Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
10 Cards in this Set
- Front
- Back
|
SStotal equals
|
SSbetween and SStotal
|
|
|
effect size tells us
two examples of effect size in OW ANOVA |
whether something is meaningful or dramatic
heta and w-hat-squared |
|
|
Effect sizes labeled
|
Small: .01
Medium: .06 Large: .14 and beyond |
|
|
heta squared tends to
|
overestimate effect size...why we do the w-hat thing as well
heta square is biased |
|
|
Only do post-hoc comparison
|
if reject Ho in omnibus original ANOVA test
|
|
|
Tukey test is for all
|
pairwise comparisons
|
|
|
q is
|
studentized (a person) range statistic in the HSD (conservative honestly significant difference)
|
|
|
n in the HSD equation is
|
number of scores PER GROUP
|
|
|
Planned comparisons
|
some researchers say should only test what interesting to you...test inherent but not omnibus test...may be pairwise or could even be more complex
|
|
|
3 constraints of planned comparisons
|
1. each hypothesis is defined a priori (before hand..kind of like choosing alpha)
2. you are able to test k-1 comparisons (all questions can be answered by k-1 or fewer comparisons) 3. The comparisons are orthogonal to each other (orthogonal means independent and that they do not overlap)...this also means that you assign coefficients to each mean |
