Chapter 8: Experimental Designs, Between-Subjects Design Flashcards
(49 cards)
between subjects experiments
uses a separate group of individuals for each of the different treatment conditions
the different groups of scores all can be obtained from same group of participants
within subject design
between subject designs are also commonly used for
nonexperimental and quasi experimental designs but they do not contain a manipulated variable
between subject design allows only one score for
each participant, for each level of the independent variable
advantages
measurements is relatively clean and uncontaminated by other treatment factors. for this reason we also call it independent measures experimental design
clean of these factors
-practice or experience in other treatments
-fatigue or boredom
-contrast effects that result from comparing one treatment to another
between subject designs can be used for a wide variety of research questions
thus a between subject design is always an option. It may not always be the best choice.
disadvantages
they require a relatively large number of participants. it is a problem when potential participants is relatively small
the primary disadvantage
is individual differences
two major concerns about individual differences
1 individual differences can become confounding variables.
2 individual differences can produce high variability in the scores (high variance) making it difficult to determine whether the treatment has any effect
there are two major sources for confounding variable in between subject designs
1 confounding from individual differences
2 confounding from environmental variables
the separate groups must be
1 created equally
2 treated equally
3 composed of equivalent individuals (the characteristics must be as similar as possible between groups)
Limiting confounding
randomization
matching groups
holding variables constant or restricting range of variability
restricted random assignment
the group assignment process is limited to ensure predetermined characteristics (such as equal size) for the separate groups
matching groups
1 identification of the variable to be matched across groups
2 measurement of the matching variable for each participant (IQ etc)
3 assignment of participants to groups by means of restricted random assignment that ensures a balance between groups
but it can be difficult or impossible to match groups on several different variables simultaneously.
holding variables constant
by using only female participants, threat to external validity
individual differences
have the potential to produce high variability in the scores
difference between treatments is described by computing the average score for each treatment, then comparing the two averages.
however simply comparing two averages is not enough to demonstrate a noticable difference. The problem comes from sometimes 10 point difference is large sometimes is small
variance
is a statistical value that measures the size of the differences from one score to another. If the scores all have similar values variance is small
with small variance
the 10 point difference between treatments shows up clearly
with large variance
10 point treatment effect is completely obscured
when a research study has a lot of variance
it is difficult to see real treatment effect, individual differences causes variance.
big differences between treatments are
good because they provide evidence of differential treatment effects. decreases variance
big differences within treatments are
bad because the differences that exist inside the treatment condition determine the variance of scores. increases variance
