Social Psych (Aronson) - Chapter 2: Methodology

Zero (Null) Correlation

(Slope=__)

Correlational Method: Predicting Social Behavior

A Method by which two or more variables are systematically measured and their relationship assessed.

p-value

A number, calculated using statistical techniques, that tells how likely it is that the results of their experiment occurred by chance and not because of the independent variable

HOMER (Hypothesize)

A testable prediction about conditions under which an event will occur

HOMER (Operationalize)

A variable defined in specific terms of the operations or techniques the researcher uses to measure or manipulate it

Confounding Variable

An extraneous variable that varies along with the IV - a consistent, unmeasured difference between conditions.

Variable

Any event, situation, behavior, or individual characteristic that varies

Extraneous

Anything other than the IV that could affect the DV

Reliability

Are the measurements obtained by the study consistent? Over time? Across multiple judges/raters?

Empirical

Based on careful observation and experimentation; must be falsifiable by these methods as well Empirical research allows us to test the validity of personal observations and folk wisdom/common sense beliefs.

Advantages of Naturalistic Observation

Behaviors are spontaneous, Doesn't rely on self-reports

Advantages of Surveys/ Questionnaires

Can assess behaviors that are hard to observe, Can measure subjective states (perceptions, beliefs, etc)

Description

Careful and systematic descriptions of social behavior so we can make reliable generalizations about how people act in various social settings.

HOMER (Evaluate)

Condition averages always will differ

Replication

Conducting a study again with different subject populations or in different situations

Independent variable (IV)

Controlled by researcher and interpreted as the cause of change in DV (Aka "manipulated variable")

Disadvantages of Correlation

Correlation ≠ Causation, Reverse-causality problem, Third-variable problem

Disadvantages of Lab Experiments

Costly, Artificial, Can be biased

Four Goals of Research?

Description, Causal Analysis, Theory Building, Application

Theory Building

Develop theories about social behavior that help us understand why people behave the way they do. Theories help psychologists to organize what they know about social behavior as well as suggest new predictions that can be tested in further research.

Conditions

Different levels of the IV (Ex. Lucky charm is one condition, no lucky charm is another)

Correlation ? Causation

Does not equal

Validity

Does the study accurately measure the relationship (and ONLY the relationship) the researcher wants to study?

Advantages of Archival Analysis

Easy access to large amounts of data, Can track changes over time, Cheap

Advantages of Correlation

Enable researchers to study problems in which intervention is impossible or unethical ie divorce, death in the family, Unlike observation, allows for prediction, Efficient: allow researchers to collect more information and test more relationships

Correlational Method

Examines the association between two variables

External Validity

Extent to which results of a study can be generalized to other situations

Advantages of Field Experiment

High external validity

Advantages of Lab Experiments

High internal validity, experimenter control

HOMER

Hypothesize, Operationalize, Measure, Evaluate, Replicate or Revise

(−) Negative correlation

Increase in one variable associated with decrease in other. Maximum: -1 (Slope= \)

(+) Positive correlation

Increase in one variable associated with increase in other. Maximum: +1 (Slope= /)

The APA's ethical guidelines for research

Informed Consent, Debriefing, Cost/benefit analysis, Institutional Review Board (IRB)

Disadvantages of Naturalistic Observation

Interference, Some interesting behaviors are very rare, Observer bias (coding system needed), Time consuming, Invasion of privacy

Solutions to Subject Bias

Keep participants unaware of hypotheses; double-blind study

Two Types of Experimental Method: Formats

Laboratory Experiments and Field Experiments

Experimental Method

Manipulates one variable to see the effect on another

Disadvantages of Archival Analysis

Many behaviors not recorded, Questions can be different than researchers want, May be wrong population

Three primary methods

Observational method, Correlational method, Experimental method

Dependent variable (DV)

Outcome being studied, response to IV (Aka "measured variable")

Random assignment

Practice of assigning subjects to conditions such that each subject has an equal chance of being in any condition

Disadvantages of Field Experiment

Random assignment difficult, Low experimenter control

Convergent validity

Replication in different settings and experiment types

Methodology and Research

Research methodology refers to the proper techniques to design a study that can help answer a given research question

Archival Analysis

Researcher examines accumulated documents of a culture

Participant Observation

Researcher interacts with people being observed, but tries not to change the situation

Naturalistic Observation

Researcher observes people in their natural habitat

Surveys/Questionnaires

Self-Report, Most common way of obtaining correlations

Disadvantages of Surveys/ Questionnaires

Self-presentation, Lack of awareness of true state

Application

Social psychological knowledge can help to solve everyday social problems

Correlation coefficient

Statistic that assesses how well you can predict one variable from another (r)

Meta-analysis

Statistical technique that averages results of two or more studies

Quasi-experiments

Structured like an experiment, but the "IV" is unethical or impossible to manipulate (Most common quasi-IV: gender)

Experimenter Bias

Subtle conscious or unconscious cues from experimenter that tell participant how to act

A strong hypothesis will be?

Testable, Falsifiable, Parsimonious, Generative, Important

Experimental condition

The "changed" group - the variable is altered in a way the experimenter believes will cause a change in the DV

Control condition

The "unchanged" group - used to establish a baseline for comparison

Causal Analysis

Tries to establish relationships between cause and effect

Observational Method

Used to describe the nature of a phenomenon

Subject Bias

motives and goals make them act differently than they normally would. Try to be "the good subject" or the "bad subject"