Download as ppt, pdf, or txt
Download as ppt, pdf, or txt
You are on page 1of 19

Research Methods

Social Psychology

The Research Process

The Research Process


Observation Theory Generate Hypotheses (Predictions) Verification (test your idea) Revise Theory New hypotheses

Verification
Many forms of verification Correlation research:

Positive correlation: As one variable increases, so does the other Negative correlation: As one variable goes up, the other variable tends to go down This data is not causal.

Need for experimental research Experimental research: studies that

Verification: Interested in Cause?

seek cause-effect relationships by manipulating one or more factors

Typically conducted in lab settings

See what the data can tell us

Experimental Method: Develop Experiment---Variables

Independent variable (IV): A variable that is hypothesized to be the cause of a behavior and is typically the manipulated variable in experimental designs.

Dependent variable (DV): A variable that is the behavior or outcome of interest and is typically the measured variable in experimental designs.

Logic of Experimental Design & Lab Studies

***Holding everything else equal, if there is a difference between groups on dv, we can assume the I.V. is the cause of the difference.***

Vary Cause Variable Measure same D.V. in both groups

Group 1

High Frustration (Experimental condition) No Frustration (Control condition)

D.V.

Group 2 (different people than Group 1)

D.V.

Think about comparison .

Experimental Method:
Develop Experiment--Operational definition Operational definition:

Independent Variables

Current study I.V.s: Alcohol & Age Alcohol (Drink vs. No Drink) Age of Person Being Evaluated (Young vs. Old)
I.V. 1
Alcohol Yes Age of Person Being Evaluate d Young No

I.V . 2

Old

Independent Variables

Can have levels within an I.V. Current study IVs: Alcohol & Age Alcohol (Drink vs. No Drink) Age of Person Being Evaluated (Young vs. Old)
I.V. 1
Alcohol Yes Age of Person Being Evaluate d Young No

I.V . 2

Old

What if I was interested in not just old and young?

Independent Variables

Can have levels within an I.V. Current study IVs Alcohol & Age Alcohol (Drink vs. No Drink) Age of Person Being Evaluated (Young vs. Old)
I.V. 1
Alcohol Yes Age of Person Being Evaluate d Young No

I.V . 2

Old

Independent Variables

Back to original study Current study IVs: Alcohol & Age Alcohol (Drink vs. No Drink) Age of Person Being Evaluated (Young vs. Old)
I.V. 1
Alcohol Yes Age of Person Being Evaluate d Young No

I.V . 2

Old

Independent Variables

Varying quantity of Alcohol? Current study IVs: Alcohol & Age Alcohol (Drink vs. No Drink) Age of Person Being Evaluated (Young vs. Old)
I.V. 1
Alcohol

I.V . 2

Age of Person Being Evaluate d

Young

Old

Independent Variables

Describing experimental research


Based on (a) number of independent variables in the study & (b) number of levels of each IV

A 2 X 2 design
I.V. 1 Alcohol Yes Age of Person Being Evaluate d Young No

I.V . 2

Old

Independent Variables

Describing experimental research


Based on (a) number of independent variables in the study & (b) number of levels of each IV

How would you describe this design?


I.V. 1
Alcohol Yes Age of Person Being Evaluate d Young No

I.V . 2

MiddleAged
Old

Independent Variables

Describing experimental research


Based on (a) number of independent variables in the study & (b) number of levels of each IV

How would you describe this design?


I.V. 1
Alcohol 1 Age of Person Being Evaluate d Young 2 3 4

I.V . 2

Old

Independent Variables

Describing experimental research


Based on (a) number of independent variables in the study & (b) number of levels of each IV

Need to be able to look at numbers at know what the design is Knowing how to describe designs Two steps
(a) Count how many numbers are there (b) look at the actual number to determine levels of each IV

What is?
5 X 3 design? 3 X 2 design? 5 X 4 X 5 X 6 X 7 design? 5 X 4 X 6 design?

Evaluating research

Lab vs. Field study on Mood & Helping Confidence in ability to say mood caused helping?

Internal validity: the extent to which


conclusions can be drawn about the causal effects of the IV on the DV.
Does the design allow us to demonstrate that the IV actually caused the DV? In research with high internal validity, we are more able to argue that the IV caused the DV, whereas in studies with low internal validity, causality cannot be inferred

External validity:

Evaluating research

Internal validity: the extent to which


conclusions can be drawn about the causal effects of the IV on the DV.
Does the design allow us to demonstrate that the IV actually caused the DV? In research with high internal validity, we are more able to argue that the IV caused the DV, whereas in studies with low internal validity, causality cannot be inferred

Lab Research & Validity:


High Internal validity, but low external validity

Field Research & Validity:

Summary/Recap

IV/DV
Describing designs

Operationalize variables Experimental vs. Correlation Evaluating research

You might also like