Nonexperimental Research Methods

1
Nonexperimental Research Methods
2
Archival Research

• Examine previously recorded data to answer a
  research question
• Census data
• General Social Survey
• Advantages?
• Disadvantages?

3
Observational Research

• Case studies
• Intensive observation of one individual or a
  small group of individuals
• Cannot generalize beyond original participants
• Naturalistic observation
• Observing behavior in the real world
• Hawthorne effect or reactance

4
Observational Research

• Participant observation
• Observer becomes part of the group being studied

5
Observational Research

• Problem How do you make observations?
• What do you observe?
• When do you observe it?
• How do you record your observations?
• Qualitative Narrative records
• Quantitiative Assign some measurement to
  behaviors

6
Interobserver Reliability

• When you use multiple observers, you need to
  assess how often they agree
• Number of agreements divided by the number of
  opportunities to agree
• Correlation between observers judgments

7
Correlational Research

• Quantifies the strength of a relationship between
  two variables
• How does the value of one variable change when
  the value of another variable changes?

8
Correlational Research

• Many types of correlations
• Pearson product-moment correlation
• Spearman rank-order correlation
• Advanced techniques
• Multiple correlation
• Path analysis
• Partial correlation
• Point-biserial correlation

9
Pearson product-moment correlation

• Represented by a mathematical score
• Ranges from 1.0 to -1.0
• Absolute value signifies strength of relationship
• Sign signifies nature of the relationship

10
Pearson product-moment correlation

• Strength of a relationship is represented by a
  mathematical score
• 1.0 Perfect positive correlation

11
Pearson product-moment correlation

• Strength of a relationship is represented by a
  mathematical score
• 1.0 Perfect positive correlation
• -1.0 Perfect negative correlation

12
Pearson product-moment correlation

• Strength of a relationship is represented by a
  mathematical score
• 1.0 Perfect positive correlation
• -1.0 Perfect negative correlation
• 0.0 No correlation No relationship!

13
Pearson product-moment correlation

• Which is stronger?

14
Interpreting Correlations

• What does strength of a correlation mean?
• Rule of thumb

• .8 to 1.0
• .6 to .8
• .4 to .6
• .2 to .4
• .0 to .2

• Very strong relationship
• Strong relationship
• Moderate relationship
• Weak relationship
• Weak or No relationship

15
Interpreting Correlations

• Large-scale study of contraceptive use in Taiwan
  found that people with more electrical appliances
  were more likely to use birth control.

16
Interpreting Correlations

• When we find a relationship between two variables
  (A and B), there are three possible explanations
• Changes in A cause changes in B
• Changes in B cause changes in A
• Changes in a third variable C cause changes in
  both A and B

17
Interpreting Correlations
18
Interpreting Correlations
19
Interpreting Correlations
20
Interpreting Correlations

• Correlations allow us to describe relationships
• Correlations allow us to predict

21
Interpreting Correlations
22
Correlation Coefficient

• Reflects the amount of variability that is shared
  between variables
• Beware the problem of restricted range
• Correlations are lower because available
  variation is restricted

23
Interpreting Correlations

• Coefficient of Determination
• Percentage of variance in one variable that is
  accounted for by variance in the other variable
• Computed by squaring the correlation coefficient