SSS Exam 2- Chap 12-14

Bonferroni Procedure

-adjusts alpha level, or prob of committing a type one error -for within subjects design

larger

A one-way within-subjects ANOVA is typically associated with ______ power than the one-way between-subjects ANOVA.

No. When a variable has only two levels, then those two levels must be significantly different following a significant ANOVA. There are no multiple comparisons to make, so a post hoc test is not necessary.

Are post hoc tests necessary following a significant ANOVA testing one independent variable with two levels (k = 2)? Explain.

Within subjects

Between or within subjects? A behavioral psychologist allows a sample of children to play with four toys of various colors and has them rate how much they like playing with each toy. The psychologist compares mean ratings for each toy.

Within subjects

Between or within subjects? A biopsychologist tests the time course for the release of a neurohormone before, during, and following a task thought to cause its release

between subjects

Between or within subjects? A college professor compares the average class grade for students in each of three sections of a statistics course.

Between subjects

Between or within subjects? A sport psychologist compares mental functioning in a sample of athletes in four different sports.

eta squared

R^2= n^2= SS(BG)/SS(T)

No, post hoc analyses are not appropriate because the ANOVA is not significant.

When you retain the null is it necessary to compute a post hoc test? Explain.

Fisher's LSD test

associated with the greatest power to detect an effect

Testwise alpha

experimentwise alpha ------------------------------------ total # of pairwise comparisons

both eta-squared and omega-squared

measure of proportion of variance for a one-way between-subjects ANOVA

eta-squared

measure of proportion of variance tends to overestimate the size of an effect in a population

k

number of groups (levels)

N

number of total participants in the study

n

numer of participants per group

Omega squared

w^2 corrects for size of error by including MS(e) corrects for number of groups by including df(BG)