Verstehen (or empathetic understanding) we should use in-depth qualitative research such as participant observation.