JMU, SOL’s and Grade Inflation
DebSF -- June 17th, 2009
The June 15th edition of Inside Higher Ed has an interesting story by Doug Lederman on grade inflation in Virginia high schools, highlighting JMU’s experience to draw a connection between SOL’s and grade inflation for college-bound HS students who take the test.
The article is based on a working paper by GMU’s Patrick Marquardt. The author notes that after the implementation of the SOL’s in 1998, Virginia required all students to pass a certain number of SOL exams to graduate. This paper examines the incentives that individual school districts put in place in order to encourage students to take the SOL’s more seriously. Examples include giving students who passed an exam a grade bump (from B to B+, say), to allowing students to use the SOL in a particular subject as their final exam, earning an A as their course grade if they passed it.
The author’s hypothesis is that these incentives drive up the GPA’s of college-bound HS students. He compared academic data on VA’s public colleges with that of 38 similar public colleges in 28 states that do not have SOL-type examinations.
Marquardt finds that the GPA’s of VA students showed a significantly higher increases than the other colleges; from 1995-2007, for example, the HS GPA of freshman enrolled in VA colleges rose by an increase of 9.9%, or .79% per year. In the alternative national sample, the increase was 6.4%, or .5% per year. Aggregate VA data also show serious jumps in 2000 and 2004, the two points during which the state sharpened the stakes for school systems, based on their SOL performance.
Are there alternative explanations? Marquardt suggests that perhaps the best VA HS students are choosing to stay in-state and attend college closer to home. If that were the case, we would expect to see SAT scores of entering freshman rise along with their GPA’s. Leaderman digs up data from the JMU Common Data Set to test Marquardt’s hypothesis, and to examine this possible alternative explanation for the GPA increase. He discounts this alternative explanation.
Looking more closely at the data, over this period, SAT scores for entering freshman remained relatively stable. JMU first started breaking down freshman HS GPA’s in the late 90′s, with yearly variations till a stable reporting form emerged in 05. From the JMU’s Common Data set (section C has SAT and GPA data):
In 98 Average GPA = 3.44 GPA> 3.5= not reported : GPA 3.0 or higher= 87.2%
In 99 Average GPA = 3.53 25 GPA percentile: 3.32, or 75% of all students had a GPA > 3.32
In 00 Average GPA = 3.53 breakdown by %age not reported
In 01 Average GPA = 3.58 GPA> 3.5= not reported : GPA 3.0 or higher= 96.5%
In 02 Average GPA = 3.64 GPA> 3.5= not reported : GPA 3.0 or higher= 98%
In 03 Average GPA = 3.66 No breakdown by %age reported
In 04 Average GPA = 3.68 No breakdown by %age reported
In 05 Average GPA = 3.673 GPA> 3.5= 68% : GPA 3.0 or higher= 97%
In 06 Average GPA = 3.69 GPA> 3.5= 70 % : GPA 3.0 or higher= 98%
In 07 Average GPA = 3.71 GPA> 3.5= 71% : GPA 3.0 or higher= 98%
In 08 Average GPA = 3.7 GPA> 3.5= 71% : GPA 3.0 or higher= 98%
Hmmm. What do you think?
While I don’t dispute the original Mason researcher’s hypothesis, the Inside Higher Ed writer jumped to conclusions about JMU without investigating all the facts.
Only about 55% of school districts in Virginia connect grades and SOL performance. Students from districts that don’t — including the giant Fairfax County — comprise a significant portion of students who come to JMU. Plus, 30% of JMU students are from out of state where there are no SOL’s. Given that these populations comprise a significant portion of our student population, the affect of SOL-related grade inflation on JMU’s student population is greatly over estimated.
Further, during the time period the Inside Higher Ed writer examines, JMU Admissions made the decision to increase the importance of grades and decrease the importance of the SAT in the application evaluation equation. Our mountains of data show that good grades is a far greater predictor of strong performance at JMU than the SAT. This Admissions decision has a great affect on the relationship of average grades and average SAT scores of the JMU student population.
Again, the researcher at Mason is likely on to something. And I am the last guy in town to defend SOLs and their affect on public schools. But had the reporter at Inside Higher Ed looked at JMU more deeply than simply using data available online, he would have found that many other factors mitigate the issue and that his conclusion was not sound.
I wonder if the researcher made an attempt to track the grading scales at the various high schools during the mentioned years. I graduated from a Roanoke high school in 2001, so SOL’s were just being put into place but there was talk about VA high schools inflating grades before SOL’s.
My high school graded on the following break down:
100-94:A
87-93:B
78-86:C
but a number of VA high schools had been switching to the college 10 point system in the 90′s. The complaint at the time was that this inflated the GPA’s of the students at schools with a 10 point system, making it easier for them to get into colleges.
It seems to me that a better way of comparing students would be on standardized methods like the SAT. I know I’m glad that JMU didn’t switch their focus to GPA until after I was admitted b/c my GPA sucked.