Lecture #13 Flashcards Preview

Stats Final > Lecture #13 > Flashcards

Flashcards in Lecture #13 Deck (10):


Eta-squared: computes effect size for the model, ranges from 0 to 1
Equation; Sum of squares between groups/Sum of squarestotal
Conventions for small, .01
Medium, .06
Large, .14


Brown-Forsythe F:

Brown-Forsythe F: alternative to the Welch test- post hoc test, where variances are not assumed.


Grand mean:

Grand mean: the mean of all means. The average of the group means, used to compute sum of squares.


Grand variance:

Grand variance:





Pairwise comparisons

Comparing two group means. Equation to find number of pairwise comparison k(k-1)/2


When would you conduct a one-way ANOVA? Why would you conduct an ANOVA rather than an independent-samples t test?

When you want to compare multiple pairwise comparisons, with 3 or more groups/levels within the one independent variable.


Be able to determine the number of pairwise comparisons that could be tested based on the number of levels of the independent variable (IV).

3 or more groups within the one independent variable-conduct a one way anova, 2 way anova is for 2 or more independent variables….3 way anova 3 IV’s
to find pairwise comparison use the equation k(k-1)/2


What are the differences between post-hoc tests and planned comparisons?

What are the differences between post-hoc tests and planned comparisons? Post hoc tests: are data driven, after the fact, they are more exploratory, and a posteriori.
The planned comparison- investigates a specific hypothesis, used before we run results. A priori test. Theory driven approach.


Be able to interpret Levene’s test for the equality of (error) variances and corresponding ANOVA SPSS output based on the results of Levene’s test.

P value Greater > than .05 you can assume equal variances, Levene’s test is NOT significant- post hoc test, tukey.
P value less