Peter Timusk: I am hoping that you can assist me with a minor, alright major statistics situation. When applying discriminant analysis would the ability to predict gender accurately at the rate of 68.2% be considered an effective or accurate measure with a processed sample population of 16,909 (8391 male cases and 8518 female cases -it has been weighted). My thoughts are that when one has a large sample like in a longitudinal study of students that 68% seems like a high error rate when one takes into account the 95% or the 90% accuracy rate commonly used in statistics or even the common plus or minus of 4% utilized in media polling. Am I correct in my assessment? If so can you explain why discriminant analysis would be good here. Do you think there is another method that may be a better fit here? Or am I incorrect in my assessment? Show me the way in this great debate. I guess in theory there is no "wrong answers" only better methods or processes for a given situation. I am not a statistician, so I am going to a person with a real background for help. If Peter can't help can someone on the board provide an assessment here. Chris Heidelberg -----Original Message----- From: air-l-bounces@listserv.aoir.org [mailto:air-l-bounces@listserv.aoir.org] On Behalf Of Peter Timusk Sent: Thursday, June 08, 2006 5:55 AM To: air-l@listserv.aoir.org Subject: Re: [Air-l] subcultures and the internet reply below but I think I misunderstand what you are asking for. I am more commenting on Western subcultures around the world. But my knowledge of Muslim culture is restricted to recently finding out about how quotes of Mohamed are attributed by a type of voting system I really know nothing about various non western subcultures. On 8-Jun-06, at 4:37 AM, jespert wrote:
Dear list members,
<snip>
. I am interested in parallel cultures, especially in Muslim parallel cultures in Denmark or in other western countries growing via the internet, but other examples on how subcultures develops and maintain them selves on the internet also have my strong interest.
If you have references for me or other input, on or off list I'll be grateful
Best Regards Jesper
I had been pointed to a punk blog from Iran before but can't find it right now This seems to be a German media report about Punks in Indonesia. Sorry lacking anything about the net. http://www.qantara.de/webcom/show_article.php/_c-310/_nr-203/i.html Punk must mean quite different things in your own culture than mine. I have read some of the references posted today to this email request. I would recommend them too. I am interested in the fact of punk rock coming along at the same time parallel to the net time vis via cyber punk fiction. You might check out feminism and music studies as I also think this sub field of music also came along about the 1990's at least in punk history but therefore net history. Peter Peter Timusk, B.Math statistics, B.A. legal studies M.A. legal studies applicant just trying to stay linear. Read by hundreds of lurkers every week. _______________________________________________ The air-l@listserv.aoir.org mailing list is provided by the Association of Internet Researchers http://aoir.org Subscribe, change options or unsubscribe at: http://listserv.aoir.org/listinfo.cgi/air-l-aoir.org Join the Association of Internet Researchers: http://www.aoir.org/