This is about my PhD thesis where I am trying to see the pattern of memory deficit among different types of hypothyroidism. I have used a between subject group design. I have 1 IV with three levels. I have 4 four groups which are:
Subclinical Hypothyroid group N= 14
Overt Hypothyroid group N= 15
Euthyroid group N = 12
Healthy Control Group N = 15
I am using multiple types of memory measures broadly speaking I have administered 1 test of verbal memory with 6 types of scores (DVs) and 2 Tests for working memory with 4 types of scores (DVs). I have been asked to apply One-way MANOVA.
Coming to the issue I am facing is that applying One-way Manova would require meeting several assumptions which my data does not. My data fails to meet the following assumptions:
1.Normality (univariate): Shapiro-wilks test came significant on few dvs.
2.Linearity within DVs: scatterplot showed non linearity among different variables and bivariate correlation also showed non significant correlation between DVs.
3. Violation of homogeneity of variance assumption.
I had applied box-cox transformation using on some variables but that did not helped me solve the problem.
I pretty confused what to do. Whether I should go for multiple ANOVA instead of One single MANOVA. But in that case the alpha would be inflated and would cause type 1 error.
Would appreciate if someone could suggest something on this
Thanks
Vatsal
The violations indicate the need to use non-parametric methods.The Kruskal-Wallace one-way ANOVA would probably work for you. The best book on the subject is Siegel, S. (1956). Non-Parametric Statistics for the Behavioral Sciences. McGraw-Hill. (Don't get the later edition!) I've attached a spreadsheet version of the test.
Thanks for the reply James.
So I need to apply Kruskal Wallis on each dependent variable??
Thanks
Vatsal
Yes. I know of no non-parametric multiple-variate, multiple-group test. One may have been created by someone, but a quick web search and Google Scholar search just now showed nothing of interest. I prefer to use the well-validated tests described by Siegel in 1956.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India