0
$\begingroup$

I would like to know if there is a way to compare the relationship between a factor (length) and a response variable (lipid content) among multiple groups? I have given up on ANCOVA, as length is continuous. I am unsure if a multiple regression allows the use of a categorical variable (group membership) as an IV.

$\endgroup$
2
$\begingroup$

Multiple regression, ANOVA and ANCOVA are all the same model.

You can include categorical variables in a regression and they can have multiple levels.

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ Forgive me if this forum's users don't use R, but I am trying to produce a scatterplot of the multiple regression results for this data with two IVs (one categorical and one continuous) and one DV...is R even capable of doing that? $\endgroup$ – Brittany Jan 7 '14 at 22:33
2
$\begingroup$

As @PeterFlom noted, multiple regression and AN(C)OVA are specific versions of the general linear model. ANCOVA's specific difference from ANOVA is that it permits modeling continuous covariates like your length factor. ANCOVA's difference from multiple regression is that it suits categorical factors like your group membership variable. ANCOVA would probably work just fine for you, but multiple regression will work too. (After all, they are the same general linear model!) You'll just need to code group membership by one of many schemes suited to multiple regression if you decide to go that route, and figure out how to interpret the results in this somewhat different format.

| cite | improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.