Saber, You may use any stats program (or do the calculations yourself in Excel) once you get shape variables from the tps programs. Suitable shape variables for standard landmark-based shape analyses are: 1. Partial Warps 2. Principle components of partial warps OR PC's of superimposed (partial procrustes) coordinates
Beware that tpsRegr and tpsRelw will yield different partial warp scores due to different default parameters in the two programs. I recommend the approach of just superimposing coordinates, calculating principle components (on covariances), and dropping zero magnitude principle components. E.g. for 2D data you will get 2k-4 shape variables (k=number of landmarks). The resulting shape variables can be entered into MANCOVA to test for group differences, using centroid size as a covariate. MANCOVA is preferable to discriminant function analysis because DFA can not handle continuous variables such as centroid size. -Thom ________________><()()(ยบ>________________ Dr. Thomas J. DeWitt, Associate Professor Departments of Wildlife & Fisheries Sciences and Plant Pathology & Microbiology Interdisciplinary Program in Ecology & Evolutionary Biology Texas A&M University 2258 TAMU College Station, TX 77843-2258 Tel. (979) 458-1684 (office) Tel. (979) 845-7522 (lab) Fax (979) 845-4096 E-mail [EMAIL PROTECTED] Web http://wfsc.tamu.edu/faculty/tdewitt/webpage.htm TAMU Map to DeWitt lab & office: http://www.tamu.edu/map/gifs/detail/FGHB.gif >>> morphmet <[EMAIL PROTECTED]> 04/04/07 9:46 AM >>> Dear Morphometricians I am a PhD student and amateur in geometrics morphometrics. I really like to works on this subject and use it for discriminating my samples. I'm using TPS series programs at present for preparing coordinated data (landmarks digitized and ....) but I can't use NTsys program for statistical analysis, so I would appreciate if you could give me some information about another compatible ( and of coure easy to use) statistical software and also send your opinion about PAST. I saw some different analyses in articles in case of discrimination analysis of populations and I'm a little confuse about that. what is your purpose? thank you very much in advance. Best Wishes saber --------------------------------- 8:00? 8:25? 8:40? Find a flick in no time with theYahoo! Search movie showtime shortcut. -- Replies will be sent to the list. For more information visit http://www.morphometrics.org -- Replies will be sent to the list. For more information visit http://www.morphometrics.org
