Types of experimental design Flashcards
Between-subjects design
certain groups exposed to certain conditions
Within-subjects design
all groups exposed to all conditions
Error variance
variability among scores caused by extraneous variables (individual differences, environment, …), low internal validity
Experimental design
manipulation of independent variable and observation of changes in the dependent variable
Simplified categories of experimental design
between-subjects, within-subjects, single-subjects
How to reduce error variance
- laboratory settings and strict control (reduces external validity)
- match subjects across groups based on characteristics that are believed to contribute
- same subjects
- refer to previous research for correct levels of variable
- randomizing
Types of between-subjects design
- single factor or multiple factor
- randomized groups design
- multiple control groups design
- matched-groups design
- matched-pairs design
- matched-multigroup design
Randomized groups design
subjects assigned randomly
Multiple control group design
different control groups for different alternative explanations (e.g. one with placebo and one with no treatment at all)
Matched-groups design
matched sets of subjects distributed randomly, one subject per group, into all of the groups
Matched-pairs design
only two groups needed
Matched-multigroup design
multiple levels of single-factor or multiple-factors
Types of within-subjects design
- single-facor or multifactor
- multilevel design
- two-level design
Multilevel design
each group all treatments of multiple levels of variable
Two-level design
each group all treatments of only two levels of variable
Mixed design
combination of between-subjects design and within-subjects design, A&B and C&D levels of two variables (treatment A or treatment B and then treatment C or treatment D for each of the established groups)
Single-factor
One variable
Multifactor
multiple variables
Two-level
two possible values for variable
Multilevel
multiple possible values for variable
