2. Methodology: How Social Psychologists Do Research Flashcards
Fundamental Principle
Social influence can be studied scientifically
Why might results of some experiments may seem obvious
-Familiarity with the subject matter
-Social influence
-Social behavior
- Hindsight bias
Hindsight bias
The tendency to exaggerate prediction of an outcome after knowing that it occurred
Like other scientists, social psychologists
Develop theories
Derive hypotheses from theory
Test hypotheses
Test hypotheses
Based on the results, revise theory
Formulate and test new hypotheses
How to: formulate Hypotheses and Theories
Previous theories and research
Science is cumulative
Dissatisfaction with behaviorism (Festinger)
Personal observation
Kitty Genovese (Latané and Darley)
The Observational Method
-Researcher observes people and systematically records behavior
-Used to describe behavior
ie: Ethnography:Description from an “insider’s point of view”
ie: Archival analysis: Researcher examines accumulated documents (archives)
Research Question
How much aggression do children exhibit during school recesses?
Interjudge Reliability
The level of agreement between two or more people who independently observe and code a set of data
Limits of the Observational Method
Certain behaviors difficult to observe
-Occur rarely
- In private
Archival analysis
-Original may not have all information researchers need
-Does not allow prediction and explanation
- Limited to description
The Correlational Method
A statistical technique that assesses how well you can predict one variable from another
Positive Correlation
Increases in the value of one variable are associated with increases in the value of the other variable
neg correlation
Increases in the value of one variable are associated with decreases in the value of the other variable
The Correlation Coefficient
Correlation coefficients range from –1.00 to +1.00
+1.00 perfectly correlated in a positive direction
0 means that two variables are not correlated
–1.00 perfectly correlated in a negative direction
Surveys
Representative sample of people asked about attitudes or behaviors
Correlations computed using responses to questions
Random Selection
A way of ensuring that a sample of people is representative of a population by giving everyone in the population an equal chance of being selected for the sample
Surveys: Disadvantages
Accuracy of responses:
People may not know the answer—but they think they do!
Surveys: Advantages
Investigate relations between variables difficult to observe
Sexual behavior & knowledge of HIV
Sample representative segments of population
Limits of the Correlational Method
Correlation does not equal causation!
Correlational method tells us only that two variables are related
Social psychology’s goal
Identify causes of social behavior
Be able to say that A causes B, not just that A is correlated with B
A Birth Control and STD Correlation
A study conducted in the early 1990s found a correlation between the type of birth control women used and their likelihood of getting a sexually transmitted disease (STD). Those whose partners used condoms were more likely to have an STD than were women who used other forms of birth control. Does this mean that the use of condoms caused the increase in STDs? Not necessarily—see the text for alternative explanations of this research finding.
The Experimental Method
Researcher randomly assigns participants to different conditions
Conditions are identical except for the independent variable (the one thought to have a causal effect on people’s responses).
Use to answer causal questions
Independent Variable (IV)
The IV is what researchers manipulate to see if it has a causal effect
(e.g., type of TV show children watch)
Dependent Variable (DV)
IV and DV Example Using Latané and Darley (1970)
Probability Level (p-value)
A number calculated with statistical techniques
Indicates likelihood results of experiment occurred by chance instead of the IV(s)
The convention in science is to consider results significant when
Probability is less than 5 in 100 that the results might be due to chance factors and not the IV
p< .05
