Week 2 Hills Chapter 10 Flashcards
What is Tukeys used for?
When there is a significant F, we can use Tukeys HSD Test
(Honestly significant difference)
Which is the most appropriate post hoc test for performing pair wise comparisons among all possible pairs of means.
Tells us which groups are significantly different from which.
Characteristics of an experimental research design…
Manipulation of an IV
Random assignment
Controlled conditions
= an experimental research design.
What is the sum of squares made up of?
The total sum of squares is a measure of the variability in all scores. It is the sum of the squared deviations of each score from the grand mean of all scores (e.g. 40 scores).
It is made up of two parts… See flash card or p. 99 Hills.
The total sum of squares is made up of what two parts?
The within groups or error sum of squares
And the
Between groups sum of squares.
What is the within groups or error sum of squares made up of?
This is a measure of the chance (or error) variability among individuals within each group. It is the sum of squared deviations of each score from its group mean.
P.99 Hills
What is the between groups sum of squares?
This is a measure of the variability. Between groups.
It is the sum of the squared deviations of each score’s Group Mean from the Grand Mean.
P.99 Hills
What is the mean square?
The mean square is actually the variance and is obtained by dividing the sum of squares by the degrees of freedom.
P.101 Hills
What is the sig. value?
The sig. Value is the probability (p) of obtaining the observed F ratio if the null hypothesis is true, and it is always a two-tailed probability.
There cannot be a one-tailed test when there are more than two groups.
P.101 Hills
P…
If p > alpha, the test is not significant, and the null hypothesis of no difference between the population means is retained.
If p</= to alpha! the test IS significant. The null hypothesis is rejected, and the alternative hypothesis that there IS a difference somewhere among the population means is accepted.
P.101 Hills
Significant F
Indicates that there is a difference somewhere among the groups, but it does not specify which groups differ from which.
Tukey’s HSD is a test to conduct pair wise comparisons to determine where differences lie.
P.101
What if the homogeneity of variance assumption is not met in the ANOVA?
A post hoc test for unequal variances should be conducted.
E.g. Games-Howell.
Post hoc tests are ONLY conducted when…
The omnibus F is significant.
P.101
That is the overall F
Partial eta squared n2 is the same as eta squared in one-way ANOVA designs, is …
The effect size measure used by SPSS, although debatable that it’s the best measure…
Calculated as the sum of squares between groups divided by the sum of squares total.
P.103 Hills
