Sociology: Theory and methods

• Interpretivists prefer qualitative methods, methods that produce non-numerical data, as they give us a 'feel' for what something is like.
• Qualitative methods

  • Unstructured interviews
  • Participant observation
  • Personal documents
• Positivists prefer to use quantitative methods, data in numerical or statistical form, that they believe will produce objective true, scientific knowledge of society
• Positivists see society as an objective reality made up of social facts. Social reality follows patterns that exist because society influences its members, systemically shapes their behaviour. These patterns can be observed and measured.
• Positivists believe they can establish cause and effect through observation and measurement
• Quantitative methods

  • Questionnaires
  • Experiments
  • Structured interviews
  • Official statistics
• What are the Practical Issues of lab experiments

  • They are closed systems whereas society is an open system
  • Individuals are complex so hard to 'match' members of control and experimental groups exactly
  • Cannot be used to study the past
  • Can only use small samples
  • Hawthorne effect
  • Expectancy effect - researchers expected outcomes my influence actual outcome
• Ethical issues of Lab experiments

  • Informed consent - though some experiments may benefit from some subjects being deceived
  • Harm to subjects- experimental group gain from experiment but control group don't - therefore there needs to be a moment where the same treatment is offered to the control group
• Theoretical Issues of Lab experiments

  • Reliability - easily replicable
  • Less likely to be representative- small sample and lack external validity due to high control and unnatural setting
  • May lack internal validity due to artificiality and Hawthorne effect
• Interpretivists argue against lab experiments as humans are different from rocks and plants (and other things natural scientists study). We have free will and choice and our behaviour is not caused by external forces. Our actions can only be understood by the choices we freely make.
• Issues with Field experiments

  • unethical: carried out without subject's knowledge or consent (can be argued no harm was done in Wood et al's experiment no harm was done)
  • Cannot be certain if true cause was identified as there are there many other variables that cannot be controlled
• Issue of comparative method

  • avoids artificiality
  • can be used to study the past
  • avoids ethical problems
  • even less certainty due to even less control
• What is the comparative method

  • identifies two groups that are alike in all major aspects except for the one variable of interest
  • compares the two groups using secondary data to see if the one difference between them has any effect
  • E.g. Durkheim's study of suicide (1897)- used official statistics to compare causes of suicide
• Practical Issues of Questionnaires

  • quick and cheap way to gather large amounts of data
  • no need to recruit and train interviewers
  • data is easily quantifiable
  • may need to offer incentives
  • Low response rate
  • inflexible
  • data is often limited and superficial
  • only provides a snapshot
• Theoretical Issues

  • Reliability- easy to replicate
  • Representative- large scale and sampling frames (can be undermined low response rate)
  • Detachment and objectivity- unbiased
  • Lying, forgetting, misunderstanding and trying to impress
  • Imposing researchers meanings
• Interpretivists see questionnaires as they see them as lacking validity. They argue to get valid data, you must be closely involved with the people you're researching so you can gain a subjective understanding of their means.
• What is imposing the researcher's meaning's

  • questionnaires are more likely to impose the researcher's framework of ideas on the respondent than to reveal the respondent's meanings
  • this is done by choosing the questions in advance
  • close ended questions force respondents to fit their views to the answers on offer
  • open ended questions causes non-identical answers to sometimes get lumped up together when quantifying
  • Shipman (1997) when researcher's categories are not the same as respondents' 'pruning and bending of data is inevitable
• Practical Issues of Structured Interviews

  • Can cover quite large numbers of people e.g. Young and Willmott (1963) interviewed 933 people in this way on their study of east London families
  • Suitable for gathering straightforward factual information e.g. person's age, job, daily routine etc.
  • Results are easily quantifiable
  • Inexpensive and easy to train interviewers
  • High response rates e.g. Young and willmott only had 54 refusals out of 987 people they approached
  • inflexible as they are standardised
  • only snapshots
• Theoretical Issues of structured interviews

  • Reliable - easy to replicate
  • Representative- quick and cheap so large numbers can be surveyed (however people may lack time and willingness to be interviewed)
• Interpretivists believe structured interviews produce a false picture as they:

  • Usually use closed-ended questions
  • Give little freedom to explain questions or clarify misunderstandings
  • People may lie or exaggerate, producing invalid data
  • May impose researcher's framework of ideas on interviewee
• Feminists also reject structured interviews and questionnaires as the relationship between researcher and responded reflects the exploitive nature of gender relationships in patriarchal society.

  • Reinharz (1983) "researchers take hit, and run. they intrude into their subjects' privacy"
  • Anne Oakley (1981) there is a strict division of labour in structured interviews. Researcher takes active role and interviewees are passive objects in the study.
• Practical Issues of official statistics

  • free source of huge amount of data
  • the state has power to compel certain individuals to supply certain data
  • allow for comparisons between groups
  • collected at regular intervals so can show trends and patterns over time
  • statistics are collected for state purposes so there may be none available for topic of interest
  • definitions state uses when collecting data may be different to sociologist's
• Theoretical issues of official statistics

  • Representative as they are very large-scale and often cover the entire population.
  • Reliable as they are compiled by trained staff who use standardise categories and collection techniques and follow set procedures to ensure they can easily be replicated.
• Interpretivists view statistics are social constructs that represent the labels officials attach to people. They argue that officials pigeonhole individuals into becoming mere statistics. For interpretivists official statistics are often considered invalid.
• Soft statistics tend to give a much less valid picture of reality. They are often compiled from administrative records from state agencies. They represent only what these agencies have decided to record. They often neglect a 'dark figure" of unreported cases.
• Hard statistics provide a more valid picture as there is little dispute on how to define the categories used to collect the data and they are often created from registration data.