As I’ve been working on my dissertation, I’ve had to come up with some new ways to compare (cross-sectional) growth in crappy fossil samples with a larger reference population. I’ve coded a procedure in the R statistical program that uses resampling to test whether two groups differ in the amount of size change experienced between various different ages (i.e. growth). This code is now available on my website.**

And how timely – a commentary in this week’s issue of Nature demands that researchers publish the codes used in their analyses (Ince et al. 2012). After all, what good is Science if it’s not reproducible? (Admittedly, the commentary is geared toward more intense, data-generating programs than anything I’ve written, which is mathematically very simple and generally comprises less than 100 lines of code. Nevertheless.)

Anyone is free to use or adapt the code, with the caveat that one must have at least a little experience using R. In many ways the procedure is similar to a method called Euclidean Distance Matrix Analysis (EDMA; Lele and Richtsmeier 1991), although unlike EDMA my program centers around the problem of making comparisons in the face of lots of missing data. And lots of fun!

**  Oh crap! I just remembered I also posted a simple resampling procedure here on Lawnchair two and a half years ago. Where does the time go…

Some inspiration
Ince, D., Hatton, L., & Graham-Cumming, J. (2012). The case for open computer programs Nature, 482 (7386), 485-488 DOI: 10.1038/nature10836

Lele, S., & Richtsmeier, J. (1991). Euclidean distance matrix analysis: A coordinate-free approach for comparing biological shapes using landmark data American Journal of Physical Anthropology, 86 (3), 415-427 DOI: 10.1002/ajpa.1330860307