chapter 10: between-subjects experimental design Flashcards
between-subjects design
Different participants are randomly assigned and then observed one time in each group or at each level of a factor
random assignment and control
Random assignment of participants to different groups ensures that each participant has the same likelihood of being selected to a given group
Can be confident that any differences observed BT groups can be attributed to the different levels of the IV and not individual differences
two independent samples
two-independent sample t test
multiple independent samples
one-way between-subjects analysis of variance (ANOVA)
independent sample
participants are independently obbserved one time in each group/level
quasi-experimental method of collecting 2 independent samples
population 1 –> sample 1
population 2 –> sample 2
ex// students who pay for their college education vs students who have help
experimental method of collecting 2 independent samples
population –> sample –> group 1 and group 2
ex// paying one group of students for higher grades vs control
two independent sample t test (independent sample t test)
1 IV with 2 levels
2 different samples
tests the overlap in scores between samples to see if manipulation (rather than individual differences) is causing differences between groups *** different levels cause change, not the different samples
quasi-experimental method of collecting 2+ independent samples
population 1 –> sample 1
population 2 –> sample 2
population 3 –> sample 3
ex// observe patrons at a gym who complete an easy, moderate, and intense workout + their food cravings
experimental method of collecting 2+ independent samples
population –> sample –> group 1, group 2, group 3
ex// select random sample of patrons at a gym and assign to complete an easy, moderate, and intense workout + their food cravings
one-way between-subjects analysis of variance (ANOVA)
1 IV with 2+ levels
2+ samples
is this significant?
tests the overlap in scores (rather than individual differences) is causing differences between groups
does not identify which paires of groups differ
ex// IV - temperature (50 degrees, 70 degrees, 90 degrees), DV - performance on cognitive ability test
post hoc test
which comparison(s) matter?
computed following a significant ANOVA to determine which pair(s) of group means are signficiantly different
pairwise comparison - what is the mean difference magnitude?
statistical compairson for the difference betweeng two group means
ex// signficant difference bt 70 degrees and test scores
assumptions of two independent sample t test
interval/ratio scale
normality (normally distributed)
random sampling
independence
equal variances
assumptions of one-way between-subjects ANOVA test
interval/ratio
normality
random sampling
independence
homogeneity of variance (variance in each group = another)
advantage/disadvantage of between-subjects design
only design that can meet all 3 reqs of an experiment (randomization, manipulation, and comparison/control group), places less of a burden on participant and researcher
BUT sample size required can be large
