Exit Exams Decrease Graduation Rates

K-12 Testing

Two recent studies report striking evidence that exit exams decrease high school completion rates, increase GED test taking, and exacerbate inequalities in educational attainment. One paper concludes tests cause a two percent increase in the dropout rate, which could mean more than 40,000 students per year nationally are denied a diploma. Both reports found that the tougher the test, the more the dropouts. A third paper confirms an increase in dropouts in Virginia, particularly for Blacks and Latinos. And California has now denied diplomas to 20,000 students solely because they did not pass the state's exit exam.


"High School Exit Examinations and State-Level Completion and GED rates, 1975 Through 2002," by John Robert Warren, Rachael B. Kulick and Krista N. Jenkins, reports that exit exams are associated with lower high school completion rates and higher rates of General Educational Development (GED) test taking.


The researchers also found the effects on high school completion were more severe with the more difficult exams; completion rates were on average about 2.1 percentage points lower in states with more difficult exams. They also found that the negative effects become more pronounced in states with higher poverty rates and more racial and ethnic diversity.


Though a 2.1 percent graduation rate decrease might seem small, Warren and his colleagues calculated that with about half of all U.S. public school students facing exit exams, 42,000 students a year are not obtaining diplomas who otherwise would have. A tiny fraction of that amount, about 1,600, are receiving GEDs, leaving a staggering number of students at a disadvantage in pursuing higher education or employment. The authors suggest that as more states have moved to more difficult tests, the dropout consequences have grown. Most researchers have concluded that while a GED is better than no diploma, it has substantially less value than a regular diploma.


Thomas S. Dee and Brian A. Jacob found a somewhat more modest impact, with the harm most severe for Black male students. In "Do High School Exit Exams Influence Educational Attainment or Labor Market Performance?" Dee and Jacob also distinguished between relatively easy and more difficult exit exams.


Students in states with fairly easy exams were about 4 percent more likely to drop out than students in states without exit exams, while students were 5.5 percent more likely to drop out in states with more difficult exit exams. Black male students, the researchers found, were 5.2 percent more likely to drop out in states with easy exams, and 7.3 percent more likely to drop out in states with more difficult exams.


Dee and Jacob also examined the effect of Minnesota's exit exam and found it resulted in a higher dropout rate in urban, high-poverty and minority districts and a lower dropout rate in suburban, low-poverty districts. They conclude that the end result seems to be "increased inequality in educational attainment."


Authors of both papers argue that previous studies, some of which found no impact from the tests, suffered from shortcomings the current studies overcome.


A third study focused on Virginia graduation rates by the Editorial Projects in Education Research Center lends weight to other reports' negative conclusions. This study found that for the class of 2003, the last class not required to pass six Standards of Learning tests to obtain a standard diploma, 75 percent of Virginia public high school students graduated in four years. The rate for African Americans was just 64.1 percent, with only 57.6 percent of black males graduating on time. The Virginia Department of Education found that for the class of 2004, the graduation rate for black students fell to 61.3 percent, while the rate for Hispanics dropped by 11 percentage points, to 66.5 percent.


News reports show others states grappling with the effects of exit exams on high school completion. California alone has now denied diplomas to 20,000 students who would have graduated except for the test, some five percent of all 12th graders across the state. Arizona, however, has reduced the rate of those not graduating solely due to the test by allowing good student grades to compensate for lower test scores. In California, however, Governor Arnold Schwarznegger vetoed legislation that would have provided alternatives to students who did not pass the exit exam (see Examiner, May 2006).


Warren's team and Dee and Jacobs recommended further study to determine whether exit exams are associated with other problems, such as teacher burnout and excessive narrowing of the curriculum, or whether there are undiscovered positive effects that may outweigh the negatives.


o The Warren, et al., paper is in the Summer 2006 issue of Education Evaluation and Policy Analysis (Vol. 28, No. 2, pp. 131-152).
o Dee and Jacob's paper is available online from the Social Science Research Network http://papers.ssrn.com/sol3/papers.cfm?abstract_id=900985.