Textbook Notes (280,000)
CA (170,000)
U of G (10,000)
PSYC (3,000)
PSYC 2360 (100)
Chapter

PSYC 2360 Chapter Notes -Statistical Conclusion Validity, Blind Experiment, Internal Validity


Department
Psychology
Course Code
PSYC 2360
Professor
Naseem Al- Aidroos

This preview shows pages 1-2. to view the full 6 pages of the document.
Research Methods - Internal Validity
Validity Threats
1. Construct validity
2. Statistical conclusion validity - whether or not you are drawing valid
statistical conclusions
3. Internal validity - concluding validly that dependent variable is caused by
independent variable
4. External validity - with what confidence can conclude that results generalize
The Importance of Experimental Control
Occurs when extent that the experimenter is able to eliminate the effects on
the dependent variable other than the effects of the independent variable
o Rare to have something that is single-y caused in psychology
Greater the experimental control, the more confidently we can say that it is
the independent variable rather than something else, that causes the
dependent variable
Created through equivalence between groups
Extraneous Variables - has effect on dependent variable only and makes it
seem that the independent variable is causing the dependent variable
o Can still have internal validity
o Constitutes random error
Confounding Variables - effects both the independent and the dependent
variables
o Problem of a third variable, can no longer conclude in valid fashion
that independent variable causes the dependent variable
Internal Validity
Independent variable causes dependent variable what you can conclude if
your experiment is internally valid
Internal validity is ensured only when there are no confounds
Confound alternative explanation
o Another variable is mixed up with the independent variable
o Does not necessarily mean that the independent variable does not
cause the dependent variable but always produces alternative
explanations
And you don’t know which one is impacting dependent
variable or if both were

Only pages 1-2 are available for preview. Some parts have been intentionally blurred.

Extraneous Variables - People are Different
Problem of individual differences
o Differences in disgust responses, response / reaction times
o Probably not going to be able to control for all extraneous variables
Limited Population Designs
o Select from limited and therefore homogeneous population (e.g.
always using college students)
o Typically used out of convenience but group is normally close in age,
live in similar environments, have similar socioeconomic statuses,
similar education background therefore there is reduced variability
and limiting influence of extraneous variables
Before-After Research Designs
o Often done in reaction time research and in clinical research
People act as their own controls
o Any differences among participants will influence both the baseline
memory measure and the memory measure that serves as the
dependent variable
o Dependent variable is measured at two different times and design
increases power of experiment by controlling for variability among
participants
o Difference between this and repeated measures is that individual is
only in one condition of experiment and first measurement of
dependent variable is baseline
o Problems?
Crates possibility of retesting effects
Fatigue, reactivity, develop strategies to do better on
second test, increased likelihood they will guess
research hypothesis
Matched Group Designs
o Participants are measured on the variable of interest before
experiment and then are assigned to conditions on the basis of their
scores on that variable
o Reduces differences between the conditions on the matching variable
therefore increases power of statistical test
o Participants can also be matched through more than one variable
o As long as matching variable actually correlates with the dependent
measure, the use of matched group design will reduce random error
and increase statistical power of the design
o Not always necessary
All participants
study and recall list
A
Participants
randomly assigned
to conditions
Study list B using
sentences Recall list B
Study list B, no
instructions Recall list B
You're Reading a Preview

Unlock to view full version