How Sociologists Do Research

1
Students Please Read If your textbook doesnt
already come with the helpful study aids weve
discussed, go to your local college bookstore or
go on-line to the Sociology home page
http//www.thomsonedu.com/sociology/

• How Sociologists Do Research

Professor Course/Section
2
Chapter Outline

• Science and Experience
• Conducting Research
• The Main Methods of Sociological Research
• The Importance of Being Subjective
• Appendix Four Statistics You Should Know

3
Research and Experience

• Experience filters perceptions of reality.
• In sociological research this occurs in four
  stages
• The experiences and passions of sociologists
  motivate research.
• Our values leads us to theories.
• Sociologists interpretations are influenced by
  previous research.
• Methods used to gather data mold our perceptions.

4
How Research FiltersPerception
5
Question

• If you possessed the money, skill, and other
  necessary resources, in which one area would you
  like to conduct research?
• Racial profiling
• The cause(s) of sexual orientation
• Sexual assault and abuse
• The effects of divorce on children

6
Types of NonscientificThinking

• Knowledge based on tradition.
• Knowledge based on authority.
• Knowledge based on casual observation.
• Knowledge based on overgeneralization.
• Knowledge based on selective observation.

7
Types of NonscientificThinking

• Knowledge based on qualification.
• Knowledge based on illogical reasoning.
• Knowledge based on ego defense.
• Knowledge based on the premature closure of
  inquiry.
• Knowledge based on mystification.

8
The Research Cycle

• Formulate a research question.
• Review the existing research literature.
• Select a research method.
• Collect data.
• Analyze the data.
• Publicize the results.

9
The Research Cycle
10
Ethical Considerations

• Researchers must respect their subjects rights
  to
• Safety
• Privacy
• Confidentiality
• Informed consent

11
Field Research

• Based on the observation of people in their
  natural settings.
• Field researchers go where people meet
• Italian-American slums
• Intensive care unit of a major hospital
• Audience of a daytime TV talk show

12
Participant Observation

• Sociologists take part in the activities of the
  people they are studying.
• Allows researchers understand beliefs, values,
  and motives.
• Requires that sociologists observe their subjects
  from an outsiders point of view.
• Tension exists between the goals of subjectivity
  and objectivity.

13
Measurement

• Operationalization is the procedure by which
  researchers establish criteria for assigning
  values to variables.
• A variable is a concept that can take on more
  than one value.

14
Measurement

• Reliability is the degree to which a measurement
  yields consistent results.
• Validity is the degree to which a measure
  actually measures what it is intended to measure.
• Generalizability exists when research findings
  apply beyond the specific case examined.

15
Causality

• Causality means a change in the independent
  variable (x) produces a change in the dependent
  variable (y).
• We establish causality by demonstrating
• an association exists between x and y
• x precedes y and
• the introduction of a causally prior control
  variable does not result in the original
  association disappearing.

16
Experiments

• A carefully controlled artificial situation that
  allows researchers to isolate hypothesized causes
  and measure effects.

17
Experiments

• Use randomization to create two similar groups.
• The hypothesized cause is introduced to only one
  of the groups.
• The groups are compared before and after one of
  the groups has been exposed to the hypothesized
  cause.

18
Experiments

• An experimental group is the group that is
  exposed to the independent variable.
• A control group is the group that is not exposed
  to the independent variable.
• A dependent variable is the presumed effect in a
  cause and- effect relationship.
• An independent variable is the presumed cause in
  a cause-and-effect relationship.

19
Steps in Experiments

• Selection of subjects.
• Random assignment of subjects to experimental and
  control groups.
• Measurement of dependent variable in experimental
  and control groups.

20
Steps in Experiments

• Introduction of independent variable to
  experimental group.
• Remeasurement of dependent variable in
  experimental and control.
• Assessment of experimental effect.

21
Field Experiments

• Retain benefits of experimental design while
  overcoming validity problems.
• Either
• Introduce the independent variable themselves.
• Observe what happens when the independent
  variable is introduced in the normal course of
  social life.

22
Surveys

• A survey asks people questions in a face-to-face
  interview, telephone interview, or paper-
  and-pencil format.
• A sample is the part of the population of
  interest that is selected for analysis.
• A population is the entire group about which a
  researcher wishes to generalize.
• In a probability sample, the units have a known
  and nonzero chance of being selected.

23
Surveys

• Three main ways of conducting a survey
• self-administered questionnaire
• face-to-face interviews
• telephone interviews

24
Question

• If a university asks you to complete an
  anonymous, written survey asking questions about
  your sexual attitudes, experiences, and
  behaviors, how likely is it you will agree to
  complete the survey?
• Very likely
• Somewhat likely
• Unsure
• Somewhat unlikely
• Very unlikely

25
Threats to Validity of Surveys

• Exclusion of part of the population from the
  sampling frame.
• Refusal of some people to participate in the
  survey.
• Unwillingness of some respondents to answer some
  questions frankly.
• Asking leading or confusing questions or
  questions that refer to multiple, unimportant, or
  noncurrent events.

26
Question

• A probability sample is
• part of the population of research interest
  selected for analysis
• the entire group about which the researcher
  wishes to generalize
• a sample in which units have a known and nonzero
  chance of being selected
• a group of people who chose to be included in a
  survey

27
Answer

• A probability sample is
• c. sample in which units have a known and nonzero
  chance of being selected.

28
Contingency Table

• A contingency table is a cross classification of
  cases by at least two variables that allows you
  to see how, if at all, the variables are
  associated.

29
Turning a Classroom Into a Contingency Table
30
TV Viewing by Aggressiveness (in )
31
Variables

• An association exists between two variables if
  the value of one variable changes with the value
  of the other.
• A control variable is a variable whose influence
  is removed from the association between an
  independent and a dependent variable.

32
Variables

• A spurious association exists between an
  independent and a dependent variable when the
  introduction of a causally prior control variable
  makes the initial association disappear.

33
Testing an Association for Spuriousness
34
Analysis of Existing Documents Advantages

• Can save the researcher time and money because
  they are available in libraries or on the web.
• Usually cover entire populations and are
  collected using rigorous methods.
• Does not require live subjects.
• The researchers presence does not influence the
  subjects behavior.

35
Rank of Countries by Measures of Well-Being
36
Rank of Countries by Measures of Well-Being
37
Strengths and Weaknesses of Research Methods
38
Strengths and Weaknesses of Research Methods
39
Strengths and Weaknesses of Research Methods
40
Strengths and Weaknesses of Research Methods
41
Statistics You Should Know

• Mean (or arithmetic average).
• Median. The value of the case at the midpoint.
• A rate lets you compare the values of a variable
  among groups of different size.

42
Correlation

• The strength of the association between the two
  variables is measured by a statistic called the
  correlation coefficient (signified as r).
• The value of r can vary from -1.0 to 1.0.
• Whether positive or negative, the magnitude (or
  absolute value) of r decreases the more widely
  scattered the markers are from the line.

43
Correlation
44
Quick Quiz
45

• 1. Which of the following filters perception and
  ensures that we can never see society in a pure
  or objective form?
• values
• theories
• research methods
• all of these choices

46
Answer d.

• Values, theories and research methods filter
  perception and ensure that we can never see
  society in a pure or objective form.

47

• 2. The attempt to observe a social milieu
  objectively and take part in the activities of
  the people being observed is called
• the Hawthorne effect
• participant observation
• validation
• surveying

48
Answer b

• The attempt to observe a social milieu
  objectively and take part in the activities of
  the people being observed is called participant
  observation.

49

• 3. Which of the following is usually the first
  step in the research cycle?
• selecting a research method
• reviewing the literature
• formulating a research question
• collecting data
• none of these choices

50
Answer c

• Formulating a research question is usually the
  first step in the research cycle.

51

• 4. The question of whether researchers findings
  apply beyond the specific case examined is a
  question of
• reliability
• validity
• generalizability
• causality

52
Answer c

• The question of whether researchers findings
  apply beyond the specific case examined is a
  question of generalizability.

53

• 5. One of the conditions that must hold before
  the relationship between an independent and
  dependent variable can be considered causal is
• an association must not exist between the two
  variables
• the presumed cause must occur before the presumed
  effect
• when a control variable is introduced, the
  original association disappears
• association between the independent and dependent
  variables is spurious

54
Answer b.

• One of the conditions that must hold before the
  relationship between an independent and dependent
  variable can be considered causal is the
  presumed cause must occur before the presumed
  effect.