Date: Mon, 11 Aug 1997 15:50:14 -0700 (PDT) From: Annelise Anderson <andrsn@andrsn.stanford.edu> To: Jamie Bowden <jamie@itribe.net> Cc: Scott Blachowicz <scott@statsci.com>, ac199@hwcn.org, "Jonathan M. Bresler" <jmb@freebsd.org>, hoek@hwcn.org, softweyr@xmission.com, chat@freebsd.org Subject: Re: FTC regulating use of registrations Message-ID: <Pine.BSF.3.96.970811151816.4558A-100000@andrsn.stanford.edu> In-Reply-To: <199708111413.KAA16436@gatekeeper.itribe.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 11 Aug 1997, Jamie Bowden wrote: > On Sun, 10 Aug 1997, Annelise Anderson wrote: > > > There's also no evidence that what students learn correlates with > > the amount spent per student. > > Please share whatever it is you're smoking. School systems in most states > are paid for by local property taxes. In wealthy suburbs property values [...] > succeed, it's a large one. Your politicians are the one's claiming that > the school's available funds don't make a difference. You go to a school > where the textbooks are 2 years out of date, and in short supply, a > library that is mostly non-existant, understaffed, and has no purchasing > power because textbooks are a priority, a building that's over 50 years > old, in need of repair, with no available funds for that either, and > underpaid teachers (who tend to be the worst the educational system has to > offer, since the better teachers in general go to schools where they don't > have the limitations, and safety concerns these schools present, and make > more), and we'll see how well you do. Don't ever claim money makes no > difference. You're either lying or naive. In the world we live in, money > always makes a difference. > > Jamie Bowden > > System Administrator, iTRiBE.net Actually the absence of correlation between inputs and results in education is rather well established. This basic research, quite extensive, was done and published by James Coleman (in a book of which I can't recall the title) some 30 years ago. During the '70s attempts were made to beat this finding down, with little success. Some studies in the '80s tried to chip away at it again with new data. In addition to Coleman there were studies that found no correlation between student achievement and teacher/student ratio, books in the library, etc. etc. What student performance does correlate with are some variables relating to the socio-economic status of the parents. I read one report some time ago that noted that a good proxy for all the socio- economic variables was the education of the mother. Now, keep in mind that a correlation doesn't identify a causative effect. All of this is, necessarily, "social science" research; it's always a bit slippery. But it's pretty clear that spending more money isn't in itself a solution. Note also that the findings of this kind of research apply to the range of values of the variables involved in the study, with, generally, no implications for values outside those ranges. No one is saying anything about what effect $20,000 per student or $500 per student might have. Also, there are a lot of debates about the measurement of student performance or achievement, and problems in comparing results with one test with those of another. There is considerable debate about whether the (rather small number of) students who have taken advantage of school voucher programs in Milwaukee, Cleveland, and other places have done better than those who have not. Here school choice is being evaluated by comparing rather short-run student performance, and there are all sorts of possibilities for selection bias and so forth; and the question of whether the public schools have been improved by this experiment is not even asked. Recently (spring 97) in New York City, the mayor got Wall Street funding and (I think) Catholic Archdiocese support for a voucher program for 1300 students; to qualify students had to be eligible for the federal subsidized school lunch program (one measure of need) and be low achievers (or perhaps going to low-achieving schools). In any case there were 16,000 applications for this program. The students were or will be selected randomly from the 16,000 applicants (I think they can attend any private school they can get into, not only Catholic parochial schools). The funding is available for a study of the results. This sounds like a good research design.....but again, doesn't test the more basic questions of whether the competition created by a voucher program will improve the public schools. Annelise
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.970811151816.4558A-100000>