# PSYB01H3 Chapter Notes - Chapter 8: Repeated Measures Design, Confounding, Random Assignment

23 views3 pages
Published on 11 Aug 2012
School
UTSC
Department
Psychology
Course
PSYB01H3
Professor
Chapter 8 Experimental Design
- Variables are kept constant by either experimental control or by randomization
- Confounding variable- variable that varies along with the independent variable
o Effects of the independent variable and an uncontrollable variable are
intertwined so you can't determine which of the variables is responsible for
the observed effect
- Internal validity- ability to draw conclusions about causal relationships from our
data
POSTTEST-ONLY DESIGN
- 2 equivalent groups of Ps are used
o To avoid selection differences- people selected to be in the conditions can't
differ in any systematic way
o Randomization or having same P do both conditions
- Introduce the independent variable
o Experimental group gets the treatment
o Control group doesn't
- Measure the effect of the independent variable on the dependent variable
o Same measurement is used for both groups
o Statistically significant test can be used
PRETEST-POSTTEST DESIGN
- In this test, the pretest is given before the experimental manipulation is given
- Makes sure that the groups were equal in the beginning of the experiment
- Not necessary is Ps were randomized to the groups
- Larger the sample, the more equal they will be when randomized
- Minimum of 20 to 30 Ps
- Pretest may be necessary to select the Ps in the experiment
- Focuses on the CHANGE from pretest to post-test
- Needed if there's a possibility that the P will drop out (attrition/ mortality) of the
experiment
- Can be too disadvantage or can sensitize Ps to your experiment
INDEPENDENT GROUPS DESIGN
Unlock document

This preview shows page 1 of the document.
Unlock all 3 pages and 3 million more documents.

## Document Summary

Variables are kept constant by either experimental control or by randomization. Confounding variable- variable that varies along with the independent variable: effects of the independent variable and an uncontrollable variable are intertwined so you can"t determine which of the variables is responsible for the observed effect. Internal validity- ability to draw conclusions about causal relationships from our data. 2 equivalent groups of ps are used: to avoid selection differences- people selected to be in the conditions can"t differ in any systematic way, randomization or having same p do both conditions. Introduce the independent variable: experimental group gets the treatment, control group doesn"t. Measure the effect of the independent variable on the dependent variable: same measurement is used for both groups, statistically significant test can be used. In this test, the pretest is given before the experimental manipulation is given. Makes sure that the groups were equal in the beginning of the experiment.