Theory/methods/debates

• Social facts


  Values/norms/structures that exist independently of the individual & can exercise social control

  • E.g. the law
• Positivism

  • Society can be studied using scientific techniques
  • Sociology should focus on social facts & observe/measure them using quantitative data
• Interpretivism

  • People's behaviour is influenced by the meanings they give to social situations
  • Sociology should interpret these meanings & provide an understanding
  • Research should focus on gaining verstehen
• Verstehen


  The ability to understand how people see the world
• Durkheim: study of suicide

  • Investigating the social causes of suicide using coroners' reports
  • People are more likely to commit suicide in times of anomie
  • Classified 4 types of suicide e.g. egoistic (social isolation)
  • Argued the study proved sociology could be scientific
• Practical research considerations:

  • Access
  • Time
  • Availability of existing data
  • Researchers' values/beliefs
  • Cost
• Ethical research considerations:

  • Personal safety
  • Witnessing illegal activity
  • Confidentiality/anonymity
  • Informed consent
  • Impact on participants
  • Impact on the researcher
• Theoretical research considerations:

  • Method
  • Research subject
  • Reliability
  • Repeatability
  • Generalisability
  • Representativeness
• Informed consent


  Participants know they are part of a study & are willing to participate
• Validity


  How accurate results are

  • E.g. whether people responded to a questionnaire honestly
• Reliability


  Whether the research results can be replicated using the same method & conditions
• Representativeness


  Whether a sample accurately represents a social group/society as a whole
• P ractical

  E thical

  R eliable

  V alidity

  E vidence of studies

  R epresentative

  T heoretical
• Townsend: 'Poverty in the United Kingdom'

  • Critical of govt estimates/measures of poverty as inaccurate: argued they were altered to fit the govt's political agenda
  • Sought to create his own measure of poverty (poverty index)
  • Questionnaires
• Townsend: 'Poverty in the United Kingdom'


  P: expensive, large sample, low response rate

  E: anonymous, informed consent, no risk of harm

  R: easily replicated but results differed hugely to official statistics

  V: some measures included people not in poverty e.g. regularly eating fresh meat

  E:

  R: used his results to generalise about national poverty levels but official statistics use a much larger sample

  T: positivist method
• Humphreys: 'Tearoom Trade'

  • Investigating men who participated in impersonal sex acts in public bathrooms
  • Inspired by overwhelmingly negative perception of these men: sought to provide a more scientific understanding
  • Covert participant observation & structured interviews later
• Humphreys: 'Tearoom Trade'

  P: very time-consuming (>1yr), inexpensive, skills required for covert observation

  E: no informed consent (interviews took place under false pretences), risk of reputational harm for the men

  R: difficult to replicate the sample selection, informal observation

  V: no Hawthorne Effect due to the nature of the observation (some didn't know Humphreys was a researcher)

  E:

  R: not necessarily a representative sample (well-educated men were far more willing to be interviewed)

  T: interpretivist/qualitative methods & positivist (produced statistics)
• Venkatesh: 'Gang Leader for a Day'

  • Studied a working class, predominantly African-American community
  • Overt participant observation
• Venkatesh: 'Gang Leader for a Day'

  P: time-consuming, inexpensive

  E: informed consent (overt), participants can't be identified

  R: informal method of observation difficult to replicate

  V: possibility of Hawthorne Effect (but participants appeared to behave as normal)

  E:

  R: sample probably not large enough to be generalisable

  E: interpretivist/qualitative method
• Barker: 'Making of a Moonie'

  • Studied members of the Unification Church (AKA 'Moonies')
  • Inspired by accusations that the Church was corrupt & brainwashed its members
  • Aimed to find out why people joined the religion & whether these accusations were correct
  • Triangulation (multiple methods): overt participant observation, in-depth interviews (2+ hours), questionnaires
• Barker: 'Making of a Moonie'

  P: very time-consuming (6yrs)

  E: informed consent, no risk of harm

  R: not easily replicated as the research was very extensive but similar results possible if the method was repeated

  V: triangulation (verstehen), Hawthorne Effect possible but she lived with them so personal relationships developed

  E:

  R: detailed findings, large sample (claimed to cover every UK Moonie)

  T: interpretivist/qualitative methods
• Content analysis


  Analysing qualitative secondary sources & quantifying it

  • E.g. sorting content of a document into categories
• Content analysis

  P: relatively cheap

  E: some documents raise ethical issues, quantifying these may be insensitive

  R: easily replicable & should produce the same results

  V: different researchers may interpret the content differently

  E: e.g. studying the use of certain words/phrases in news reports

  R: quantifying the data makes it more generalisable, can't improve a small/unrepresentative sample

  T: produces quantitative data but uses qualitative (so positivist & interpretivist)
• Official statistics

  P: easily to access, inexpensive

  E: anonymised & already publicly available

  R: standardised, method used to collate the statistics not always available

  V: produced by the state so socially constructed & potentially biased

  E: e.g. census data, crime statistics

  R: large sample

  T: quantitative data (positivist)
• Lab experiments


  Take place in a controlled environment so variables can be isolated
• Field experiments


  Carried out in a natural setting & participants are generally unaware they're involved; researcher manipulates 1 or more variables to see its effect on the participants
• Lab experiments

  P: expensive, time-consuming, hard to identify & control all variables

  E: potential harm to participants, informed consent

  R: standardised method

  V: presence of the researcher could create Hawthorne Effect

  E: Stanford Prison Experiment (Zimbardo), Bobo Doll Experiment (Bandura)

  R: samples are generally small, not necessarily representative

  T: positivist (uses scientific method)
• Field experiments:

  P: expensive, time-consuming

  E: lack informed consent, potential harm to participants

  R: uncontrolled environment so method might not be replicable

  V: less artificial setting, no Hawthorne Effect

  E: Pygmalion Effect (Rosenthal & Jacobson)

  R: samples are generally small, not necessarily representative

  T: interpretivist (not a controlled environment)
• Surveys

  P: large sample, inexpensive, low response rate

  E: informed consent is implicit

  R: questions are identical so easily replicated

  V: researcher isn't present so can't affect responses, participants can give false/vague responses (social desirability), participants may not understand the question, researcher interprets the responses

  E: Making of a Moonie (Barker), Poverty in the United Kingdom (Townsend)

  R: generally large sample, low response rate

  T: produce quantitative data
• Social desirability effect


  Participants try to present themselves in a positive light rather than producing genuine responses/behaviours
• Questionnaires

  P: inexpensive, quick, accessible to participants

  E: implicit informed consent, can be anonymised, sensitive answers cannot then be reported

  R: questions are identical so replicable

  V: participants may not feel the possible responses to a closed question are accurate, participants may lie/misunderstand

  E: census

  R: easy to get a representative sample e.g. through the electoral register, low response rate may change the demographics

  T: positivist (closed questions), interpretivist (open questions)
• Structured interviews


  Asking the same predetermined questions to all participants in the same order & rating the responses with a standardised scoring system
• Semi-structured interviews


  Combining a predetermined set of open questions with the opportunity for the researcher to explore certain themes/responses further
• Unstructured interviews


  Flexible & informal interviews (like a conversation)
• Group interviews


  Researcher speaks to a group of participants rather than 1 person at a time
• Interviews:

  P: time-consuming, interview training required for best results

  E: informed consent, unstructured allows rapport to be built, researcher with poor skills can cause distress, sensitive topics may be distressing

  R: structured easily replicated, semi-structured/unstructured hard to replicated

  V: verstehen, participants may change their answers to achieve social desirability

  E: Making of a Moonie (Barker), 7 Up (semi-structured)

  R: generally smaller samples, semi-structured/unstructured are quite personalised

  T: positivist (structured), interpretivist (semi-structured/unstructured)
• Observations

  P: time-consuming, access may be difficult e.g. Gang Leader for a Day

  E: participant observation may be covert so no informed consent, researcher may witness illegal activity/be at risk of harm

  R: hard to replicate because it is in an uncontrolled environment

  V: verstehen, Hawthorne Effect negatively affects non-participant

  E: Gang Leader for a Day (Venkatesh), Learning to Labour (Willis)

  R: generally small sample, results are often specific to the group & not generalisable

  T: interpretivist/qualitative
• Case studies


  Investigating an individual/group in great detail rather than trying to obtain a representative sample
• Case studies

  P: time-consuming, difficult to find willing participants

  E: informed consent, sensitive topics may be raised

  R: specific to the group so impossible to replicate the results

  V: verstehen, Hawthorne Effect

  E: Learning to Labour (Willis)

  R: specific to the group, not meant to be representative

  T: interpretivist (qualitative data)
• Personal documents


  Created with the intention of keeping to oneself & generally express personal thoughts/feelings not intended for public view