Thank you very much! Julien Claude <[email protected]> 于2021年3月24日周三 上午12:29写道：
> Dear Xuziqiang35, > > the problem is that for several groups, you have just one or very few > observations. > It would be non sense to add more variables or to extrapolate them, and, > then to expect that you will find "significant" differences. Some > differences here can not be tested as within group sample variance can not > be estimated.The one P value you observed would be better interpreted as > Non Available pvalues. > Think that the estimates of the mean, even if you have more than one > observation, are also estimate and that they might be influenced by > sampling too !!! > > There has been several recent papers posted here recently (related to > BgPCA) but not only (cf recent paper of J. Rohlf posted at the end of last > year), maybe you should have a look. > As some species are represented by just one specimens, I would recommend > regrouping species and reformulating the problematic according to your > sample (for avoiding to test something that cannot be tested). > > On one other point, it would be good in the future to put an object on > message (it might help other people for archiving message they got from the > list). > > best regards, > > Julien CLAUDE > Institut des Sciences de l'Evolution de Montpellier > > > > ------------------------------ > > *De: *"徐自强" <[email protected]> > *À: *"Morphmet" <[email protected]> > *Envoyé: *Mardi 23 Mars 2021 10:01:24 > *Objet: *[MORPHMET2] > > Hello, teachers. > I am a student who recently learned to apply geometric morphometry to > fossil wing variation studies, but I have run into some challenges I cannot > solve on my own and would like to seek support. > First, I considered using landmark, but the variability is not obvious. > Since many data related to vein sequence curvature are not included, I next > considered introducing semiland for analysis, but some important vein > sequences are missing, making it difficult to obtain semiland points > effectively. > Finally, can the effect of too much semiland on the landmark data be > reduced? > Ag Bc Ca Nm No Sg Sp Ss represent different species respectively. The > dashed part in the picture is repaired according to other fossils, but some > fossils are more or less difficult to repair in the red circle part of the > vestiges. > I would like to get some advice, thank you very much! > > -- > You received this message because you are subscribed to the Google Groups > "Morphmet" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/morphmet2/CADfVnE_LAWBwWbBB%2BVRPciT%3DO1XPEQsr6aFpbaVZ6PdCgdbM0w%40mail.gmail.com > <https://groups.google.com/d/msgid/morphmet2/CADfVnE_LAWBwWbBB%2BVRPciT%3DO1XPEQsr6aFpbaVZ6PdCgdbM0w%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > > -- You received this message because you are subscribed to the Google Groups "Morphmet" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/morphmet2/CADfVnE9TFp-Qm3nPRgp_yxp7Ba3JtBo%2B6Kio-uvOkOtkGrxbZw%40mail.gmail.com.
