PSY 290A Study Guide - Final Guide: Observer-Expectancy Effect, External Validity, Construct Validity
13 Dec 2016
School
Department
Course
Professor
Final Exam Study Guide
Chapter 13:
1. Explain how quasi-experiments can be either independent-groups designs or within-
groups designs.
• Independent-groups design:
o at least one treatment and one comparison group, but participants are not
assigned randomly
• Within-groups design (repeated-measures design):
o participants experience all levels of independent variables; researcher
takes advantage of an already-scheduled event, a new policy or
regulation, or a chance occurrence to manipulate the independent
variable
2. Define the following quasi-experimental designs: nonequivalent control group design,
interrupted time-series design, and nonequivalent groups interrupted time-series
design.
• Nonequivalent control group design:
o An independent-groups quasi-experiment that has at least one treatment
group and one comparison group, but participants have not been
randomly assigned to the two groups.
• Nonequivalent control group design pretest/posttest :
o An independent-groups quasi-experiment that has at least one treatment
group and one comparison group, in which participants have not been
randomly assigned to the two groups, and in which at least one pretest
and one posttest are administered.
• Interrupted time-series design:
o A quasi-experiment in which participants are measured repeatedly on a
dependent variable before, during, and after the "interruption" caused by
some event.
• Nonequivalent control group interrupted time-series design:
o A quasi-experiment with two or more groups in which participants have
not been randomly assigned to groups; participants are measured
repeatedly on a dependent variable before, during, and after the
"interruption" caused by some event, and the presence or timing of the
interrupting event differs among the groups.
3. Explain whether quasi-experimental studies avoid the following threats to internal
validity: selection, maturation, history, regression, attrition, testing, instrumentation,
observer bias, experimental demand, and placebo effects.
• Selection effects:
o can be avoided through matched groups and wait-list design.
• Design confounds:
o can be ruled out by inspecting data carefully
• Maturation:
o can be ruled out through a comparison group
• History:
o can be ruled out through a comparison group
• Regression:
find more resources at oneclass.com
find more resources at oneclass.com
o can happen if the groups are selected based on extreme high scores or
extreme low scores
• Attrition:
o can be controlled for by systematically checking the participants' scores
that dropped out, if there is a systematic drop out there could be an
attrition effect
• Testing and Instrumentation:
o can be ruled out through a comparison group
• Observer bias, demand characteristics, and placebo effects:
o observer bias can be ruled out through blind and double-blind designs
4. Using both the design and the results, analyze whether a quasi-experimental design
allows you to rule out internal validity threats.
• Let's researchers take advantage of real-world opportunities
• Enhanced external validity, if study takes place in a real-world setting
researchers don't have to ask if the study applies to the real-world
• Ethical concerns can be eliminated using quasi-experiments
5. Explain the trade-offs of using a quasi-experimental design.
• Internal validity is not as important in quasi-experiments, however the external
and construct validity is usually really good
6. Interrogate quasi-experimental designs by asking about construct validity, external
validity, and statistical validity.
• Construct validity:
o involves assessing whether the variables were manipulated and measured
in ways consistent with the theory behind the experiment.
• External validity:
o involves asking whether the experiment's results can be generalized to
other people or to other situations and settings.
• Statistical validity:
o starts by asking how strongly the independent variable affects the
dependent variable (effect size) and whether the effect is statistically
significant.
7. Explain three differences between small-N and large-N experiments.
• Small-N:
o Small number of participants
o Each participant is treated as a separate experiment
o Individuals' data are presented
• Large-N:
o Large number of participants
o Participants are grouped
o Data are presented as group averages
8. Describe three small-N designs (stable-baseline designs, multiple-baseline designs, and
reversal designs) and explain how each design addresses internal validity.
• Stable-baseline design:
o A small-N design in which a researcher observes behavior for an extended
baseline period before beginning a treatment or other intervention; if
find more resources at oneclass.com
find more resources at oneclass.com
