ANOVA & ANCOVA assumptions For Idiots

Question 0

ASSUMPTIONS FOR ANCOVA.

P.173 Hill

Answer 0

1. The covariate(s) needs to be at interval or ratio level of measurement. Dichotomous sometimes.
2. For reliability of results, the correlation between the COVARIATES and the DV should be significant.
3. Multiple COVARIATES need to be chosen judiciously, based on sound theoretical understanding of their relevance… Not too highly correlated with one another.
4. COVARIATES normally distributed each group.
5. Linear relationship between COVARIATES and DV (each group).
6. COVARIATES should be reliable.
7. Homogeneity of regression…relationships between the COVARIATES and the DV must be the same across all groups.

Question 1

ASSUMPTIONS FOR ONE-WAY ANOVA

Hill p.92

Answer 1

1. DV should be interval or ratio level measurement.
2. Scores randomly sampled from the population of interest.
3. Scores are independent…participants appear in one and only one group; the groups are not related to one another; and the score for each participant is independent of the scores for all the other participants.
4. Sores on DV normally distributed in each population.
5. Homogeneity of variance…variances in groups / popns. equal.