SAT: Part hoax and part fraud?

<p>

Note that the 0.82 correlation mentioned above occurred between SAT scores of students who took the test in the 1970s and their scores on the Armed Service Vocational Aptitude Battery. I’m not not familiar with the history of how the armed services test has changed since the 70s, but the current version sounds similar to the SAT with sections on Mathematics Knowledge, Arithmetic Reasoning, Word Knowledge , and Paragraph Comprehension. However, it also has some sections that seem specific to various sub-groups within the armed services, such as a section on Electronics and a section on Automotive and Shop Information. The same study mentions that the Armed Service Vocational Aptitude Battery had a 0.56 correlation with the Lorge-Thorndike IQ test. While this is a notable correlation, it is different enough to assume the armed services test is not the same as a standard IQ test. Consistent with this, the study found a 0.48 correlation between SAT scores and scores on Raven’s Advanced Progressive Matrices, which attempt to measure a component of g… a notable correlation, but far weaker than 0.82. Also note that the SAT has undergone some significant changes since the 1970s data used in the study, changes that move it further away from its origins of being based on the Army Alpha test (an Army intelligence-like test). The author of the referenced study addresses this issue by saying:</p>

<p>Unfortunately, the usefulness of Equation 2 may not last long. Currently, the SAT is being revised to shift the focus from general reasoning ability to academic achievement. The objective is to test content knowledge rather than intelligence (Barnes, 2002).</p>

<p>The Frey and Detterman study was a good find, Data10.</p>

<p>Here is another reference:</p>

<p>"1) Treena Eileen Rohde and Lee Anne Thompson: Predicting academic achievement with cognitive ability</p>

<p>This study is likely the weakest only because they used a group of college students from an elite university. Not that there is anything wrong with this, but when you see the samples in the studies below, it is a noticeable concern.</p>

<p>Their major contribution was that in predicting (standardized) academic achievement, speed of information processing and spatial ability can explain small, but significant, amounts of variance unexplained by general vocabulary (Mill Hill) and perceptual organization (Raven’s Matrices), although the latter two tests, hands down, did the best in predicting academic achievement across various indicators.</p>

<p>In their own words:</p>

<p>General cognitive ability measures (Raven’s, Vocabulary) and specific cognitive abilities (working memory, processing speed, spatial ability) collectively accounted for between 16% [GPA] and 54% [SAT] of the variance in academic achievement."</p>

<p>end quote
54% of the variance shared between “cognitive ability” measure and SAT implies a multiple R of about +.74
The study was criticized because of sample selection from am elite school. I think the study was from 2007.</p>

<p><a href=“http://www.gnxp.com/blog/2007/01/iq-academic-achievement.php”>http://www.gnxp.com/blog/2007/01/iq-academic-achievement.php&lt;/a&gt;&lt;/p&gt;

<p>So, if you were a college looking to select eager young students, would you prefer the hard-worker who has acquired content knowledge via his educational experience, or the very intelligent one who doesn’t have to work hard to learn? Seems to me that the combination of IQ-test-ish SAT plus grades lets you measure (to some extent) both qualities, whereas an SAT that is based on curriculum mastery takes away one attribute. You may end up selecting just the “good” students, which is fine if that’s your objective. What if you’d like to find the potentially brilliant scholar of classics who just didn’t care about biology and math? </p>

<p>

Their own words also state that, “The largest correlation was between SAT verbal and Vocabulary (r=0.709),” and “The correlation between GPA and Vocabulary was not significant.” So one would expect that estimates of cognitive ability focusing on vocabulary will find an especially large difference between the how much of variance in GPA can be explained and how much variance in SAT score can be explained. That said, I don’t doubt that cognitive ability tests correlate better with SAT than GPA…</p>