
Advantages of BetweenSubjects Designs
Includes a control or comparison group that helps to control for many extraneous variables.
Can randomly assign participants to groups, which helps to balance extraneous variables across the groups.

Disadvantages of BetweenSubjects Designs
You can’t always randomly assign participants to groups.
Large individual differences between participants may make it difficult to observe meaningful group differences.

If the subjects are used more than once or matched, this design is called a
WithinSubjects Design

Advantages of WithinSubjects Design
They take fewer participants.
They typically have more statistical power (such as a matched ttest)

Disadvantages of WithinSubjects Design
You have to worry about practice effects and carryover effects

equalizing the amount of practice, for example, you could test one person in the B1, B2, B3 order, another with B1, B3, B2, until every order is tested; used to avoid practice and carryover effects
counterbalancing

a factorial design in which there is at least one betweensubjects factor and at least one withinsubjects design
mixed design

Mixed Designs are useful when:
You might end up testing the subjects too much in a wholly withinsubjects design

total variance=
systematic + nonsystematic variance

"between group" variance; experimental variance due to the IV; due to confounding variables
systematic variance

error variance or variance within the groups; due to chance factors
nonsystematic variance

If betweengroup variation is not significantly ______ than withingroup variance, then group differences probably do not contribute anything beyond chance variability
greater

statistical anaylsis comparing more than 2 sets of scores
ANOVA

Sources of sequence/carryover effects
learning, fatigue, habituation, sensitization, contrast, adaptation

ways to deal with carryover effect
random order; counterbalancing

advantages of matched design
can control for subject variables that may otherwise obscure the effect of the IV

disadvantages of matched design
matching variable needs to be reliable, difficult finding enough participants who are matched, impossible to match on every possible variable that might influence the IV

when greater than 2 IV, we can examine _______ and ______
main effects; interactions

overall effect of one IV (ignoring the other)
main effect

effect of one IV depends on the level of a second IV
interaction effect

