Peter Coles (1) claims that the statistical analyses in our study (2) are “strongly undermined by an inappropriate choice of variables.” This rather harsh statement is surprising and inappropriate especially because Coles seems to generalize his criticism although two completely different statistical methods were used for the data analysis; the connectivity analysis was carried out by using the statistical parametric mapping, SPM2 (well validated in numerous brain imaging studies), and hemispheric asymmetries were tested with ANOVA and t tests, as described in the article (2). It is true that the construction of an asymmetry index has its difficulties, and there are numerous publications since Geary's paper in 1930 (3) suggesting different approaches to tackle the problem. A long-tailed distribution is the major focus of Cole's comment (1). Our index actually showed a normal distribution (Fig. 1). This outcome is in line with the well known central limit theorem, which states that when the sample size n is big the sampling distribution of the mean becomes approximately normal regardless of the distribution of the original variable (it is a rule of thumb that this is true if df > 30).
Fig. 1.
Frequency distribution of calculated cerebral asymmetries, (right hemisphere volume − left hemisphere volume)/(right hemisphere volume + left hemisphere volume/2), including all four study groups. One outlier is not included in the figure, but recalculation of the data excluding this outlier does not change the results.
Nevertheless, in response to Cole's letter we have retested the asymmetry data with nonparametric statistics (Wilcoxon signed-ranks test for within-groups analysis and the Mann–Whitney U test and Kruskal–Wallis test for between-groups analysis).
The results stand significant.
Footnotes
The authors declare no conflict of interest.
References
- 1.Coles P. Statistical errors and asymmetry indices. Proc Natl Acad Sci USA. 2008;105:E82. doi: 10.1073/pnas.0806646105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Savic I, Lindström P. PET and MRI show differences in cerebral asymmetry and functional connectivity between homo- and heterosexual subjects. Proc Natl Acad Sci USA. 2008;105:9403–9408. doi: 10.1073/pnas.0801566105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Geary RC. The frequency distribution of the quotient of two normal variates. J R Stat Soc. 1930;93:442–446. [Google Scholar]