Sociological Research Methods

1
Sociological Research Methods
2
Research Methods

• Methodology
• A system of rules and procedures that guide
  scientific investigation research guidelines.
• Characteristics of scientific methodology
• 1. concerned with objectivity or exclusion of
  bias
• 2. strict rules for admissible evidence
• 3. careful records
• 4. replication of findings

3
Scientific Methodology

• Science
• Logical systematic accumulation of knowledge
  based on direct, systematic observation.
• Empirical evidence evidence we can verify with
  our senses and experiences.
• Scientific assumptions
• 1. there is order in the universe
• 2. correlation patterns of relationships
• 3. causation cause and effect relationships
• 4. prediction

4
Scientific Methodology

• Variable
• Any characteristic that can change or have
  different levels, rates, or statuses.
• Typical sociological variables age, sex, race,
  education, social class, etc.
• To operationalize a variable specifying exactly
  what will be measured to indicate the variable.
• Two types of influences among variables
• 1. independent variable the one that causes an
  effect.
• 2. dependent variable the one that is affected
  by another variable.

5
Scientific Methodology

• Causation
• When a change in the status of the independent
  variable (variable A) causes a corresponding
  change in the dependent variable (variable B),
  then there is causation, or a causal relationship
  between the two variables. (A?B Changes in A
  cause changes in B).
• Generalizations
• Recurrent relationships among particular
  variables allow us to make generalizations, or
  probability statements about the likelihood of
  future occurrence.

6
Scientific Methodology

• Correlation
• A regular relationship between 2 or more
  variables the beginning point in establishing
  causation, or cause-effect relationships.
• High versus Low correlation.
• High correlation is when great changes to the
  independent variable are associated with
  similarly great changes to the dependent
  variable. Low correlation is when great changes
  to the independent variable are associated with
  only minor changes in status of the dependent
  variable.
• Positive versus Negative correlation.
• When an increase in the independent variable is
  associated with an increase in the dependent
  variable, this is a positive correlation. When an
  increase in the independent variable is
  associated with a decrease in the dependent
  variable, then this is a negative correlation.

7
Scientific Methodology

• Correlations may be causal or spurious.
• Causal correlation
• When the changes in the dependent variable are
  due to the changes in the independent variable.
• Spurious correlation
• When the changes in the dependent variable are
  not due to changes in the independent variable.
  Both are correlated they seem interrelated
  but it is due to some other outside variable.
• Example Going to a hospital is correlated with
  dying, but it is a spurious correlation because
  it is a disease or injury - not the hospital -
  that causes dying.

8
Scientific Methodology

• Controls
• Ways of excluding the influence of outside
  variables so they do not influence the
  relationship of interest.
• We want to look exclusively at variables A and B
  and how they relate to each other, so we need to
  control for the influence of other variables (C,
  D, E, etc)
• We use controls to determine whether a
  correlation is causal or spurious.

9
Scientific Methodology

• Example does smoking cause lung cancer?
• 1. Is smoking correlated with lung cancer rates?
• 2. If so, is this relationship causal or
  spurious?
• To determine, we must use controls to rule out
  the influence of other variables, like workplace
  pollution, exposure to solvents, gasoline and
  other possible cancer causers. If we control for
  those other variables and examine exclusively the
  relationship between smoking and cancer, we find
  a causal relationship. Indeed, it is a relatively
  high, positive correlation. Therefore, there is a
  high probability that smoking does cause lung
  cancer.

10
Difficulties in Sociological Research

• 1. The mere act of investigating social behavior
  may alter the very behavior being studied.
• The Hawthorne Effect when people know they are
  being studied, they alter their behavior.
• 2. People have emotions and moods - even
  sociologists and this affects performance.
• 3. The origins of behavior are usually complex.
• 4. Ethical considerations limit the types of
  research methods we use.
• 5. Social patterns change constantly.

11
Methodology split to address these difficulties
and theoretical differences

• Scientific Sociology adopts a quantitative
  approach, with emphasis on scientific methods,
  exclusion of bias, and statistical measurement
  techniques. This is highly precise, but dry.
• Interpretive Sociology adopts a qualitative
  approach, with emphasis on verstehen (empathy, or
  understanding the subjective reality of the
  research subjects). This is prone to bias, but
  yields rich insights.
• Critical Sociology, related to Karl Marx, is a
  qualitative method that focuses on ethical issues
  involving the need for social change where the
  researcher is also an activist. It is prone to
  bias, but yields rich insights.

12
Four Basic Research Methods

• 1. Experiment
• 2. Survey research
• 3. Observational study
• 4. Existing sources or secondary data analysis

13
1. Experiment

• A method of investigating cause-effect
  relationships under controlled conditions. The
  idea is to study the relationship between 2
  variables while controlling for other factors.
• Typical technique
• 1. divide people into 2 identical groups, similar
  in all respects. One group is the experimental
  group, the other is the control group.
• 2. Introduce the independent variable to the
  experimental group, but not the control group.
• 3. Then, afterwards, compare these 2 groups to
  see if they are still identical. If not, it is
  probably due to the independent variable.
• Advantages ability to use controls ability to
  perform statistical analysis.
• Disadvantages Hawthorne effect limited to
  narrowly defined issues and smaller scale
  research.

14
2. Survey

• Scientific survey a method of systematically
  obtaining accurate, standardized information
  about a population.
• Survey population the total category of people
  we are interested in. This population must be
  carefully defined.
• Census ask every member of the population very
  expensive and time consuming for large
  populations but very accurate.
• Sample ask a small but representative group
  pulled from the survey population. This is more
  efficient, but less accurate.
• Sample size
• 100-3000 is the typical size for a simple survey.
  Above 3000 is overkill, and below 100 is too
  small to make generalizations.

15
2. Survey, cont

• Representative sample
• Individuals are selected in such a way as to
  accurately represent the population.
• Allows generalization to the entire survey
  population.
• Individuals must be randomly selected, so that
  every single member of the population had an
  equal chance to be selected to participate.
• Survey response rate
• The percentage of the sample/census that agrees
  to participate. The higher the response rate the
  better.

16
2. Survey, cont

• Survey questions
• Must be worded to avoid bias
• Straightforward
• Unemotional
• Clear or unambiguous
• Neutrally worded
• Balanced, not loaded questions.
• Survey administration mail, phone, face to face

17
2. Survey, cont

• Evaluating a survey
• 1. Was the population clearly defined?
• 2. If there were generalizations to the
  population, was the sample size 100 or more?
• 3. If there were generalizations to the
  population, was it a representative sample?
• 4. Was the response rate sufficiently high?
• 5. Were the questions clear, fair and balanced?

18
2. Survey, cont

• Advantages of survey method
• Good for examining macro social forces.
• Can do time studies to see how something is
  changing.
• Most scientific surveys permit statistical
  analysis where we can generalize to the larger
  population.
• Disadvantages
• People give false information.
• Subjects give opinions on topics they know
  nothing about.
• It can be difficult to assure representative
  sample.

19
3. Observational Studies

• A common research method in Interpretive and in
  Critical Sociology.
• Takes an in-depth look at a particular group,
  event, or social process.
• Emphasis on verstehen.
• Lab versus Field observation
• Lab researcher invites people to their
  laboratory.
• Example the focus group
• Field researcher goes to where the people are
  their natural habitat.
• Example the case study

20
3. Observational Studies

• Two types of observation
• 1. detached observation researcher remains
  aloof, distant, anonymous and observes from a
  distance.
• 2. participant observation researcher is
  directly involved in the group process.

21
3. Observational Studies

• Advantage of observational studies
• Yields deep insights
• Rooted in real life situations
• Disadvantage
• Less precise
• Prone to bias
• Cannot generalize to wider populations

22
4. Existing Sources

• Secondary data analysis involves re-examining
  data/information collected by others.
• Census data
• Body of sociological research and literature
• Courthouse and other official records
• Advantages
• Inexpensive
• Accessible anyone can do it.
• Disadvantages
• Errors in the primary data are passed on to the
  secondary data analysis

23
Ethics in Research

• Five basic research dilemmas
• 1. harm to participants
• 2. invasion of privacy
• 3. informed consent
• 4. applications of research
• 5. deception

24
End of Chapter 2