Reader Comments
Post a new comment on this article
Post Your Discussion Comment
Please follow our guidelines for comments and review our competing interests policy. Comments that do not conform to our guidelines will be promptly removed and the user account disabled. The following must be avoided:
- Remarks that could be interpreted as allegations of misconduct
- Unsupported assertions or statements
- Inflammatory or insulting language
Thank You!
Thank you for taking the time to flag this posting; we review flagged postings on a regular basis.
closeAbout material
Posted by tjlamsa on 21 Dec 2006 at 12:30 GMT
First of all thanks for the very intersting study.
You state in the methods section that you have used two different data source and study subjects to obtain brain size and testes size. Doesn't this act as a major bias in calculating the correlations?
There is only one mate per species. The statistical power would be enhanced by having more subjects and fewer different species to clarify the statistical testing of hypothesis.
RE: About material
schillaci replied to tjlamsa on 27 Jan 2007 at 16:48 GMT
Thank you very much for your comment. I agree with you regarding the limitations to my research associated with the sample sizes I used. Estimates of taxon-specific brain and testes sizes would certainly benefit from larger sample sizes. Similarly, statistical power would also benefit from the inclusion of additional species. In my future research I hope to remedy these analytical limitations to my study of primate brain evolution.
I am not sure, however, that I agree with your comment on statistical bias. Using different data sources for estimating the linear relationship between brain and testes sizes will introduce error and uncertainty to my estimates of correlation, but there is no reason to assume that using different data sources would bias systematically the direction or strength of those estimates.
Concerning statistics
caio_maximino replied to schillaci on 28 Mar 2008 at 21:03 GMT
This will be redundant to annotations, but I would like to emphasize the point that the statistical analysis should be clarified. It is not clear whether ANCOVAs were corrected for the phylogenetical relationships among the species studied, and p-values for the correlations were not presented. Since there was a considerable phylogenetic signal in the traits analysed (even though the method used to assess that is not the best available), one cannot consider them to be independent data points, and Bonferroni corrections should be applied to p-values of correlations.
Aside from that, the article is very insightful, and the hypotheses tested are intriguing. I also would like to add that the alterations proposed in the statistical analyses (if necessary) will probably give the same conclusions that the author reached; it is only a matter of correctedness.