Different types of experimental designs Flashcards
manipulating a variable quantitatively
–> changing the amount of the variable to which the group of praticipants is exposed –> parametric design
Changing a variable qualitatively
–> varying the variable by changing the variable –> nonparametric design
Between subjects design
Each treatment is administered to a different group of subjects
Within subjects design
A single group of subjects ins exposed to all of the treatments, one at a time
–> data is averaged
Single subject design
All subjects are exposed to all treatments
–> data is not averaged but analyzed for single subjects
Error Variance
the variability among scores caused by variables other than your independent variable
–> caused by uncontrolled variables
How to reduce error variance
Holding the extraneous variables as constant as possible
How to randomize error variance across groups
In in between subject designs you can randomize error variance by randomly assigning subjects to treatment conditions
Single factor randomized groups design
- -> Between subjects design
- Subjects are assigned randomly to different groups
Randomized two groups design
You randomly assign subjects to just two groups
Randomized multigroup design
- -> expansion of the randomized two group designe
- you add on or more levels of the independent variable
levels of a variable
different options for a variable
Multiple control group design
- -> variation of the randomized multigroup design
- includes multiple controle groups to rule out alternative explanations of your results
Matched-groups design
Matched sets of subjects are distributed at random (one subject per group) into the groups of the experiment
–>used if you suspect that some subject characteristics correlate strongly with the dependent variable
Matched-Pair design
Matched groups equivalent to the randomized two-group design
-consists of two groups that are not randomly assigned
