Dear all- I have been trying to use the tpsRegr to look at differences in morphology of juvenile sockeye salmon based on the type of rearing habitat (river or lake) that they come from. Initially, I wanted to look only at changes in shape due to the habitat that they come from, but I soon found that my results were being confounded by changes in shape due to size as well. Because of the field sampling techniques, the fish that were sampled from different sites (3 sites for each rearing habitat type) were not all the same size (some sites had overall larger or smaller juveniles, whereas some had a broad range of sizes).
I used the test for common regression slopes that is highlighted by tpsRegr, in order to see if the relationship between size and shape was the same for all sites sampled, hoping that I could then continue on to test for shape differences between the 2 different rearing habitats with size held constant. The tests for common slopes however showed that the slopes of the relationship between size and shape were (highly) significantly different between the different sites and the different rearing habitats. I am hoping for some advice as to what to do next. I think that the significant difference in slopes may actually be due in large part to the large sample sizes (between 100-300) for each site tested. In fact, I went back and looked at each site separately (randomly dividing each site into two groups), and tested for a similar slopes between fish from each site. These results also showed to have significantly different slopes.... even when they were from the same site from the same habitat type!!! Has anyone encountered this type of problem before, and are there any tips as to what to do next? I am a beginner with this stuff, so any advice would be much appreciated. My email is [EMAIL PROTECTED] Thanks very much, Renae Mackas -- Replies will be sent to the list. For more information visit http://www.morphometrics.org
