Task 8 Flashcards
Between-subjects design
Each treatment is administered to a different group of subjects
Within-subjects design
A single group of subjects is exposed to all treatments, one treatment at a time
Single-subject design
Same as within-subjects design, but data is focused on the behavior of a single subject under different treatment conditions
Error variance
The variability among scores caused by variables other than the independent variable
Single-factor designs
Only one independent variable is manipulated
Single-factor randomized-groups design
Subjects are assigned at random to different groups and each group is exposed to a different level of the independent variable
Matched Group design
Matched sets of of subjects are distributed at random, one subject per group, into the groups of the experiment
Sources of carryover effect:
Learning
Fatigue
Habituation
Sensitization
Contrast
Adaptation
Counterbalancing
The different treatments are assigned in a different order for different subjects
Complete counterbalancing
Provides every possible ordering of treatments and assigns at least one subject to each ordering. Every treatment follows every treatment equally often across subjects and every treatment appears equally often in each position
Partial counterbalancing
Includes only some of the possible treatment orders; there is a restriction that each treatment appears equally often in every position
Latin square design
Each treatment appears an equal number of times in each ordinal position
Minimize or eliminate carryover effect:
Pretrain subjects
Allowing time for subjects to adapt / habituate
Allowing breaks
Factorial design
Design that includes more than one independent variable
Mixed (or split-plot) design
Combines between-subjects and within-subjects factors
