PL3105 Midterm 1.1 and 1.2

Social Psychology

Social Psychology

The scientific study of the way people’s thoughts, feelings, and behaviours are influenced by the real or imagined presence of other people 

• internal processes and external outcomes

• Emphasis on context for behaviour (situational forces)

Most effects in social psych are between ___ and ___ in magnitude

Weak and medium

• Effect size: r between .10 and .30 → explains 9% variance approx. 

  • Makes sense as if effect is large (ex. r=1.00), means that social factors are deterministic to our behaviors 

Harry Harlow’s Contact Comfort Theory

Cloth vs. wire mother

• refuted existing “cupboard theory” (primates attached to caregivers because they provide food)

• While the monkeys reached out to wire mother for food, majority of time spent clinging to cloth mother 

Hindsight bias

Exaggeration of how event could be predicted before it occurred 

• Results tend to feel predictable after the fact!

Three Types of Methods

1. Observational Method (Description)

• Asks about the nature of the phenomenon 

• Observes people + records measurements/impressions of behavior

2. Correlational Method (Prediction) 

• 2+ variables systematically measured, relationship between variables is assessed (from knowing x, can we predict y) 

• Correlation Coefficient: Assess how well one variable can be predicted from another, ranges from -1 to +1

3. Experimental Method (Causality) 

• measures using IV and DV

Observational Method Types

1. Ethnography: Understand group by observing from the inside (wo imposing preconceived notions) 

• Consistent observations, ensures objectivity 

• Observation, ecological momentary assessments (EMA), electronic activated recorder (EAR) 

2. Archival analysis: Examine accumulated documents (ex. diaries, speeches, music lyrics) of a culture

• Limitations: 

  • Certain behaviours are difficult to observe → may only occur rarely/in private

  • Reveals patterns but does not give an explanation why its happening 

Correlational Method Types

Surveys: Representative sample asked questions about attitudes and behaviour 

• Pros: 

  • Allows researchers to judge relationship b/w variables that are difficult to observe (ex. Sexual behaviour) 

  • Able to sample representative segments of population (most effective through random selection)

• Cons:

  • Self report —> Not always accurate

Internal validity

Nothing besides IV effects DV→ observed effect is caused by what we thought 

• Random assignment to condition (Equal chance of being in any condition) to minimize participant differences, lab control 

External validity

Ability to generalize to real life conditions (tricky for experiments done in artificial settings, ex. lab) 

• Generalizability across situations + people → can improve through random selection of participants from population

• Field experiments: Researchers study behaviour outside of lab (in natural setting) 

  • Same design as lab experiment, just different environment 

  • High external validity 

Psychological realism

Extent to which psychological processes triggered in experiment are similar to psychological processes occurring in everyday life 

Probability level (p-value)

•  Likelihood that results of experiment occurred by chance rather than independent variable 

  • Results are significant if probability is less than 0.05 that results are due to chance 

Tests of Validity

1. Replications: Tests external validity. Replication of the same study, different populations/settings

2. Meta-Analysis: Tests internal validity. Average results of 2+ studies (p-values), see if effect of IV is reliable.

Guidelines for Ethical Research: Institutional Review Board (IRB)

• Protects safety + dignity of participants 

• Includes at least one scientist, one non-scientist, and one person who is not affiliated with institution 

• Must review and approve all research before it is conducted 

• Procedures judged to be overly stressful/upsetting must be changed or deleted

The Open Science Movement

• Focused on making the scientific process more open to other scientists + the public. Share methods, data, and data analysis files 

• Attempts to change incentive structure (publish or perish)

• Null studies and replication studies are being published in high impact journals (null findings as important as statistically significant studies)

• Improved science research practices to reduce false positives → cultural change for scientists to do good science instead of a lot of science Allows for replicability of research findings 

Reproducibility Project (2015)

Large scale replication, 100 studies from 3 different journals, close/exact replications 

• Teams around the world worked on this project from open materials and data 

• Out of the 100 studies conducted, less than half replicated (39)

• Ranges from 30-77% in other versions of this project