Grade Inflation Sources

ARE GRADES ACTUALLY GOING UP?

College — Clifford Adelman, “A’s Aren’t That Easy,” New York Times, May 17, 1995, p. A19 – describes a five-year study of “the records of 21,000 students from more than 3,000 universities, community colleges, and trade schools” that found “grades actually declined slightly in the last two decades.”  Original source:  The New College Course Map and Transcript Files: Changes in Course-Taking and Achievement, 1972-1993, 2nd edition. Washington, DC: U.S. Department of Education, 1999, esp. pp. 198-201.  Newer Adelman report: Clifford Adelman, Principal Indicators of Student Academic Histories in Postsecondary Education, 1972-2000. Washington, DC: U.S. Department of Education, Institute of Education Sciences, 2004.  (Self-report surveys finding to the contrary are reported in Arthur Levine and Jeanette S. Cureton, When Hope and Fear Collide [San Francisco: Jossey-Bass, 1998], pp. 124-27.)

High School – A 2002 RAND Corp. review of grades and standardized test scores of 23,900 high school students found no real grade inflation, at least in mathematics, from the early 1980s to the early 1990s.  (Dan Koretz and Mark Berends, “Changes in High School Grading Standards, 1982-1992.”  Available at www.rand.org/publications/MR/MR1445.)  Gerald Bracey did his own comparison of SAT scores with grade-point averages, looking at differences between 1977 and 1995, and similarly found no evidence to support charges of grade inflation.

SAT SCORES RISING AT SELECTIVE INSTITUTIONS

(For simplicity, only verbal scores are given below.   Math scores, which are available from the original sources, show almost exactly the same pattern in each instance.)

* At Harvard:   Mean verbal SAT score for class of ’85:   659.  For class of ’04: 738.
(Source, which also contains data for years in between): www.college.harvard.edu/dean/annualreport2001/annual_report_2001.html

* At the nation’s most selective colleges:
SAT scores were converted into national percentile scores to facilitate comparison across years, tests, and colleges.
At Rank 1   (“most competitive”) colleges, scores were as follows:
1960    92
1972    95
1982    95
1996    96

At Rank 2 (“highly competitive plus”) colleges:
1960    86
1972    92
1982    92
1996    93

Source: Caroline M. Hoxby, “The Return to Attending a More Selective College: 1960 to the Present,” n.d.
Available at http://post.economics.harvard.edu/faculty/hoxby/papers/whole.pdf.

* At all private colleges (actual verbal SAT scores):
1985     543
1992     543
1999     558

Source:  Trends in College Admission 2000, March 2002, Table 4.15.
Available at  www.airweb.org/trendsreport.pdf.

SAT BEING TAKEN BY GREATER PROPORTION OF HIGH SCHOOLERS (CONTRIBUTING TO OVERALL SCORE DECLINE)

See Condition of Education annual reports by the National Center for Education Statistics.  For example, Supplemental Table 22-1 (“SAT Test-Takers as a Percentage of High School Graduates…”) in the 1996 edition, drawing from College Entrance Examination Board data, provides percentages for each year from 1972 to 1995.  From 1972 to 1983, the percentage of (roughly one million) high school graduates taking the SAT fluctuated only slightly, from 31.0 to 34.1.  At that point it began to grow; by the mid-1990s, it hovered in the low 40-percent range.  (Available at http://nces.ed.gov/pubs/ce/c9622d01.html.)
In 2002, the College Board reported that an unprecedented 46 percent of high school students took the SAT.

When the SAT was taken by only a self-selected elite, we would naturally expect average scores to be higher.  Thus, a drop in scores paralleled by a rise in the proportion of eligible students taking the test offers no evidence of a decline in quality of U.S. students overall, or their schools.  Indeed, a review of SAT scores by state (in a given year) reveals a whopping negative correlation (r = -.86) with the percentage of students in each state taking the exam.  (Source:  Jeff McQuillan, The Literacy Crisis: False Claims, Real Solutions [Portsmouth, NH: Heinemann, 1998], p. 96n3)

GRADES & TEST SCORES DO NOT PREDICT CAREER SUCCESS (OR SATISFACTION)

Medicine — See Robert C. Davidson and Ernest L. Lewis, “Affirmative Action and Other Special Consideration Admissions at the University of California, Davis, School of Medicine,” Journal of the American Medical Association, vol. 278, no. 14 (October 18, 1997): 1153-58.

Applicants to medical school who were given special consideration – that is, those who were admitted as a result of an affirmative action policy or some other criteria despite the fact that their undergraduate grades and standardized test scores were relatively unimpressive – did indeed get lower grades in medical school but did just as well as their peers once they started working in hospitals, and then went on to pursue career paths indistinguishable from those with better grades or scores.

Law – David L. Chambers, Richard O. Lempert, and Terry K. Adams, “Doing Well and Doing Good: The Careers of Minority and White Graduates of the University of Michigan Law School,” University of Michigan Law School Law Quadrangle Notes, Summer 1999: 60-71.

“There is a strong, statistically significant relationship between LSAT [scores] and [undergraduate] GPA, on the one hand, and grades at the end of three years of law school on the other, but no significant relationship between the LSAT or UGPA with regard to what matters much more – the achievement of students after graduation” in terms of earned income or career satisfaction.  “There is a significant correlation, however, between [grades and scores] and our index of [community] service: in all decades, those with higher admissions index scores tend to contribute lessunremunerated service to society” (pp. 70-71; emphasis added).

General – Gordon E. Samson et al.., “Academic and Occupational Performance: A Quantitative Synthesis,” American Educational Research Journal, vol. 21, no. 2 (1984): 311-21.

A review of 35 studies revealed that academic indicators (grades and tests) from college accounted for less than 3 percent of the variance in eventual occupational performance as judged by income, job effectiveness ratings, and job satisfaction.  Moreover, these indicators had no predictive power whatsoever for M.D.s and Ph.D.s.

DOES STRINGENT GRADING PRODUCE BETTER PERFORMANCE?

Tougher grading was initially correlated with higher test scores, but long-term effects were negligible – with the exception of minority students, for whom the effects were negative: Julian R. Betts and Jeff Grogger, “The Impact of Grading Standards on Student Achievement, Educational Attainment, and Entry-Level Earnings,” National Bureau of Economic Research, September 2000, Working Paper 7875.
To order: www.nber.org/papers/w7875

For a list of seven other studies, only three of which found any boost in class performance (on multiple-choice tests) as a result of more stringent grading, see Kohn, The Schools Our Children Deserve (Boston: Houghton Mifflin, 1999), p. 255n88.

EFFECTS OF GRADES (AND OTHER REWARDS) ON LEARNING

See Kohn, Punished by Rewards and The Schools Our Children Deserve; follow index listings for “grades.”

OTHER SOURCES CITED (in order of mention):

Harvey C. Mansfield, “Grade Inflation: It’s Time to Face the Facts,” Chronicle of Higher Education, April 6, 2001, p. B24.  [ADDENDUM 2022: For a sampling of Mansfield’s views on other topics, see this website or this Twitter thread.]

Henry Rosovsky and Matthew Hartley, Evaluation and the Academy: Are We Doing the Right Thing?:  Grade Inflation and Letters of Recommendation  (American Academy of Arts and Sciences, February 2002)

Alfie Kohn, “Two Cheers for an End to the SAT,” Chronicle of Higher Education, March 9, 2001, pp. B12-13.

Ohmer Milton, Howard R. Pollio, and James A. Eison.  Making Sense of College Grades  (San Francisco: Jossey-Bass, 1986).