LECTURE 1: Research Methods in Social Psych
Learning to be a Scientist
Jan 15 2013
Check this out!!
Teenagers and sex
- Teens exposed to sex more likely to engage in sex after
- Alternative thinking is a way to critically analyze certain theories or assumption
Goals of Science
a. Predict (hypothesize outcome using various known facts and elements)
c. Control (prisoners and human rights are examples)
The Scientific Method:
- Provides a number of logical steps that allow us to make statements about causal relations
between two or more variables
- Hume’s Rules for Causality
• A means of distinguishing between explanatory (causal) versus descriptive
Punishment causes aggression
- Does punishment cause aggression
Punishment at Home(per day) Teacher’s Ratings of Aggressiveness
• 1 to 5 times • 3
• 6-10 times • 6
• 11-15 times • 9
• 16-20 times • 12
• 21 or more times • 15
- Perfect correlation (therefore) accurate assumption?
• A could cause B or
• B could cause A
• Because child is more aggressive, they will be further punished
- Third factor too: financial situation; environment
- Could also be genetics too.
• C causes A and B
Hume’s Rule (cont)
1. Constant Union -- As A varies B varies
• (Descriptive or correlation) • Need to show more correlation
2. Temporal Antecedent -- a change in A must always precede a change in B
3. Spatial-Temporal Contiguity-- A and B must be shown to co-occur in space and time
The Church Clock Fallacy (Dawkins)
- Has all these elements; constant union, temporal antecedence, spatial-temporal contiguity
- So what do we do?
- Well, we experiment, using the scientific method.
• Have one clock ring and see if other does
• Have one ring randomly and see if other follows rings
Four Steps of the Scientific Method
1. Formulate Hypothesis(es): Changes in one or more variables cause a change in behaviour.
- What do we call the cause
• Independent variable
- What do we call the effect
• Dependent Variable
Note: A hypothesis is a statement about indirectly observable behaviour.
- E.g., Punishment causes aggression; why indirectly observable?
• Abstract concept – we now must make them concrete
2. Deduce consequences from the hypothesis. Make a statement about directly observable
- Heart of experimentation
- Operational definition of independent and dependent variable
- research hypothesis: observable version of Step 1
Let’s use our classroom situation discussed earlier, and our hypothesis that punishment causes
a. What would be an operational definition of our independent and dependent variables,
b. Our research hypothesis version of our hypothesis that punishment causes aggression?
- We could have:
• Punishment – no recess
• Aggression – Yuan’s aggression scale (YAS)
3. Observation time
- X = treatment,
- O = Observation,
- O1 = Observation 1,
- O2 = Observation 2
- In this case:
• X = no recess
• O = YAS
Design 1: One case design
- X, O1 - OK??
• Immediate observation
• Not good as need to know behavior after and before
Design 2: Pretest-Post-test Design
- O1, X, O2
• Compare pre and post treatment
Not as good as sleeping time and wakefulness need to be measured – 3 variable rd
Design 3: No Treatment Control Added
- MRS. BROWN’S CLASS 01 X 02
- MR. BROWN’S CLASS 01 02
• Can test another class
Not as good as environments may differ -> selection factors
- R = Random assignment added to Pre-test Post-Test design
- R , O1, X, O2
- R, O1, , O2
• Random assignment of student into recess and no recess group
• GOOD TEST
- Post-test Only Design (with random assignment)
- R, X, O1
- R, , O1
- Do not necessarily need pre-test as less chance of influencing subjects
4. Make statement about hypothesis based on the observations
- Internal validity: systematic source of error or can we conclude that we have causation?
• random assignment
- External validity : even if we have shown causation, does the phenomenon generalize to the
• What kinds of things do we need to establish external validity?
must do experiment again; replication – different age, populations, etc.
- Ethics – cannot slap a kid, scare a kid, harm them, etc. – Ethics Review Board
- The social psychology of the psychological experiment: demand characte