On Dec 3, 9:45 pm, djh <halitsk...@att.net> wrote: > I have some data which I think you'll find interesting regarding the > regressions of the new average slopes against singleton lengths(these > are the same kind of regression for which I exhibited an S/C split in > a CI plot several posts ago.) > > But I don't want to sling these data at you if you're busy thinking > other thoughts, so I won't post until you tell me it's OK.
OK, now that the sample size error is fixed.
> > Also - quick related question ... can I operate on the special covs as > I've done for the average slopes, i.e. regress them om lengths. Or > are they a different kind of animal?
Comparing covariances of estimates is even trickier than comparing standard errors, but if you have nothing better to do then fine, look at their regressions on length. Plots will probably be more informative than regression coefficients etc.