Fred, Glad you explained it. I always thought that the number from Rymut was the current number and if smaller many have left Poland of died off. I never thought that adding your data to Rymuts was the correct population of a particular surname. This sure makes a difference in my thinking. I am sure glad you posted this. As far a Rymut on my surname it is inaccurate. I know more people with my last name then he accounts for. Kuba On Mar 16, 2010, at 11:50 AM, Fred Hoffman wrote: > Hi, > > Kuba wrote: > >> There is a name in Polish Surnames....... by Fred Hoffman. >> Siemiątkowski. There were 1994 of them in Poland in 1990. >> >> However >> I searched the site >> http://www.moikrewni.pl/mapa/kompletny/siemi%25C4%2585tkowski.html >> >> and got the following information. >> >> W Polsce są 982 osoby o nazwisku Siemiątkowski. >> Zamieszkują oni w 137 różnych powiatach i miastach. Najwięcej >> zameldowanych jest wWarszawa ,a dokładnie 107. >> Dalsze powiaty/miasta ze szczególnie dużą liczbą osób o tym nazwisku >> Sierpc (37), m. Płock (37),Lipno (32), m. Toruń (30), zachodni Warszawa >> (28), Grodzisk Mazowiecki (24), Brodnica(22), Rypin (21) i Płońsk z liczbą >> wpisów 20. >> >> It shows 982 people with that name in Poland. >> then is shows the distribution by town. i.e. 137 in Warszawa, 37 Sierpc >> and so on. > > Good point! The discrepancy in numbers is easily explained, however. The > data I used in my book was from Prof. Rymut's compilation of 1990 PESEL > data, the same data you can search at http://www.herby.com.pl/indexslo.html. > That data did not separate masculine and feminine forms of surnames, so that > all Siemiątkowskis and Siemiątkowskas were grouped together under the form > Siemiątkowski. > > The data used by Moikrewni site is from Rymut's 2002 compilation of PESEL > data, which presents masculine and feminine forms separately. So to get the > full picture for the name Siemiątkowski, you also have to add in those > listed on the page for Siemiątkowska: > > http://www.moikrewni.pl/mapa/kompletny/siemi%25C4%2585tkowska.html > > Add the 982 Siemiątkowskis and the 1,051 Siemiątkowskas and you get 2,033 > people registered under this name's masculine and feminine forms as of 2002, > or 39 more than the 1990 database showed. By and large, the 2002 number is > somewhat larger than the 1990 number, though there are cases where it's > smaller, probably because misspellings have been corrected. The point is, > comparing 1990 data to 2002 data is sometimes a matter of comparing apples > and oranges (to use a cliche that probably needs to be given a decent > burial). > > If you're interested in why they took this approach, let me explain. The > 2002 data is more accurate than the 1990 data was, and is also > comprehensive -- it covered essentially 100% of the Polish population, > whereas the 1990 data covered only about 93% of all Polish citizens. Rymut > realized that linguists -- perhaps not used to computers and their tiresome > way of doing exactly what you tell them to do -- made some mistakes > compiling the 1990 data in regard to masculine and feminine forms. Some name > forms ending in -ska are not actually feminine forms -- for instance, > LASKA -- so it was not appropriate to lump them in together with apparently > corresponding -ski forms. The 1990 data showed no LASKAs, but 3,354 LASKIs; > the more accurately compiled 2002 data shows 65 LASKIs vs. 3,534 LASKAS. As > you see, the 1990 data can be very misleading. I think that's because Rymut > and his collaborators were mostly linguists, not computer experts. They > didn't realize computers are only as smart as their programming, and can > make mistakes a human would instantly recognize and correct. I hate to think > how horrified Rymut must have been when someone called and said, "Yo, > Kazimierz, take a look at your 1990 LASKI/LASKA data. Something ain't right > here!" > > Also, adjectival masculine/feminine forms of some adjectival surnames, such > as BURY vs. BURA, had not been recognized as such; so the 1990 data gives > them separately, which is inconsistent. In addition, Rymut recognized that > there is a small but significant number of Polish women who choose not to go > by the distinctive feminine forms; for instance, of the 68,743 KOWALSKIs > registered in 2002, 61 were females who, for whatever reason, preferred not > to go by KOWALSKA. (With some surnames, the percentage of females using the > masculine form is much higher.) > > My point -- and believe it or not, I do have one -- is that Rymut knew > scholars would probably use this data for statistical analysis, and he > wanted it to make it as accurate as possible. So he left the data for > masculine and feminine forms separate. And the Moikrewni site, which uses > that data, also presents them on separate pages. That's why, to get the full > picture for the name Siemiątkowski, you also have to add in those listed on > the page for Siemiątkowska: > > I know I took the long way around to explain this, but I thought some of you > might like to know the details. The 2002 data is valuable for many purposes, > but to use it well, you have to know the ins and outs of the compilation and > formatting. The 1990 data on the Herby site lumps masculine and feminine > forms together, sometimes incorrectly. The 2002 data on the Moikrewni site > separates them. It's kind of a pain in the rear, but there is some > justification for that approach. > > Fred Hoffman > > > ------------------------------- > To unsubscribe from the list, please send an email to POLANDBORDERSURNAMES-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message Kuba Przedzienkowski