Sociology: Research Methods

• Quantitative data


  A research method that produces data of a numerical kind.
• Qualitative data


  A research method that produces data of a written kind.
• Representativeness


  How well the sample used mirrors the group the researcher wants to study.
• Reliability


  The extent to which results stand up to re-testing.
• Validity


  The extent to which the results paint a true picture of real life behaviour.
• Verstehen


  A German word meaning 'empathy'; the extent to which the researcher can empathise with a person's behaviour by putting themselves in their place.
• Positivism


  The view that sociology can be studied as a science, uses quantitative research methods.
• Interpretivism


  The view that sociology cannot be studied as a science, uses qualitative research methods.
• Operationalisation


  The process by which a sociologist defines or measures key variables precisely e.g. social class might be measured in terms of occupation.
• The Hawthorne Effect


  When participants know they are being watched and change their behaviour because of this.
• Triangulation


  This is where sociologists use a combination of quantitative and qualitative research methods.
• Ethical issues


  What is morally right or wrong to do in relation to studying people?
• Hypothesis


  A prediction to be tested i.e. a cause and effect statement, favoured by Positivists.
• Aim


  A broad idea of what they want to find out of the investigation, favoured by Interpretivists.
• Random sampling


  When every person in a target population has an equal chance of being selected in the sample.
• Official statistics


  Quantitative data gathered by the government or other official bodies. e.g. births, deaths, educations, crime and health statistics.
• Non-official statistics


  Published by non-government organisations e.g. trade unions and churches.
• What are two ways to collect official statistics?

  Registration (e.g. parents having to register births by law). Official surveys (e.g. the Census and the General Household Survey).
• Emile Durkheim


  Studied suicide statistics to try and compare the suicide rates between Protestants and Roman Catholics, only discovered the more a person is socially controlled they are more likely to commit suicide.
• How do Interpretivists see official statistics?

  As social constructions e.g. police statistics do not record all crime.
• Example of triangulation

  Barker studied the Moonies and used both overt PO and questionnaires
• Observation - Unstructured Participant

  Example: Paul Willis (12 working class males)Strengths: -Valid/gain insight (T)-Preserves Naturalness (P)-Avoids subjects change of behaviour (P)-High 'rapport' (T)Limitations:-Lacks Reliability and Representativeness (T)-Loss of objectivity (T)-Very time consuming (P)-Costly (P)
• Observation - Structured Participant


  Example: Flanders - did research into teacher and pupil interactions.Strengths- Reliable (T)- Representative (T)- Quick (P)- Cheap (P)- Easy to analyse (P)Limitations- Low validity (T)- No reasons/ insight (T)- Inflexible (P)
• Observation - Covert Participant

  Example: Laud Humphreys (tearoom trade), James Patrick (gangs), Griffin (racism). Strengths:- Avoids people changing their behaviour (P)- Preserves naturalness (P)- High in validity (T)- Gain insight (T)- Sometimes the only method able to use (P)Limitations:- Deceitful (E)- Researcher has to keep up an act (P)- Hawthorne effect may still happen (P)- Guilty Knowledge (E)- May have to participate in illegal activities (E)- Hard to access certain groups (P)- Time consuming (P)
• Observation - Overt Participant


  Example: Venkatesh (gang leader), Barker (moonies)Strengths:- Informed consent (E)- Reliable (T)- No act needs to be kept up (P)- The researcher can ask questions (P)- Interpretivists would like it (T)- Gains depth and insight (T)Limitations;- Not representative (T)- Hawthorne Effect (P)- Researcher can become overly subjective (T)- Costly (P)- Time-consuming (P)- Researcher may need training (P)- Can be hard to gain access (P)
• Interviews - Structured

  Example: Laud HumphreysStrengths: - Training is straightforward and cheap (P)- Can cover a large amount of people (T)- Representative (T)- Easy to analyse (P)- Quick to conduct (P)- Standardised procedure (T)- Suitable for hypothesis testing (P)Limitations:- Do not gain depth or insight (T)- Closed questions = restrictive (T/P)- No explanations (T/P)- Social desirability effects (P)- Interviewer led (P)- Lack validity (T)- Inflexible (T/P)- Not good for sensitive issues (E)- Interviewer effects (P)
• Interviews - Unstructured

  Example: Dobash and Dobash, BarkerStrengths:- High in validity (T)- Supported by Interpretivists (T)- Suitable for sensitive issues (E)- No deception (E)- Gained informed consent (E)- Have a right to withdraw (E)- Gain reasons (T)- Confidentiality (E)Limitations:- Interviewer effects (P)- Needs training (P)- Researcher characteristics (P)- Time-consuming (P)- Lacks reliability (T)- Lacks representativeness (T)- Costly (P)
• Interviews - Group

  Example: Paul Willis - Carried out group interviews with 12 working class males, he found out that they created an 'anti-school subculture' Strengths:- Participants may feel more comfortable (P/E)- Participants often throw ideas around the group which could stimulate more ideas and topics to discuss (P)- Useful way of generating initial ideas to be followed up (P)- Can observe group dynamic (P)- Higher in validity (T)- Higher in representativeness (T)- Ethical (E)- Interpretivists would support (T)Limitations:- May be peer pressure (E/P)- Data generated from group interviews are more complex and difficult to analyse (P)- Hard to maintain focus (P)- One or two individuals may dominate the discussion (P)- 'Group think' may occur (P)- Lacks reliability and representativeness (T)
• Documents - Personal


  Example: Hey- Made use of notes that girls passed to each other in class, and then threw in the bind, to understand friendship patterns.Strengths:- Rich qualitative data (T)- Interpretivists like this (T)- High in validity (T)- Cheap (P)- Saves time (P)- High in verstehen (T)- May be the only source available (P)Limitations:- Difficult to access (P)- May require special skills and knowledge (P)- Analysis t​akes a long time​ (P)- There may be l​ack of informed consent (E)- Invasion of privacy (E)- May be highly sensitive (E)- Low reliability (T)- Not representative so results cannot be generalised (T)- Small sample size (P)- Positivists dislike (T)
• Documents - Public


  Example: McRobbie analysed magazines in the 1970's and 1990's and found huge differences in the way females were being represented.Strengths:- Easy to access (P)- Cheap (P)- Quick (P)- Ethical (E)- Reliable (T)- Representative (T)- Positivists are more likely to use this method (T)Limitations:- Lacks validity (T)- Lacks meaning (T)- Not supported by interpretivists (T)
• Experiments - Lab

  Examples: Milgram, BanduraStrengths:- Reliable (T)- Highly controlled environment (P/T)- Positivists like (T)- Representative (T)- Quick (P)Limitations:- May cause harm (E)- Unrepresentative (T)- Lacks validity (T)- Hawthorne effect (P)- You can't study the past (P)- Time-consuming (P)- Relatively expensive (P)- Difficult to gain a sample (P)- Impossible to identify and control all variables (P)- Interpretivists dislike (T)
• Experiments - Field

  Example: Rosenthal and Jacobson, studied the effect of teacher expectation on student performance and progressStrengths:- Inexpensive (P)- High in validity (T)- Interpretivists like (T)- Natural setting = natural behaviour (P)Limitations:- Time Consuming (P)- Unrepresentative (T)- May psychologically harm participants (E) - Lacks Reliability (T)- Less control over variables (T)- Access can be problematic (P)- Positivists will reject this method as it is unreliable and unrepresentative (T)
• Experiments - Comparative

  Example: Durkheim's study of suicide (1897) Strengths: - It can be used to study past events (P)- No ethical problems (E)- It avoids artificiality (T)- Saves money (P)Limitations:- Researcher has less control (T)- Low in validity (T)- Time-consuming (P)- Interpretivists dislike (T)
• Official Statistics

  Example: Durkheim, comparison on suicide between Catholics and Protestants. Strengths:- Quick (P)- Cheap (P)- Useful to make comparisons (P)- They can identify trends and patterns (P) - It collects large-scale data (T)- Reliable (T)- Representative (T)Limitations:- Lacks validity (T)- Socially constructed (P)- No explanations (P/T)- Definitions can be changed (P)
• Questionnaires


  Example: the censusStrengths:- Quick (P)- Easy access to a large sample (P)- Highly ethical as participants aren't pressured to complete and their info can be kept confidential (E)- Highly reliable T)- Representative (T)- Positivists like (T)Limitations: - May need to offer incentives (P)- Low in validity (T)- You don't know who completes the questionnaire making it less representative (T)- Low response rate (P)- Interpretivists don't like (T)