Subscribe Bookmark RSS Feed

Comparing regression lines of different data

Hi, I would like to compare the regressions of two different sets of data.

One set is the number of Sea urchin spine breaks vs the treatment

The other set is Level logger reading vs Site

If you need context: I have field samples of urchins, I quantified spine breaks and noted their collection location as treatment. Alongside this study, we placed sensors that log waves, and i have them organized by site. Id like to see if there is any correlation between these two regressions.

Thanks in advance.
I have edited the data so that site (nominal, S) can be used for both spine breaks (continuous, B) and logger data (continuous, L).

So now my question is, can I compare slopes of these comparisons (S v B and S v L) if they don't have the same number of data points?

For example, for site 1, there may be 75 'B' points, but 89 L points.

Also, I've heard something about using ANCOVA for this, but I also heard something about dummy columns..I'm not sure how to run this (I have 10 sites)

Any help would be greatly appreciated.

Super User


Jun 23, 2011

I am not sure what you mean by "compare slopes" here. Site is nominal so S v B or S v L alone would imply comparing means by e.g. ANOVA (assuming S is independent and B or L response variables).

If you have different number of rows for B and L, stored in separate columns, and include them in the same model I think JMP wold not include rows where on or more numbers are missing in the analysis.

Can you provide more details about what you want to do?