Re: Fw: SAT

["GERALD BRACEY" <gbracey1@verizon.net>]

Loos to me like the table shows that the three parts of the SAT are 
incredibly correlated.  You get .48, .47, and .51 from each part alone and 
only .53 for all three taken together.  Whether or not and increase of .08 
in predictive power (.54 vs .62) is worthwhile probably depends on whether 
or not you're making a lot of money from the test.
----- Original Message ----- 
From: <aburke5054@aol.com>
To: <arn-l@interversity.org>
Sent: Tuesday, July 15, 2008 4:58 PM
Subject: Re: [arn-l] Fw: SAT


> I don't know how amusing this is, but the following is Table 5 from the 
> recent CB validity study:
>
> Unadjusted and Adjusted Correlations of Predictors with FYGPA
>
> 1. HSGPA                                            0.36     0.54
> 2. SAT-CR                                            0.29     0.48
> 3. SAT-M                                              0.26     0.47
> 4. SAT-W                                             0.33     0.51
> 5. SAT-M, SAT-CR                                0.32     0.51
> 6. HSGPA, SAT-M, SAT-CR                  0.44      0.61
> 7. SAT-CR, SAT-M, SAT-W                   0.35     0.53
> 8. HSGPA, SAT-CR, SAT-M, SAT-W      0.46    0.62
>
> Clearly, HSGPA and SAT together have more predictive punch than either 
> alone.  Note too that if A and B have equal correlation with C is does not 
> follow that A and B are redundant in predicting C.  The ability of A and B 
> together to predict C is, other things equal, a function of the overlap 
> between A and B.
>
> Art
>
> -----Original Message-----
> From: George K Cunningham <gkc@louisville.edu>
> To: arn-l@interversity.org
> Sent: Tue, 15 Jul 2008 1:25 pm
> Subject: Re: [arn-l] Fw: SAT
>
> Historically, the SAT was a slightly better predictor of first year in 
> college GPA than High School GPA.  In the latest research they appear to 
> be equal predictors, but the two combined is a much better predictor than 
> either by themselves.
>
> George K. Cunningham
> Professor, Ret.
>
> > > > "GERALD BRACEY" <gbracey1@verizon.net> 7/15/2008 1:56 PM >>>
> This letter appears in today's Washington Post.  Monty, Bob and some other 
> ARNers will recognize this as the same argument that engendered a 
> long-running, acrimonious dispute between George Hanford, then President 
> of the College Board, and SAT critic, Jim Crouse (The Case Against the 
> SAT, 1988, with Dale Trusheim).  It will be amusing to see what, if 
> anything, the Board has to say this time around.
>
> Jerry
>
> ----- Original Message -----
> From: GERALD BRACEY
> To: letters@washpost.com
> Sent: Thursday, July 10, 2008 9:04 AM
> Subject: SAT
>
>
> Laurence Bunin of the College Board inadvertently gives the game away in 
> his letter defending the SAT.  He wrote "Recent data compiled by the 
> College Board show that the SAT...is almost precisely as predictive of 
> college success as are four full years of high school."  In other words, 
> the SAT is redundant with grades and useless.  As long as you've got the 
> high school record, you don't need the SAT.
>
> Carl Campbell Brigham, the principal developer of the SAT, saw it a 
> "merely as a supplementary record" and the question has always been, "When 
> the SAT is added to the high school record, does it increase the accuracy 
> of predictions about college success."  It does not.
>
> Gerald W. Bracey
> 1797 Duffield Lane
> Alexandria, VA 22307
> 703-317-1716
>
> The writer is a former Director of Research, Evaluation, and Testing for 
> the Virginia Department of Education and author of Put to the Test: An 
> Educator's and Consumer's Guide to Standardized Testing.