lecture 8 - Factorial design

simple experiments

one iv with two levels

reasons to add levels

test boundary conditions(where an effect stops), identify curvilinear relationships (low/med/high) test multiple treatments (control, CBT, DBT) 

costs

lower stastical power (more type II errors), requires larger sample size and more resources

adding variables

experiments with two or more IVs, can include participant variables (measured like age and gender)

why add variables

test boundary conditions (generlaizialbilty, moderators) test theories (whether an effect changes under specific IV combinations)

milgram 

obedience difers by lab coat and culture 

costs

lower power - more type two errors, larger sample, more complex more resources

basic design

independent groups (between subjects), within groups (repeated measures) matched groups

factorial design structures

2×2 between groups ANOVA, 2×2 within groups ANOVA, 2×2 mixed ANOVA (one IV between, one within)

between groups 2×2

both IVs are between groups, IV1 - movie type (happy/sad) IV2 = social context (alone/group)

within groups 2×2

all particpants’s experince all 4 conditons

mixed 2×2 

one iv is within groups the other between groups 

main effect 

the overall effect of one IV averaging, across levels of the other IV 

interaction 

the effect of one IV depends on the level of another IV, if simple effect differ - interaction, if simple effect match - no interaction

more complex factorial designs 

affng more variables 2×2×2 = 8 conditions, adding more levels 2×2×4 = 16 conditions