Dear Dr Kath, I do not know if I am getting correctly the proposed compositions for each subset. But at least for forensic purposes I think that the groups should be based both on ethnicity and nationality. In Forensic Anthropology what is commonly termed as "ancestry determination" is a key factor to identify the victim. The problem is that this concept has little to do with biology, as it rather deals just with the perception of physical appearance. Thus in my opinion the main problem with this approach would be that this is a very subjective concept. My bet is that both physical characteristics and ethnicity self-identification will vary quite a lot from the American to the Dutch population. And in Italy, where probably sociological factors are very similar to those in my country, most people may even simply deny to place themselves within any ethnic group, so that it will depend almost exclusively on the observer.
Therefore I think that, if forensic anthropologists are taken into account as potential users of the dataset (and I would really like to have access to the data myself!), at least 3 different subsets of a "reasonable" size should be defined for each country. This would multiply the total sample size by 3 (300 individuals per country). For paleoanthropological studies 3 balanced groups of 100 individuals would probably be more than enough, though. And my guess is that, after the preliminary exploratory analyses, you would indeed end working with a single pooled sample for each sex for most comparative purposes. I would really like to know the opinion on the matter of more experienced people in the list. Cheers, Luis _______________________________ Luis Cabo Unidad de Antropologia, Dpto. Biologia Organismos y Sistemas, Universidad de Oviedo. "...as they grow up, either turn thieves for want of work, or leave their dear native country, to fight for the Pretender in Spain, or sell themselves to the Barbadoes." (Jonathan Swift; "A Modest Proposal for preventing the postgraduate students from being a burden on their parents or country, and for making them beneficial to the publick") ----- Original Message ----- From: <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Monday, January 06, 2003 6:33 PM Subject: 3-D human body data > Hello all, > Sender: [EMAIL PROTECTED] > Precedence: bulk > > The Society of Automotive Engineers (SAE) is responsible for > distributing the data from the CAESAR project. They will be charging a > fee for it. Because the data contain more than 13,000 3-D scans the > cost for the entire set is more expensive than a lot of people may want > to pay, particularly small companies or universities. (Someone told me > the entire set from North America costs $20,000, but I am not sure about > that price.) Martha Swiss at SAE asked me if I thought it would be > worthwhile to create a smaller subset of the data...say 100 subjects or > so...and sell that for a more affordable amount. I thought some of you > might have an opinion about that, and you might like an opportunity to > influence the product. Do you think a small subset like this might be > useful for morphometrics? If so, what kind of a slice through the data > should they make? The data were sampled by gender, ethnicity (3 > groups), and age (3 groups). There are 3 3-D scans, 73 3-D landmarks, > 99 tradi! > tional style measurements, and demographic data for each subect. The > data were also sampled by country and include data from the US (well > North America really), The Netherlands, and Italy. Would 100 subjects > randomly selected from each country be useful? That would be 300 scans > (3 per subject) for each country. Or would a sample that has an equal > number of subjects from each ethnic group be more valuable since > presumably it would represent more shape diversity? Any comments or > advice would be welcome. > > Thanx, > Kath > == Replies will be sent to list. For more information see http://life.bio.sunysb.edu/morph/morphmet.html.
