If you have two quantative factors (with interaction) with different level of each, with a given number of replication, what is the best way to analyze the data?
It is obvious that we can use a factorial ANOVA follow by multiple comparison of means, but it is not a proper way to handel such data.
ANOVA is not proper to analyse quantitative data and quantitative factors ! It's used to analyse quantitative data if you applied qualitative factor such as temperature with different levels ! Are you sure concerning you experiment that you have quantitative factors ?
Dear Sohrabi,
I would ask you to clarify: On one hand you talk about "levels", implying that your quantitative Variable has only a few discrete values, which would render it practically nominal. In this case, it would be appropriate to run an ANOVA (if all conditions are met). On the other hand, you say both variables are quantitative (interval scale?), and then I agree it would be a waste of data to run an ANOVA (albeit not forbidden).
The crucial question you have not said anything about yet is: What kind of dependent variable do you have? Depending on that, a kind of regression analysis would probably most appropriate.
Dear Aenne
The factors are different level of nanosilicone (say 0.0, 0.1, 0.2, 0.3, 0.4) and plastic (0, 0.5 and 1). 6 quantative response variable were measured (drainage, elasticity, ...). Also, We want need to explore interaction effects as well as determining the best combination.
Dear Hormoz
When interaction effect was significant in Factorial design, there is a Proc in SAS called Slice. You can find it in help manual of SAS.
Dear Hormoz,
as far as I can judge from your description, you could go for a linear regression. You However, one more thing you have to consider if you aim at investigating the effects on 6 outcome variables is whether or not these outcomes are dependent on each other. If so, a multivariate ANOVA is definitely NOT appropriate.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA