Sociology - Methods (G674)

  • Created by: SophieMae
  • Created on: 14-01-14 10:40


  • 3 types - structured, semi-structured, unstructured. May also be a group interview, including focus groups (see later card).
    • Unstructured (and group)  = open questions.
    • Structured = closed questions.
    • Semi-structured = open and closed questions.
  • Open questions allow probing and exploring = qualitative data, favoured by interpretivists = depth, detail, meaning and insight = validity.
  • Closed questions = quantitative data, favoured by positivists = numerical, statistical, trends and patterns.
  • Where conducted? - familiar place = respondents relaxed so give honest responses = validity.
  • Rapport - establish one to make respondents feel relaxed = validity.
  • Ethics - CAPRI - confidentiality, anonymity, permission (under 18), right to withdraw, informed consent. If BSA guidelines are followed respondents should feel relaxed and comfortable and give open, honest answers = validity.
  • By whom? - social characteristics - if not the same then may not feel comfortable and not give honest answers = validity.
  • Values of interviewer - may affect selection of data for results - bias.
1 of 33


  • Social desirability - the respondent could give responses they think the interviewer wants to hear.
  • Body language - interviewer can observe this and compare with answers of the respondent (verbal cues) and a judgement can be made on truthfulness of answers = validity.
  • Pilot - allows checks for clear and ambiguous language, that the sample can understand, allow interviewer to practice skills (particularly if unskilled) = reliability.
  • How recorded? - can be notes but this may be off-putting for the respondent and interviewer cannot write it all down so will select key points (bias).
  • Better to audiotape - allows interviewer to focus on non-verbal cues, taping adds to the reliability as it allows for rechecking, by the original researcher or another one, for misunderstadings/misinterpretations. Taping may initially affect answers as they may feel uncomfortable but interviewees soon forget the tape giving honest answers.
  • Videotaping is the best method - allows rechecking of both verbal and non-verbal cues.
2 of 33

Group Interviews/Focus Groups

  • A group interview may be quite wide ranging whereas a focus group is usually concerned with a specific topic. Organised group discussion about an issue chosen by researcher.
  • Researcher is the facilitator of discussion - guides it, normally with a list of issues for discussion.
  • Gathers qualitative data - favoured by interpretivists.
  • Researcher can see the ways in which members change their views when being challenged by others - can see how strongly held their views are.
  • Participants can concentrate on issues they consider to be important as they have more control over the discussion.
  • Focus groups can throw up issues that the researcher may not have thought of.
  • Researchers can also geta realistic insight into the views of the participants as they can challenge and argue with each other.
  • Focus groups are thought to be more 'naturalistic'.
  • Reduce interviewer effect of other traditional interviews.
  • Allows access to a larger sample in a short space of time - represenative - generalisations.
3 of 33


  • Advantages
    • More people can be involved.
    • Interview more people in less time.
    • Members of the group can bounce ideas off each other - less artificial situation - like a conversation.
    • Groups empower people.
    • Researchers can obtain in-depth information.
    • Participants can reflect and rethink what they say, potentially making the data more valid.
    • Participants more relaxed - more likely to give open and honest answers - validity.
  • Disadvantages
    • Can be difficult to record.
    • Dominant participants may 'take over'.
    • Participants may say what they think the researcher (or their peers) wants to hear.
    • Difficult to quantify results - reliability and cannot verify individual responses - validity.
    • Could divert away from research area.
    • Not reliable - open questions - cannot be replicated.
4 of 33


  • Can contain open and/or closed questions.
    • Open - interpretivists - depth, detail, insight and meaning - qualitative data - harder to analyse.
    • Closed - positivists - numerical, statistical, trends and patterns - quantitative data - easier to analyse as they can be put into a computer.
  • Quick, cheaper method.
  • Ethics and BSA guidelines followed? - validity.
  • People take them away - can be influenced by people at home or soemone else could fill in - confering - validity.
  • May not get the questionnaire back if taken away - affects response rate - representativeness?
  • If filled out in front of a researcher - may feel pressured to answer in a certain way - validity.
  • Reliability - another researcher using the same method will get similar or same results - only if questionnaire contains all or mostly closed questions.
  • Can be used in conjuction with other methods - fuller picture - validity.
  • Mailed ones can be distributed across a wide geographical area - improve representativeness of a sample.
  • Questionnaires are socially constructed - values of the researcher in the questions?
5 of 33

Pilot Studies

  • Small scale prelimiary study conducted before main research to check the feasibiltiy or to improve the design of the research.
  • Often used when designing questionnaires but can be used with most methods.
  • Allow the researcher to identify and resolve potential problems such as sample and access, method of distribution, wording of questions, potential misunderstandings, language used, if you are collecting the data needed to answer your aim.
  • Researcher may ask those who took part for feedback as an additional way to recognise any issues.
  • Saves time and money by sorting out any problems before the main research is carried out - if a pilot is not carried out and there are issues then the whole project would have to be scrapped causing a waste of time and money.
  • Carried out on members of relevant population but not those who will form part of the final sample - may influene the later part of the research - know what to expect etc.
  • Test the questions for ambiguity, see if data collected fulfils aims, to see if the sampling frame is suitable, help researchers develop ways of getting the full cooperation of those being studied (Gavron - necessary to establish a rapport - spent time chatting informally prior to interviews), develop research skills of those taking part (Rex & Tomlinson - trained amateur interviewers during pilot studies) and to ultimately decide whether or not the research goes ahead.
6 of 33


  • The selection of a group to study from the target population.
  • Various factors need to be considered when selecting a sample - identification of the targe population, identifying what sampling frames are available, finding how to access the sampling frame and target pop, understanding who is included in the sampling units, deciding what size the sample should be and working out what sample strategy to use.
  • Sampling frame - list of all those in the target population (e.g. Electoral Register, Postcode Address File.)
  • Sampling population - everyone in the group being studied.
  • 2 types of sample = random - simple random, systematic random, stratified random. Non-random - volunteer, opportunity, snowball, quota, purposive.
  • Random sample - everyone in the population has an equal chance of being selected, seen as objective and scientific, used by positivists who want a representative sample.
    • Systematic random - making one random choice of a starting point in a list and then selecting people on a systematic basis (every nth name for example).
    • Stratified random - sampling a stratum in a group/catergory that has characteristics in common (e.g. males/females, ages, social class).
    • In many research situations it is not possible to draw random samples as there is no obvious sampling frame to be used.
7 of 33


  • Cluster sampling - the target population is sub-divided, then a random sample is selected from the sub-divisions, then a further sample from those samples and so on until a final sample of the required size is achieved.
  • Non-random - could be argued to be subjective, used by interpretivists who are more interested in finding specific types of individuals/groups than ensuring representativeness.
    • Opportunity - who is there at the time, asking people who happen to be about and fit the criteria to take part.
    • Volunteer - those who volunteer to take part, often involves advertising for people who fit the criteria to get in touch.
    • Purposive - researcher deliberately seeks out those who meet the needs of the project.
    • Snowball - initial sample asked to name others who might fit the criteria, often used if a group is difficult to
    • Quota - populaiton divided into catergories and researcher given instructions as to how many people they need from each catergory to fit the quota.
  • Random sampling eval - can be repeated and a similar group will probably emerge (reliable), if the sample is large enough it is likely to be representative and therefore it is safe to generalise. However, time consuming and if the sampling frame fails to provide relevant info such as age or ethnicity, stratified sampling is not possible = biased samples may be produced.
  • Non-random sampling eval - useful when no sampling frame available, when a specific type of person/group is required or if a group is difficult to reach. However, unlikely to be representative.
8 of 33


  • The need for sociologists to transform a research question or hypothesis into something that can be measured through research.
  • Subject matter will influence the choice of research method.
  • Chosen hypothesis needs to be a clear statement which makes a casual link between two variables.
  • Research population often needs to be operationalised in terms of social class, age, gender and ethnicity. Also, need to be precise when identifying the population.
  • Social class is often central to a research hypothesis and it is important to make clear how you operationalise this concept. Used to use job type as the basis for social class but since 2000 social class has been operationalised by asking questions related to the nature of their employment.
  • Ethnicity requires careful thought too. For example, if you study 'Asians', you need to be precise in terms of country of origin, religion etc. Terms like 'black' and 'white' require some discussion to make your precise intentions clear.
  • Hypothesis - needs to be broken down into measurable components. If any sociological terms have been used these need to be clearly defined and questions constructed which measure aspects of them.
9 of 33


  • Also need to think through how you might ask about sensitive topics (e.g. racism). Such topics might require the use of attitudinal scales or it might simply be a list of +ive and -ive statements which you ask respondents to tick.
  • Need to state clearly what you are hoping to achieve by using these particular statements and how they relate to your hypothesis.
  • If you are using an observation schedule, it is important to justify the catergories you are hoping to observe. This may involve the construction of 'tally sheets' or diagrams.
  • If you are using a content analysis schedule or semiotic analysis it is important that the catergories or interpretations that you apply to media language or image operationalise your hypothesis accurately.
10 of 33

Triangulation/Methodological Pluralism

  • Using more than one method. Usually one of the methods balances the disadvantages of another.
  • Triangulation - cross checking the data for reliability and validity. Reliability - another method to make the results easier to replicate by another researcher. Validity - compare responses from different methods to see if they complement/support each other. E.g. Denzin (1970).
    • Investigator Triangulation - use of different researchers - checks for observer and interviewer bias.
    • Data Triangulation - collecting data at different times from different people in different places. Can also involve combining primary and secondary data.
    • Methodological Triangulation - 2 forms - within method triangulation = uses a variety of techniques within the same method (e.g. open and closed questions). Between method triangulation = combination of a number of research methods.
  • Methodological Pluralism - to get a fuller picture of the research. Recognises that each method and type of data has its particular strengths and weaknesses. Combined they are seen to produce a more comprehensive and rounded picture of social reality. Can also provide new insights and new directions for research. E.g. Eileen Barker's study of the Moonies (1984).
11 of 33


  • Non-participant and participant.
    • Non-participant - researcher is not part of the group.
    • Participant - researcher observes from within the group.
  • Covert or overt (key variation of participant observation).
    • Covert - undercover - the group does not know the true reason for the person being there - the ethics can be questioned BUT sometimes justified.
    • Overt - open and honest about what is happening, what the research is about etc.
  • Advantages of overt observation;
    • allows the observer to ask questions,
    • observer can retain some detachment,
    • researcher doesnt have to lie,
    • no risk of being uncovered, bringing the end to the research.
  • Disadvantages of overt observation;
    • observer may influence the subjects' behaviour,
    • some groups will not accept being observed,
    • difficult to become full participant,
    • harder to understand the subjects' behaviour.
12 of 33


  • Adantages of covert observation;
    • respondents may act more naturally,
    • difficult to access some groups,
    • observer has same experiences and may understand the group better.
  • Disadvantages of covert observation;
    • unethical to mislead subjects,
    • difficult to opt out of illegal/immoral activities,
    • observer may go native - become too much one of the group and lose objectivity.
  • In non-participant observation the Hawthorne effect could be shown - pople act differently because of the presence of an observer - uncomfortable etc. However, over the period of time people are likely to forget the observer and become more relaxed.
  • It is the observer's interpretations and judgements - subjective. In order to make it more objective, more than one observer could be used and then compare notes/ideas.
  • Observation is favoured by interpretivists - gaining qualitative data.
13 of 33

Mixed Methods

  • Using more than one method.
  • Improve quality and validity of the data.
  • Gain a fuller picture.
  • Qualtitative and quantitative data - mixing positvists and interpretevists.
  • Realist approach - mixing and matching theoretical viewpoints.
  • Cross checking for validity and reliability.
  • Advantages include;
    • The possibility of adding narrative to numbers. This can help add meaning to the numbers, rather than relying completley on quantitative data.
    • Numbers can be used to add more precision to qualitative data.
    • Strengths of qualitative data are combined with those of quantitative data.
    • Broader range of research questions can be explored - can explore how and why (qualitative) rather than just what/how many.
    • Strengths of one method can be used to overcome weaknesses of another.
    • Mixed methods research is often more generalisable and it is often possible to collect more data because of the benefits of using methods that complement each other.
14 of 33


  • Disadvantages include;
    • Costly in time, resources and money.
    • Demands high levels of skill in the collection and analysis.
    • Researcher must ensure that the methods will complement each other, which often makes the research design process very demanding.
    • Careful and skillful data analysis is needed.
    • Due to the large amount of collected, the research can become too complicated and fail to answer the research questions.
    • Mixed methods research has in the past been unpopular with social policy makers who prefer the clarity of one method approaches and results BUT this may have changed in recent times.
15 of 33

Case Studies

  • This is a detailed study of a particular group or organisation - focuses on one group.
  • Very detailed, gives in-depth information but the intense scrutiny may miss wider issues due to its concentration.
  • May stand alone or be part of a large piece of research.
  • Once a 'case' has been identified then methods can be selected - a case study is NOT a method.
  • Use a variety of methods and many are based on the colleciton of qulaitative data such as observation or unstructured interviews, others use a combination of quantitative and qualitative data to gain as wide-ranging as insight as possible of the 'case' being studied.
  • Bryman says when a researchers uses one method it cannot be described as a case study.
  • McKee and Bell (1985) - studied a small community to explore the effect of high rates of unemployment on family relationships.
  • Blackman - the youth underclass - qualitative study of homeless young people in Brighton.
    • Kept a field diary and wrote it up as a series of narratives. The field diary would have included as much as possible of what he observed and the conversations he had with the youth. He wanted the narratives to "allow a view inot the personal and subjective world of the youth."
16 of 33


  • Advantages
    • Can act as a start to more wide-ranging research and might throw up facts that need further research.
    • Can be used to generate a hypothesis.
    • Provide an insight into a 'case' which other might find difficult to access (e.g.deviant groups).
    • Can be conducted by a single researcher.
    • Tend to be high in validity - range of methods means data can be cross-checked and time spent can produce in-depth data.
    • Researcher can develop a rapport.
    • Possible to gain verstehen - likely to spend a long time with the individuals or groups and develop a sensitivity to their situation.
    • Where quantitative data is collected = reliable.
  • Disadvantages
    • Unlikely to be based on a representative sample and so it is not safe to make generalisations.
    • Platt argues there is a danger that even experienced researchers begin to make generalised comments when it is not safe to do so.
    • Difficult to repeat the case study due to the close relationship usually developed.
    • Risk of the over involvement of the researcher - 'going native'.
17 of 33

Longitundinal Studies

  • Carried out over a period of time (usually years) in order to study changes or developments over time.
  • Two types - panel and cohort studies.
    • Panel - data is collected from a sample selected from sampling frames (e.g. the Postcode Address File).
    • Cohort - based on people with the same social characteristic, often their age.
  • Data usually collected by questionnaires or interviews.
  • Many use a mixed methods approach (see earlier card).
  • Some participant observation studies take 3/4 years - could therefore be described as longitudinal studies.
  • Cost of the research and analysis is high - selection of initial sample is extremely important and often involves a number of complicated stages.
  • Decision about how long the research will be financed for needs to be made, as well as how often the data should be collected.
  • Decision needs to be made regarding what to do about attrition - members of the sample dropping out of the research or becoming uncontactable.
18 of 33


  • Advantages
    • Changes in opinions/attitudes can be tracked over time.
    • Correlations can be identified.
    • Possible to make comparisons with data from other studies.
    • Data from a study can be used to inform policy making.
    • Data can be qunatitiative, qualitative or both.
    • In-depth can be obtained - validity.
    • A relationship of trust can be established as the respondents feel at ease with researchers.
  • Disadvantages
    • Time consuming and expensive to run.
    • Access to an appropriate sample is sometimes probelmatic.
    • Attritions will occur - lead to a distorted sample and skewed results.
    • If attrition is high - representativeness may be lowered and generalisations cannot be made.
    • Researchers could go native or become too close to their sample.
    • Hawthorne effect can occur as respondents build relationship with researcher. They know waht might be expected and either change their behaviour.
    • Attrition of researchers - people change job or retire.
19 of 33

Secondary Data

  • Second hand data collected by somebody else, as opposed to primary data which is collected by the sociologist.
  • Seen to have equal status with primary data.
  • Can be just as difficult to collect and interpret as primary data.
  • Bryman - 5 catergories:
    • Life documents - diaries, letters, photos, video recordings, email - may be written down or just visual/aural.
    • Official documents - from the state/government, official statistics/reports etc.
    • Other documents - other organisations, charities, profit-making organisation etc.
    • Content of the media - articles, documentaries, internet, magazines.
    • Previous sociological research.
  • Webb et al - also identified trace measures - physical changes produced by human action.
  • 3 main approaches to secondary data:
    • Extraction - taking the statistics/research from the orginal texts.
    • Content Analysis - documents/sources examined in great detail to find themes - qualitative and quantitative.
    • Semiotics - study of signs. 
  • Scott - there are 4 criteria to use when assessing the quality of secondary data - authenticity, credibility, representativeness and meaning.
20 of 33


  • Advantages
    • Saves time if the information required already exists - researcher does not have to repeat reserach.
    • The researcher may use the original data to re-examine previously publised data or studies in order to interpret them in a new theoretical light.
    • Useful when sociologists want to look back in history but there is no one available/able ro provide a life or oral history. 
    • Sometimes it is difficult to access a group directly.
    • There are groups engaged in activities they do not want sociologists to study.
  • Disadvantages
    • Could be biased or distorted or contain error.
    • Gov't statistics are often constructed in a way to throw a positive light on events/statistics.
    • Private organisation are concerned to produce a positive image of themselves.
    • Life documents give a very one sided view of what happened and are almost always bound to be sympathetic to the writer.
    • Historical sources contain the possibility of bias. May - influenced by particular historical events or cultural ways of thinking.
21 of 33

Official Statistics

  • Figures collected by the government or state agencies. Quantitative data gathered through surveys by both national and local government agencies.
  • Divided into two - registration data and survey data.
    • Registration data - recording of sociologically important events such as births, deaths, marriages and divorces.
    • Survey data - gathered by social surveys which happen at particular times. Often representative samples. E.g. The Census - every 10 years. General Household Survey - annual information.
  • Advantages
    • Easy and cheap to access.
    • Up to date.
    • Surveys - huge, carefully constructed representative samples - generalisations.
    • Positivists - see them as 'hard' reliable facts as they've been collected in a standardised, systematic and scientific fashion.
    • Trends can be identified - comparisons can be made year on year.
    • Comparisons can be made between social groups.
    • Quantitative - reliable.
    • Effect of legislation can be tracked. Can inform/support social policies.
22 of 33


  • Disdavantages
    • May only give a partial picture of a sociological problem.
    • May tell us more about the people involved rather thean the social trend they claim to describe.
    • Surveys are very one dimesional in thier validity - tell us very little aout the interpretations/experiences that underpin them.
    • Open to political abuse - can be manipulated or 'massaged'.
    • May be based on operational definitions that sociologists would not agree with.
    • No chance to prompt and probe - low validity.
    • Do not give a full picture of some topics (e.g. crime statistics only show reported crimes).
23 of 33

Content Analysis

  • Recognised way of studying media and media messages.
  • Can be quantitative or qualitative.
  • Quantitative (or formal) CA:
    • Involves counting how frequently a particular item appears in the media under review.
    • E.g. how often boys/girls are presented in stereotypical ways in reading schemes (Best & Loban)
    • Before the media can be quantified in this way the researcher has to devise a CA grid so that the data can be recorded systematically and objectively.
  • Thematic CA:
    • Researcher looks for themes common to the catergories devised in the quantitative analysis.
    • E.g. boys presented as - more adventurous than girls, physically stronger etc. Girl presented as - caring, more interested in domestic matters etc.
  • Qualitative CA:
    • Textual analysis which investigates the latent and hidden meanings. Often called semiotics.
    • Emphasises the importance readers attach to signs within the text or imagery.
    • The central concept of semiotics is that a sign represents something - the sign symbolises something else and is understood to have a meaning to others.
    • Tend to favour one interpretation.
24 of 33


    • Sociologists have used CA of the media to decode these images. Two forms of codes - denotative and connotative.
    • Denotation tends to be described as the literal, obvious or common sense meaning of a sign.
    • Connotation tends to be the associations of a sign.
    • Many sociologists, but not exclusivley feminists, have used the method of CA to decode the messages being used in magazines, TV programmes and advertisements.
  • CA is a useful tool in helping sociologists to research media content.
  • Can be used in a quantitative or qualitative way but normally the two will be combined so that both types of data are collected, giving a fuller picture.
  • CA is often criticised for ignoring the audience - it is the researcher's own interpretation - subjective.
  • To answer this criticism researchers can question/interview the audience to see what messages they are recieving/interpreting. Adds methodological pluralism/triangulation.
25 of 33

Interpretive Approaches

  • Bryman - interpretive sociology sets out to understand varieties of human behaviour by empathising with it.
  • Weber used the term verstehen to describe the process of looking through the eyes of the people involved.
  • Opt for methods that give qualitative data because:
    • uncovers meaning of social action - only by understanding how individuals buils their patterns of interaction can a full understanding of society be presented.
    • contexts - studies small scale groups and specific situations so researcher can preserve the individuality of each - relates to specific circumstances.
    • unanticipated phenomena and influences - researcher doesn't necessarily have clear idea o what they are looking for - start with an interest in certain area but no idea where it might lead. A grounded theory emerges from the actual process of the research. More 'open' to ideas than a positivist approach.
    • Process - interested in the dynamics of the situation - the 'process' rather than the outcome.
  • Most common methods:
    • observation, qualitative interviews, focus groups, secondary sources.
    • Note - ethnography - study of a way of life - usually observation and in depth interviews to get in depth understanding of the group being studied.
  • Interpretivists argue:
    • Human beings are not inanimate - cannot be studied in the same way as phenomena in the natural world.
    • Not possible to identify cuase and effect because human behaviour is unpredictable.
26 of 33


    • Methods of natural sciences are not appropriate when researching the social world - social world is about meanings rather than social facts.
    • People have subjective understandings of the structures of society.
    • Findings are bound to influenced by researcher's values - not possible to be value free.
    • Social world and researcher affect each other.
  • Also argue that the social world does not exist outside of the social realities people construct.
  • Important to gain a rapport with the people being researched and then the data will increase in validity.
  • Want a deep understanding and to be reflective. Reflexivity - researcher standing bak and considering the research from own perspective and as far as possible from the perspective of those being researched.
  • There is a need to identify the group or issue to research and to work with a gatekeeper or use another way to gain access to the group or individuals. Also, obtain qualitative data and identify where the research will take place.
  • Evaluation:
    • May be affected by the Hawthorne effect.
    • The social characteristics of the researcher may affect the research.
    • Important to know exactly how the researcher affects those with whom they are interacting.
    • Can be time consuming to collect qualitative data.
    • Vital to keep good fieldwork notes.
    • May be subjective and low in reliability - difficult to replicate.
27 of 33

Positivist Approaches

  • Argue the follwing points:
    • There are social facts that can be observed and studied. In addition, casual relationships can be sought.
    • Social facts are external to individuals.
    • Society is more important than individuals.
    • Individuals are subject to external forces such as capitalism or competition in the education system.
    • The social world is predictable.
    • It is possible to study the social worldd in a logical and coherent way.
    • It is possible to seek laws that govern people's behaviour.
    • Research questions and hypotheses can be tested.
  • Want to do the following in conducting research:
    • Identify patterns and trends.
    • Look for cause and effect.
    • Identify what is predictable.
    • Make comparisons over time and from one group/culture to another.
    • Look for correlations.
    • Gain data that is reliable.
    • Conduct research that is objective and value free. Also, large-scale data collections/sample - representativeness.
28 of 33


  • Positivists make a number of decisions prior to research:
    • Choice of topic - depend on funding source, may not be a free choice.
    • Operationalising key ideas so that they can be measured.
    • Hypothesis of research question.
    • Identify sampling technique.
    • Choose method that will gain quantitative data.
    • Deciding whether or not to do a pilot study.
    • Identifying location.
    • How to record and analyse data.
  • Methods used:
    • Surveys, experiments, content analysis, comparative methods and offical stats.
  • Produce numerical data and deriving graphs, charts etc from this to identify trends and patterns.
  • High in reliability and often large scale - representativeness is possible - generalisations.
  • Evaluation:
    • Argued that it is impossible to be objective and value free.
    • All variables in the social world cannot be controlled.
    • Ignores meanings and interpretations - low validity.
29 of 33

Realist Approaches

  • Bridges the gap between postivism and interpretivism.
  • Argue that sociological research can follow logic and methods of natural sciences BUT they work with a different interpretation of science from positivists, distinguishing between open and closed systems of science.
  • Natural science = closed system - generate hypotheses, make predictions, collect empirical data.
  • Social science = open system - causal mechanisms underlie regularities, but these regularities are rarely observable and can be subject to rapid and unpredictable change.
  • Sociological research cannot be value free - always linked to theory to some extent.
  • Sociologists should try to gather data which is reliable and transparent BUT also has to access and explore the subjective dimensions of an individual's social experiences.
  • Realists are not committed to positivism or interpretivism - use methods that fit their purpose, often opting for mixed methods.
  • Realism is a contested concept in social research - can be used in relation to philosophical issues and has different interpretations.
  • Undertaking research relalists want to understand structural mechanisms, be as objective, systematic and logical as possible and uncover underlying causal mechanisms that lead to observable events.
  • Argue that research cannot be reliable. Churton - science claims to be reliable BUT is rarely tested.
  • No way of testing realist theories because they argue that sociological structures are like open system in science - cannot be observed.
30 of 33


  • Immersing themselves in the lives of the people under study, gain in-depth understanding.
  • Purpose - describe the culture and lifestyle of the group of people being studied.
  • Key decisions - extent of involvement and amount of info the sociologist gives the group.
  • Extent of involvement:
    • External observer with no contact with the group (non-participant observation) or complete immersion in the group (participant observation).
    • In making this decision - decide what they wish to obtain from their research and weigh up the advantages and disadvantages of the role they adopt.
    • What is possible? What is ethically correct? What method will produce the most valid results? Will becoming a full member of the group actually improve the quality of the research?
  • Amount of info:
    • Completely honest (overt observation) or they can tell the participants nothing (covert observation).
    • What is possible? What is ethically correct? What method will produce the most valid results?
  • Gold suggests that the result of making these decisions can lead to the researcher taking 1 of 4 roles:
    • Complete participant - fully functioning member acting in a covert role.
    • Participant as observer - joins in as a participant but is overt about their role.
    • Observer as participant - mainly there to interview and observe.
    • Complete observer - simply observes what is going on around them, no attempt to interview or discuss.
31 of 33


  • Making contact/gaining access - most researchers use a gatekeeper/key informant who opens the door for them in deviant groups. But not all groups are deviant and many researchers simply ask colleagues if they can study them or get a job or joing a society where they can observe people.
  • Acceptance - role and relationships. Role refers to the decision to be covert or overt, most will adopt the role that gives them the greatest change of getting the research material they want, may be limited by ethical issues regarding how much harm may be caused by acting covertly. Relationships refers to the similarities and differences between the researcher and the group being studied.
  • Recording the activities - particularly problematic for researchers engaged in covert observation. Number of answers - remember as much as possible and write this up as soon after the events as possible (adv - pay full attention, disadv - bound to forget things which affects validity) or make notes wherever possible as the action is unfolding (adv - great accuracy, disadv - disrupt normal social interaction). Selective in what to note down - subjective - affects validity.
  • Getting at the truth - influencing the group/being influenced by the group - hard to remain objective, sociologist's feelings almost always slip into the field diaries and notes, closer to the group the researcher gets the more likely it is that bias will creep in. 
  • Leaving the group - when to leave - Glaser and Strauss - correct time to leave is when new information does nothing buy confirm what the sociologist has already found out (theoretical saturation). How to leave - may be emotional to leave. On the other hand, if enganged in deviant behaviour it may be dangerous to leave and so a strategy musst be developed.
32 of 33


  • Form of research in which all the variables are closely controlled, so that the effect of changing one or more variables can be understood.
  • Less common in sociology as it is almost impossible to isolate a social event from the real world around it. This means that researchers cannot control all the variables which is the essence of an experiment.
  • Experimenter effect = awareness of being in an experiment affects the normal behaviour of the participants.
  • Two types - laboratory and field.
  • See handout for details on each type.
33 of 33


No comments have yet been made

Similar Sociology resources:

See all Sociology resources »See all Sociological research methods resources »