[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
Re: playing with numbers
- To: <arn-l@interversity.org>
- Subject: Re: playing with numbers
- From: "Monty Neill" <monty@fairtest.org>
- Date: Mon, 29 Sep 2003 17:32:44 -0400
- References: <20030929102042.647CA3F35@inter.interversity.net>
Note that this article failed to provide adequate information on why states
are using the confidence intervals. Even as the congressional leadership was
figuring out AYP, researchers Kane and Staiger reported that looking at
annual test scores changes was more noise than signal for groups of fewer
than about 65; Walt Haney suggests groups of 100 or more are needed.
The feds would not allow OR to use a group size of 70 - I think 50 is the
largest any state has. Meaning, in small schools, esp. in next few years in
states not beating (excuse me, testing) the kids in 3-8 inclusive, the
accuracy of data is real shaky.
None of this would you know from this article.
Post a Message to arn-l: