Study: Exit exam hurts girls and minorities

Researchers with Stanford University’s Institute for Research on Education Policy and Practice found that the graduation rates of girls and non-white students have plummeted as a result of California’s high school exit exam requirement.

Looking at four California school districts, researchers analyzed the graduation rates and other data for kids who tested poorly (in the bottom fourth of all students) on standardized reading and math tests that they took earlier, in the eighth and ninth grades.

The study compared what happened to those low-performing students in the Class of 2005 — before the new graduation requirement took effect — with similar students at the same schools in the classes of 2006 and 2007. 

The study found that girls and non-white students fared the worst as a result of the policy change. The graduation rates of the academically struggling black students and girls in the study dropped 19 percentage points after the requirement kicked in, compared to just 1 percentage point for struggling white students and 12 points for boys, according to the report.

Other findings? Little evidence that the high school exit exam is causing kids to drop out of school, as feared, or that it’s raising academic achievement, as hoped. It is, however, causing graduation rates to dip, overall. The study estimates that 3.6 to 4.5 percent of California students — up to 22,500 a year — don’t graduate because of this new policy.

State Superintendent Jack O’Connell, who championed the exit exam, quickly issued a news release in response to the study, saying:

… The heart of this report speaks to why I’ve called out California’s racial achievement gap and why I am so committed to implementing the14 recommendations made by my P-16 Council aimed at closing these gaps. The recommendations include the creation of a statewide strand of culturally relevant pedagogy and a culture survey of our students and education staff to discuss and address issues of unconscious racial bias in our schools.

I believe that the biggest mistake we could make is to view this report as a reason to lower our expectations for any student, but especially for our students of color and females. While reports like this call for us to redouble our efforts to improve instruction and effective interventions, I remain wholly committed to maintaining a high standard of expectations for all students. As a result, I have asked my staff and HumRRO, the CAHSEE evaluator, to conduct further review of the study so we can look for ways to better meet the educational needs of all students and help them succeed in school, on the CAHSEE, and in life.

But the report recommends that California “explore new ways to improve education and hold our high schools accountable for the academic achievement of their students without the negative and inequitable effects of the exit exam.”

Do you think California should keep this requirement intact? Does the study convince you otherwise?

photo from luminafoundation’s photostream at flickr.com/creativecommons

Katy Murphy

Education reporter for the Oakland Tribune. Contact me at kmurphy@bayareanewsgroup.com.

  • Nextset

    The exit exam is a good thing and should be retained. It’s an easy test, passing is 8th grade performance level. Maybe we should use it as a test for enrollment in high school.

    Bad students, and that’s what the failing people are, need to be kept out of the graduation ceremonies and not given diplomas. I’m not buying that they weren’t “taught” enough. They didn’t want to learn or are (less likely) unable to learn due to retardation. Yes there are some people who are so dull that they cannot read, write and process information below the level of the average 14 year old at 18 or 19. These kids should have been in special ed anyway and not enrolled in a normal high school.

    Normal students can pass the thing years before graduation. those who are up to the last minute and haven’t passed often pass with intensive study (of 8th grade reading/writing). If the “intensive study” had been done previously maybe they would have passed earlier.

    Dull students need strong discipline to perform – something they never get in the typical urban school. That can be fixed.

    Without this test the HS diploma would be meaningless, just a piece of paper the government schools run off on a ditto machine and throw from windows.

  • Nextset

    Oh, and “girls and minorities” may be having the problems they are on this test because indiscipline seems to be greatest with them. That’s real easy to fix.

  • Oakland Teacher

    Certainly Nextset is not serious?

  • Jonathon

    For the sake of Oakland children’s, I hope that Nextset is in fact the opposite of serious.

  • Bay Area resident

    It’s not the exam that is hurting these students, it’s the education system that didn’t provide them with a solid education. Getting rid of the exit exam won’t make the students any more prepared to succeed; improving the education system is what is needed.

  • Pingback: Good news about Oakland’s graduation rate? - The Education Report - Reporter Katy Murphy’s blog on Oakland schools()

  • Nextset

    Oakalnd teacher: I assume you refer to post #2.

    In my experience public schools are not in the habit of “riding” minorities. I suppose they either don’t want to fight with them or don’t expect anything from them anyway – so the schools don’t correct their bad language & grammar, get in their faces about behavior and attitudes, or push them to do anything other than what is comfortable & politically correct. In other words, the only school discipline they are getting is during football/volleyball practice. Maybe that’s not true but it’s my working hypothesis. This is what I see as a guest speaker in high school classes (not in OUSD), from my time a lifetime ago as a sub teacher, and from what I see at the mall, the courts and around town.

    As far as the girls, I think they are far more boy crazy lately than when I was in high school. Spending time worring about your “boyfriend(s)” instead of yourself and your family at age 15 is a problem and a reflection of the training the girls are getting at school. We don’t fight with minorities and girls enough for their own good. Maybe we assume someone else will take care of them all their lives. We shouldn’t.

    If the students knew that any significant failure to perform in school would result in a transfer out to a lowest common denominator school and a loss of your friends at the current school, they would tend to perform at their capability. Private schools operate on this basis and don’t retain F students (as far as I’ve experienced). There is no good reason the public schools don’t operate the same way for their academic programs.

    And 8th grade level “exit exam” would not be an issue with academic schools by the time anyone got to 12th grade if the schools were run appropriately. The exit exam isn’t a problem elsewhere. The bottom 25% will have problems and even then those problems should be addressed every year, not just in 12th grade. Those students in such distress with math/verbal skills should have been in special ed/voc ed by 10th grade.

    I don’t care how bad their families are. The schools have enough face time with the kids to make something out of someone even if the parent(s) are trash. We’ve used the public schools for over a Century to promote social mobility and we should do so again – overtly. I see plenty of (poor) immigrant kids rising even with public school education and I know we can get much better results with the “minorities”, etc if we start working them harder not softer.

    The test isn’t the problem. Bad students are the problem. Bad students can either be corrected or transferred out to more suitable (for them) schools.

  • TheTruthHurts

    Clearly, this needs to be looked at. But why we would expect the test to show the opposite of every other test (an achievement gap exists), I don’t quite understand.

    I could care less whether we have THIS exit exam, but we need SOMETHING. If a diploma is to mean more than an 18th birthday, we have to have “PROOF” that these kids are learning something.

    If the test is biased, fix it. Hell, bias it towards women and minorities. Whatever we do, don’t get rid of proof. It is embarrassing that with all the complaints of teaching to the test, we still can’t get more students to pass the test.

  • Katy Murphy

    I think the point they’re making is that the student outcomes on this exam, broken down by gender and ethnicity, aren’t consistent with how the same kids do on other standardized tests.

    All of the students included in the study were generally lower-performing, regardless of gender or ethnicity.

    So why are the outcomes of girls and boys, whites and minorities, so much different for this particular exam, when their scores were comparably poor by other measures?

  • TheTruthHurts

    Those are great questions and should be answered. I just don’t want this to be another parade of test-bashing. If someone wants to propose a tax so we can afford “alternative” performance measures, I’m all ears. Maybe Berkeley should admit based on essays and we should graduate based on science projects? Someone will have to pay for those evaluations and they are likely far more subjective and fraught with bias.

  • Pepe

    No one is arguing that we lower standards because of these students, but we do need to take into consideration the biases of such high stakes testing. These sort of inconsistencies of outcomes seem typical across the board when it comes to many of the different types of tests that students take. Placing so much emphasis on one questionable test is not ethical.

    While I normally disagree with everything else he says, I agree with Nextset (did I just say that?) that we need to start working students harder. Minorities often aren’t pushed to learn because their teachers are afraid of them or have lowered expectations for them.

    However, as Katy pointed out, this issue seems to go further than that. On the exit exam, certain populations of students who have met the standards or even excelled in other forms of assessment did not do well on the exit exam. I’ve noticed similar results on the Star test. As a school volunteer, I tutored one boy who could have taught someone the entire first year of algebra by the end of the year, but this student scored below grade level on the math Star test. Unfortunately, testing often only measures a person’s ability to do well on that test, not the amount of knowledge a person may have acquired.

    How can we get more authentic measures of student knowledge and growth?

  • Nextset

    Pepe: Was that student able to re-test? It’s always possible that a single test administration was thrown off by something that was going on with the student.

    Another point – the failed exit exam is typically the last thing in a series of failures with the troubled students. I once reviewed a large number of Urban Public High School transcripts for a scholarship program & interviewed applicants. There was nothing subtle about the students who couldn’t read. Rather than a single bad score there were a series of test failures for people who read at the 6th grade level at (end of) 11th grade. One of these candidates, a girl, had the nerve to come in for the interview with some nonsense about wanting to be accepted at UC Davis and go pre-med.

    Delusional thinking. I pointed out her history of childish reading/writing levels and asked her how she expected to function a single week at University level. Her answer was that no one had ever told her there might be a problem (and that’s what gets me mad). Of course her grades didn’t reflect her severe inability to read and write. Her verbal scores stank her entire school career. And she said she’d never even been talked to about remedial programs, summer schools or anything else to work on the issue. I’m not sure if I asked her to read aloud from Business Week or the Wall Street Journal which I had handy. Or even Time Magazine… I did that occasionally. It is not difficult to detect inability to read.

    In my experience the problem is not a matter of ability to do well on “a test” but rather people with clear & acute problems not being addressed – just patted on the head until quitting time.

    And it’s rough out in the Brave New World. Don’t think an employer or examiner wouldn’t just ask an applicant to read something aloud.

    The public school candidates who got those scholarships were immigrants. Who would have guessed.

    Brave New World.

  • Pepe

    Nextset, there are no retakes on the Star test, and my experience has been echoed by many teachers that I have talked with about it.

    Your explanation does not address the inconsistent results brought up in Katy’s article. What’s your take on this? It can’t just be clear and acute problems not being addressed when there is such a clear pattern over a large sample group.

    I think one of our problems is the fact that the testing industry is such a cash cow. Because our politicians rely on these companies financially, they also rely on the (mis)information these companies feed all policy makers. How are these testing companies held accountable?

  • Catherine

    The math portions of the test are not biased. Perhaps some discussions in the language arts portion discuss things about the Grand Canyon, etc. and are biased, but high school students begin taking this exam in the 10th grade. There are practice exams all over the place. Students can practice the exams if they want to do well.

    I think we need to talk about motivation. I also think that a student should be able to go into the computer lab at school and print out sample tests. Of the students who did not pass, how many actually took sample tests home and studied? This would be an interesting question to ask the students who failed.

    If a person did not pass the written driver’s license test, yet did not study we would not say that they should be given a license anyway or that the test was biased, we would expect them to study. If they wanted to be a driver of a big rig we would require that they pass the test, I don’t understand why we are not expecting the same of high school students.

  • Bay Area resident

    With regards to this exit exam being biased while other STAR tests aren’t, that simply can’t be the case. The high school exit exam (CAHSEE) uses the exact same types of questions as the STAR test. For math, in fact, most of the questions are plucked from the 7th & algebra STAR tests verbatim. The CAHSEE is an amalgamation of standards/test items from other grade-level STAR tests, so differing bias outcomes between STAR and CAHSEE tests would be very confusing.

  • Bay Area resident

    Upon reading the study in more detail, and not just the article summary, the authors do not claim that the CAHSEE is biased. They instead come to the conclusion that the unexpected underperformance of minorities and women (relative to what their prior STAR performance would predict) is due to the “stereotypes threat” theory related to the high-stakes nature of the test. It’s an interesting point.

  • Katy Murphy

    Thank you for pointing that out, Bay Area Resident. I was planning to make note of that distinction as well.

    I hadn’t heard of “stereotype threat” before reading this study. Is this a prominent theory?

  • Pingback: California’s high school exit exam: out the window? - The Education Report - Reporter Katy Murphy’s blog on Oakland schools()