Research Methods in Sociology

1
Research Methods in Sociology

• Doing and Evaluating Research

2
Steps in the Research Process
3
Formulating the problem

• Find a problem
• Pick a general topic to study
• Review the literature
• Summarize previous research and identify
  important concepts
• Concepts are mental abstractions of the social
  objects to be studied
• Specify a research question
• What is the relationship between the concepts?
• Develop a hypothesis - an expectation or
  prediction derived from theory
• Concepts are stated as variables
• Independent (IV) - creates change or effect
• Dependent (DV) - change or effect
• Control - constant

4
Measurement

• Translate abstract concepts into concrete
  variables
• variable - characteristic or trait that can be
  measured
• Specify the variables in the hypothesis
• Operationally define the variables
• Specify how concepts will be turned into
  variables to be empirically measured
• validity - Are you measuring what you think
  youre measuring?
• reliability - Would you get the same results if
  you repeated the measurement?

5
Population Sample

• Determine the population and sample
• Population - the entire group of people who are
  the focus of the research and to whom the
  findings will be generalized
• Sample - A subset of the population that is being
  studied research participants
• Sampling - selecting units from a population in
  order to make generalization back to the larger
  population
• Random Sample - Based on the laws of probability
  and used to reduce bias, subjects are selected so
  that every member has an equal chance of being
  selected

6
Research Design

• How will you collect data?
• Experiment
• Secondary research
• Survey research
• Field research

7
Research Design

• Experimental Research
• A systematic attempt to test a causal hypothesis
  by manipulating the independent variable
• Use of Experimental and Control Groups
• The experimental group is exposed to the
  independent variable (control group may receive a
  placebo)
• Random assignment
• To insure that the experimental and control group
  are equivalent
• Pre-test and Post-test measurements
• Pre-tests are used as a baseline for comparison
  while Post-tests are used to determine whether
  significant change has occurred
• Comparison
• Are there any differences in scores for the
  dependent variable across the two groups?

8
Research Design

• Secondary Research
• Data and information compiled for other purposes
  are accessed and reanalyzed
• Survey Research
• Uses questionnaires and interviews for data
  collection
• Provides data for quantitative analysis
• Field Research
• Research conducted in the natural setting of a
  social phenomenas occurrence
• Typically descriptive work (More theory-building
  than testing)
• Provides data for qualitative analysis

9
Data Collection Questionnaires (survey)

• Closed-ended
• Respondent is provided with a pre-existing set of
  responses
• Response choices must be mutually exclusive and
  exhaustive
• Data is easy to summarize

• Open-ended
• Respondent is asked to provide his or her own
  answers
• Provides detailed data (validity)
• Time consuming to analyze
• Decreased reliability

10
Example of Survey Data Use

• Do income and race relate to who finds life
  exciting?
• What is your hypothesis?
• Independent variable?
• Dependent variable?

11
Data CollectionInterviews (survey)

• Personal Interviews
• Pros
• Better quality results
• Response rates nearly 95
• Expensive and time-consuming
• Cons
• Expensive to conduct
• Potential of interviewer bias
• Possibility that the interviewee may withhold
  information

• Telephone Interviews
• Pros
• Swift data collection
• Broad geographical coverage
• Cheaper to conduct
• Cons
• Interviewee impatience
• Cannot use visual aids or pick up on visual cues
• Unable to control factors that may distract the
  interviewee

12
Data CollectionParticipant Observation (field)

• Participant-observer role
• Researcher observes phenomenon in its setting,
  but does not directly participate
• Informs that research is being conducted
• Less deceptive and fewer ethical dilemmas
• Possibility that researcher may influence
  behavior of those he or she studies

• Complete participant role
• Researcher engages directly in the phenomenon
  they are studying
• Does not inform that research is being done
• Can be emotionally and fiscally costly
• Many ethical dilemmas

13
Data CollectionSecondary data sources/archives

• Very popular form of data collection
• Relatively inexpensive
• Broad coverage
• Commonly used secondary data sources
• The United States Census (www.census.gov)
• The General Social Survey (www.icpsr.umich.edu808
  0/GSS/homepage.htm)
• The Uniform Crime Report (www.fbi.gov/ucr/ucr.htm)
  

14
Data Analysis and Interpretation

• Choose how to analyze data
• Quantitative analysis
• Numerical data
• Qualitative analysis
• Non-numerical data
• Summarize and interpret
• Decide to retain or reject hypothesis

15
Commonly Used Statistics for Analysis and
Interpretation
16
Measures of Central Tendency

• The mean is commonly called the average
• It can be greatly influenced by extreme scores
  also called outliers
• The median is the middle of a distribution
• The median is less sensitive to extreme scores
  than the mean and this makes it a better measure
  of central tendency.
• The mode is the most frequently occurring score
  in a distribution
• The mode is subject to sample fluctuations and is
  therefore not recommended for use as the single
  measure of central tendency

17
Measures of Variation

• Range
• Simply the difference between the largest and
  smallest values in scores for the study
  population
• Tells little about the variation in scores
  because it only consists of two values
• Standard Deviation
• More useful measure of variation since it
  measures the average distance of each score from
  the mean
• Tells how tightly the scores on a variable are
  clustered around the mean
• One standard deviation away from the mean in
  either direction accounts for 68 percent of the
  scores
• Two standard deviations away from the mean
  account for 95 percent of the scores
• Three standard deviations account for 99 percent
  of the scores

18
Measure of Association

• Pearsons Correlation Coefficient (r)
• Used to determine both the strength and the
  direction of a linear relationship
• It takes on values from -1 to 1
• The further from 0, the stronger the relationship
  between two variables
• Whether the score is positive or negative
  indicates the direction of the relationship
  between two variables
• Negative relationships mean the variables move in
  opposite directions while positive relationships
  mean the variables move together

19
Always remember.

• Correlation DOES NOT prove causation!
• Ice cream sales increase as electricity usage
  increasesdoes this mean that eating more ice
  cream causes people to use more electricity or
  vice versa?
• An important goal for sociologists is to
  determine causality, but it is a difficult one to
  realize.

20
Determining Cause and Effect

• Time Order
• The cause must precede the effect in time.
• Correlation
• A statistically significant association must
  exist.
• Elimination of Confounds
• The association must not be due to a third
  variable.
• If a third variable is found to be the source of
  change, the initial relationship is determined to
  be spurious.
• Theory
• Logical link explaining the relationship

21
Evaluation of Results

• Consider the importance of research for theory
  and any applied possibilities
• Does your research support an existing theory?
• Does it suggest the need for reformulating a
  theory?
• What further research should be done?
• What policy implications does your research have?
  
• Prepare a formal report
• To share with the social science community and
  general public

22
ASA Code of Ethics

• Disclose research findings in full and include
  all possible interpretations of the data.
• Safeguard the participants right to privacy and
  dignity while protecting them from harm.
• Protect confidential information provided by
  participants.
• Acknowledge research collaboration and disclose
  all financial support.