GPA equivalencies: Private, Public, Talent Adjusted

[DunninLA]

<p>This morning I read something about GPA that caught my eye for the first time. The provacative quote came from a March, 2010 paper <a href=“http://www.gradeinflation.com/tcr2010grading.pdf[/url]”>http://www.gradeinflation.com/tcr2010grading.pdf&lt;/a&gt;&lt;/p&gt;

<p>co-authored by Stuart Rojstaczer, creator of the website gradeinflation.com, and retired Duke Professor of Hydrology. Let me summarize what I learned from prior reading at gradeinflation.com:</p>

<p>1) Ave. GPA has increased about 0.1 per decade over the past six decades
2) Ave. GPA at Private schools is on ave. 0.3 higher than Public
3) Ave. GPA for Science/Math majors averages 0.3 lower than Humanities majors</p>

<p>Putting 2) and 3) together, on average, a Public science GPA of 3.0 compares to a Private Humanities GPA of 3.6</p>

<p>However, I have always wondered what would happen if identical twins with identical GPA/SAT performance were split at 18, one attending, say Duke, the other, say UCLA. They major in the same discipline. Would the Duke GPA be 0.3 higher? That question could not be answered because the ave. GPA differences published have thus far not controlled for Talent of student (i.e. GPA/class rank/SAT).</p>

<p>In my first time reading this morning of the March, 2010 paper by Rojstaczer, the following unsupported statement appears at the bottom of page 2: “The above two equations suggest that private schools are grading 0.1 to 0.2 higher on a 4.0 scale for a given talent level of student.” </p>

<p>Also for the first time in the March, 2010 paper, a further breakout of ave. GPA difference by area of concentration is given. They are as follows:</p>

<p>GPA Social Science = GPA Science + 0.1
GPA Engineering = GPA Science + 0.15
GPA Humanities = GPA Science + 0.3</p>

<p>Taking all these ave. GPA difference between disiplines, and the ave. GPA difference between Public and Private, I believe we can construct an equivalency matrix as follows:</p>

<p>GPA equvalencies Public/Private Raw/Private Talent Adjusted </p>

<p>…Talent-
…Raw …Adjusted*
………. Public ……. Private ……. Private
Science ……… 3.0 ……… 3.3 ………. 3.15
Soc. Sci. ……. 3.15 ……. 3.45 ………. 3.3
Engineering ……. 3.2 ……. 3.5 ………. 3.35
Humanities ……. 3.3 ……. 3.6 ………. 3.45</p>

<p>Questions:</p>

<p>What do you make of these historic ave GPA difference between Public/Private, and between Humanities/Engineering/Social Sciences/Sciences?</p>

<p>What practical significance does this have, if true, for a person selecting colleges and majors, in part, on predicted raw GPA necessary for graduate school admission to Medical or Law schools?</p>

<p>Will or should graduate school admissions committees (assuming they have not yet done so) formalize a GPA equivalency table similar to the above in making admissions decisions?</p>

[sefago]

<p>All public schools are not the same for starters.</p>

[DunninLA]

<p>^ If you are suggesting that the ave. GPA data BY SCHOOL published at the bottom of the gradeinflation.com website should be used instead of a the Class of Public v. Private, that makes sense to me. I just wouldn’t want to be a Brown graduate subjected to such a normalization process, but I suppose I would want to be a Swarthmore or Harvey Mudd grad subjected to the same normalization.</p>

[kvilledeac]

<p>I think a more detailed chart comparing all schools’ level of inflation/deflation would be useful to graduate schools. They already do this to some extent but its probably based on vague reputation and not real data.</p>

[xiggi]

<p>

It is perhaps easy to see why graduates from certain private schools dominate placement in top medical schools, law schools, business schools, and why certain private schools are overrepresented in Ph.D. study (E. Bernstein, 2003; J. Burrelli et al., 2008). </p>

<p>They grade easier and there is a tendency for graduate schools, professional schools, and some employers to confer extra stature to those who have attended selective private schools.

</p>

<p>Students get into top medical or law schools because they attend schools that grade “easier?” Is it easy to see … or easy to confuse idle speculation and data analysis. </p>

<p>Have the authors of this “study” considered the importance of the LSAT and MCAT in admissions to what they call top schools?</p>

<p>PS I hope that many know Stuart Rojstaczer for his other claim to fame. :)</p>

[DunninLA]

<p>^ you mean his little jazz band?</p>

<p>Apparently they’re upgrading the March article taking into account feedback from readers. Incidentally, their chart showing delta between sciene/humanities ave. GPA shows a 0.3 delta, whereas the commentary identifies a 0.4 difference. I assume they’ll clean that up.</p>

[xiggi]

<p>“^ you mean his little jazz band?”</p>

<p>Nah, it’s more along the lines of his “positive” or “serious” contributions to the arcane world of college rankings.</p>

[Erin_s_Dad]

<p>I find the shapes of the Deviation from Predicated Grade Curve interesting. Engineering definitely does seem to have a left moded(?) curve but the lower end of the curve is higher than LACs. Public Commuter schools are left moded as well (explained by having to work through school?). Public Satellites have a huge hump at the top end which no other curves have (rewarding those who are definitely a cut above at the Satellite campuses?). The Public Flagship and LACs both seem to be mostly normal (with a minor shift right?).</p>

<p>It’s interesting but how good are we at predicting college success using SATs? And adding selectivity of the school helps how?</p>