Chapter 2 Methodology - How Social Psychologists Do Research

Chapter 2: Methodology- How Social Psychologists Do Research A Fundamental Principle  Social influence can be studied scientifically  Social psychologists apply the scientific method Formulating Hypotheses and Theories  Like other scientists, social psychologists: o Develop theories (conceptual frameworks for explaining phenomena) o Derive hypotheses (specific predictions) from theory o Test hypotheses  Based on results, revise theory, formulate and test new hypotheses  Where do ideas (hypotheses) come from? o Previous theories and research  Dissatisfaction with existing theories (e.g. Festinger proposed dissonance theory) o Personal Observation- Kitty murder Research Designs  Observational- Description: What is the nature of the phenomenon? o Researcher observes people and systematically records behavior (describe behavior)  Ex. Ethnography- Description of group or culture from an “insider’s point of view”  Archival Analysis- Researcher examines accumulated documents (archives) of people, cultures, etc.  Correlational- Prediction: From knowing X, can we predict Y?  Experimental- Causality: Is variable X a cause variable of Y? Observational Method Example  Research Question- How much aggression do children exhibit during school recesses?  Method- Behaviors concretely defined  Observer unobtrusively and systematically looks for and records behaviors (ethnography)  Accuracy of observer is assessed o Interjudge reliability Limits of the Observational Method  Certain behaviors difficult to observe (e.g. willingness to help, prosocial behavior)  Does not allow prediction and explanation (e.g. if kids exhibit aggression during recess, why?) o Limited to description  Cronbach’s Alpha- What are the two disciplines of psychology: individual differences (correlational) psychology, experimental psychology The Correlational Method  Two or more variables are systematically measured and the relation between them is assessed (e.g. violent media watched and aggressive behavior in children) o Correlation coefficient- A statistical technique that estimates the strength of the relationship between variables  Range from -1.00 (perfect positive) to +1.00 (perfect negative)  Absolute size indicates strength of the correlation  0 means that two variables are not correlated, perfect is rare o Positive Correlation- Increases in the value of one variable are associated with increases in the value of the other variable  E.g. aggression and viewing violent media are positively correlated  More aggressive children tend to watch more violent television  Height and weight are positively correlated o Negative Correlation- Increases in the value of one variable are associated with decreases in the value of the other variable  Vaccination rate correlates negatively with disease rate Surveys  Research method (i.e. data source) in which a representative sample of people are asked about attitudes or behavior o Often used in correlational research  Correlations are computed using summed or averaged responses to questions o E.g. gun control attitudes correlated with voting preferences  Random Selection (i.e. sampling) o Ensures a representative sample by giving everyone in the population an equal chance of being selected for the sample o Answers are not useful if they are limited to the sample of people actually tested (nonrepresentative) Surveys: Advantages and Disadvantages  Advantages o Enable investigation of relations between variables difficult to observe (sexual behavior and knowledge of HIV) o Can sample representative segments of population  Disadvantages o Accuracy of responses  People may not know the answer but think they do Limits of the Correlational Method  Correlation does not equal causation  Correlational method tells us only that two variables are related  Social psychology’s goal o Identify causes of social behavior (A causes B, not just that A is correlated with B) Correlation Does Not Equal Causation  3 possible causal relations when a correlation is found (e.g. aggressive sexual acts and pornography) o X causes Y, X correlates with Y, an external variable is influencing  Porn causes viewer to become more sexually aggressive  People who are sexually aggressive are more interested in pornography  Correlation is caused by something else (e.g. upbringing, subculture) Experimental Method  Researcher randomly assigns participants to different conditions o Random assignment controls extraneous variables  Conditions are identical except for the independent variable (the one thought to have a causal effect on people’s responses) o Use to answer causal questions  Independent Variable- the IV is what researchers manipulate to see if it has a causal effect  Dependent Variable- the DV is the “effect”, what researchers measure to see if it is affected by the IV o Children aggression (DV), violent TV (IV) IV and DV Example using Latane and Darley  IV- Number of bystanders  DV- Helping behavior Internal Validity in Experiments  Making sure that nothing besides the IV can affect the dependent variable  Maximized by using tightly controlled lab conditions: o Control extraneous variables o Randomly assign people to experimental conditions Random Assignment  Ensures that: o All participants have equal change of being in any experimental condition o Differences in participant’s personalities or backgrounds are distributed evenly across conditions  This powerful technique is the MOST IMPORTANT PART of the experimental method Limits of Experimental Method  Experimental situations can be: o Artificial o Distant from real life (Tradeoff of increasing control over the situation to make it similar for all participants) External Validity in Experiments  The extent to which the results of a study can be generalized to: o Other situations (i.e. from the lab to real life situations) o Other people (i.e. from the participants to people in general) Generalizability Across Situations  Psychological Realism o Extent to which psychological processes triggered by experiments are similar to psychological processes in real life o May use a cover story (i.e. disguise true purpose) to make studies as realistic as possible (e.g. Milgram’s studies) Generalizability Across People  Random selection of participants from population (an ideal goal) o Impractical and expensive for most experiments o Replicate studies using different samples (e.g. from different cultures)