PA Officials Want to Replace Bad Keystone Exams with Bad College Entrance Exams

Screen Shot 2019-07-15 at 8.34.43 AM

Pennsylvania officials are scandalized that the Commonwealth is wasting more than $100 million on unnecessary and unfair Keystone Exams.

They’d rather the state spend slightly less on biased college entrance exams.

State Auditor General Eugene DePasquale and State Sen. Andy Dinniman held a joint press conference last week to introduce a new report compiled by DePasquale’s office on the subject which concludes with this recommendation.


Replacing bad with bad will somehow equal good?

Under the proposal, elementary and middle school students would still take the Pennsylvania System of School Assessment (PSSA) tests. However, instead of requiring all high school students to take the Keystone Exams in Algebra I, Literature and Science, the report proposes the same students be required to take the SAT or ACT test at state expense.


This is certainly an improvement over what the state demands now, but it’s really just replacing one faulty test with another – albeit at about a $1 million annual cost savings to taxpayers.


The report does a good job of outlining the fiscal waste, lack of accountability and dubious academic merits of the Keystone Exams, but it fails to note similar qualities in its own proposal.


From 2008 to 2019, the state already paid Minnesota-based Data Recognition Corp. more than $426 million for the PSSAs, Keystone Exams and Classroom Diagnostic Tools (an optional pretesting program). The federal government paid the company more than an additional $106 million. Officials wonder if this money couldn’t have been better spent elsewhere, like in helping students actually learn.


DePasquale, who recently launched a congressional bid, puts it like this:


“When the federal law changed in 2015, why didn’t Pennsylvania begin to phase out Keystone Exams? I could understand if they use them for a short period of time after that, but it’s been four years, and will cost taxpayers nearly $100 million by the end of the contract for tests our students do not even need to take.”


The federal government dropped its mandate four years ago and the state legislature did the same last year.


Originally, state lawmakers intended to make the Keystone Exams a graduation requirement, but in 2018 they passed legislation to make the assessments one of many avenues to qualify for graduation starting in 2021-22. Students can instead pass their core courses and get into college among other things.


“The Department of Education itself said they [the Keystone Exams] are not an accurate or adequate indicator of career or academic readiness,” Dinniman said. “So what I’m always surprised about is, they said it and then they continue to use it. These tests have faced opposition from almost every educational organization that exists. And when we got rid of the requirement and put in [more] pathways to graduation, this was passed unanimously by both the Senate and the House.”


The federal government also changed its testing mandate. It used to require all public school students to take state-specific assessments in grades 3-8 and once in high school.


When Congress reauthorized the federal law overseeing education in 2015, it offered states more flexibility in this regard. Elementary and middle school students still have to take a state-specific test. But now the high school portion can be fulfilled with college admissions tests – and, in fact, a dozen other states legislate just such a requirement.

Democrats DePasquale and Dinniman think the SAT and ACT test are an improvement because students who taken them are more likely to go to college. But that’s a classic case of confusing correlation and causation.


Students motivated to go to college often take these exams because they are required to get in to a lot of these schools. Taking these tests doesn’t make students MORE motivated and determined to enroll in post-secondary education. They’re ALREADY motivated and determined.


Moreover, one of the faults the report finds with the Keystone Exams is that the assessments measure student’s parental income more than children’s academics.


Kids in wealthier districts almost always do better on the Keystone Exams than those in poorer districts. In fact, the report notes that of the 100 state schools with the highest scores, only five were located in impoverished districts —where the average household income is below $50,000.


Yet the report fails to note that this same discrepancy holds for the SAT and ACT tests. Poor kids tend to get low scores and rich kids get the highest scores.


In fact, the College Board – the corporation that makes and distributes the SAT – recently started adjusting scores on its test in an attempt to counteract this effect thereby accounting for high schools and neighborhoods “level of disadvantage.”


Does this creative scoring actually work? Who knows – but it’s kind of like being forced to swallow poison and an antidote at the same time when any sensible person would simply refuse to swallow poison in the first place.


And that’s the best solution state officials have for our children.


They’re suggesting we replace discriminatory Keystone Exams with discriminatory college entrance exams.


To be fair, DePasquale and Dinniman are somewhat constrained by boneheaded federal law here.


Though the Every Student Succeeds Act (ESSA) is an improvement over No Child Left Behind, it still requires all high school students to take standardized tests.


Given what we know about the limits and biases of these assessments, policymakers should remove that hurdle altogether. But until the federal government gets its act together, one could argue that DePasquale and Dinniman’s policy suggestion may be the best available.


When you can’t do right, maybe it’s best to do less wrong.


But we must acknowledge that this isn’t the ultimate solution, it’s only a stopgap. We must continue to push for intelligent assessment policy that’s best for our children.


Standardized testing should be eliminated altogether – especially in high stakes situations. Instead we should rely on classroom grades, portfolios of student work and/or other authentic measures of what children have learned in school.


Accountability – the typical reason given behind these assessments – should be determined by the resources provided to students, not a highly dubious score given by a corporation making a profit off of its testing, test prep and ed tech enterprises.


The most we can expect from DePasquale and Dinniman’s program if it is even considered by the legislature is a band-aid on a gaping wound.


Read the full report, Where Did Your Money Go? A Special Report on Improving Standardized Testing in Pennsylvania.


Like this post? I’ve written a book, “Gadfly on the Wall: A Public School Teacher Speaks Out on Racism and Reform,” now available from Garn Press. Ten percent of the proceeds go to the Badass Teachers Association. Check it out!


24 thoughts on “PA Officials Want to Replace Bad Keystone Exams with Bad College Entrance Exams

  1. You have to know this is coming, eh Steven!

    “Moreover, one of the faults the report finds with the Keystone Exams is that the assessments measure student’s parental income more than children’s academics.”

    The tests don’t measure anything. By continuing to use incorrect, purposely misleading and many times outright lies of the educational malpractice supporters we lend credence to their bullshit.

    There is a correlation between parental income/wealth with standardized test scores no doubt. But a correlation is not a “measurement”. To confuse and conflate the two is wrong and only serves to obfuscate the intentions and results of those who push (yes as in a drug pusher) these malpractices.

    Quit using there falsehoods. . .

    . . . please!


    • Duane, I was quoting the report’s findings about the Keystone Exams. I was trying to explain that the same logic they use against the Keystones could be used against the SAT and ACT. You and I both know these tests are meaningless measures. I was trying to show that the report is inconsistent.

      Liked by 1 person

  2. A few years ago Maryland had a plethora of tests that could be used for graduation since parents were revolting against PARCC. It was so confusing and no one could figure out if their child would graduate. They set minimum scores for SAT, AP, and ACT, also IB, Accuplacer or PARCC. Now, it’s all gone and all we have left is PARCC…even though ALL the high school kids start taking PSAT in 9th grade and are encouraged to pay to take the SAT. ALL the tests need to go away!!!

    Liked by 1 person

  3. I think having a second opinion is useful. Looking at standardized test scores along with grades will give parents a more complete understanding of the education their children are receiving. It is probably better to use a national exam rather than a state based exam, so I think this is an improvement.


    • TE, it depends on who’s giving the second opinion. Doesn’t it? Why would you trust a big corporation who is incentivized to fail students and then makes more money off the remediation materials? It’s beyond foolish – unless you are somehow profiting off of this unnecessary and greed-based industry. But you’re entitled to your opinion. Most wealthy people agree with you.


      • In my experience test scores are correlated with academic ability. A student with an ACT math score of at least 30 (about 700 SAT) will not have trouble with the mathematics of intermediate microeconomics, a key course for the undergraduate major. Students with scores of no more than 18 on the ACT math exam (about 500 SAT) will have a great deal of difficulty with the course are are likely not to pass.

        At my university, ACT scores are highly correlated with student success for the students whose high school academic GPA is below 2.4.


      • TE, you could find the same correlation by looking at applicants parental income. Would that be a fair criteria for college acceptance? But in relying on these test scores, you clothe inequality in “respectable” statistics. That’s not acceptable.


      • Steven,

        My university does not use test scores for admission. Any student with a 2.0 high school gpa or better in academic courses is automatically admitted. That is why we have students with math ACT scores below 18 and above 30.


      • TE, what exactly is your point? Do you mean to suggest that certain students are too dumb to take your classes? That it’s not your responsibility to teach certain children? Students must come up to your standards before you’ll condescend to spend your valuable time trying to teach them? If so, that’s a dangerous and wrongheaded point of view often espoused by some educators at the college level. I’ve addressed this before in detail in this article: College Remediation is Less About Bad Students Than Academic Elitism –

        Before I go, a note about what comments I let on to my site. As you know, I let the vast majority of your comments onto my blog. Because of that, the WordPress algorithm automatically lets most of your comments through without my approval. When I see them, I either let them stand or suspend them temporarily or otherwise. I’d prefer nothing to get into this page without my approval first, but WordPress either doesn’t work that way or I’m just ignorant of how to properly navigate it. As such, some of your comments get taken down before I decide to let them stand. I need time to think about them and decide. I know that upsets you because you often send rapid fire responses. Please give me time to think. I do not exist only on-line. I have things to do other than moderate this blog. Thank you.


      • It isn’t that difficult to change the settings for a WordPress Blog so all comments must be approved by you. Back in 2013 when I was tricked into a flame war, I ended up changing all my blogs so all comments had to be approved by me before they’d appear.

        The link takes you to a WordPress tutorial on YouTube that hopefully will help you do what I did years ago.

        Liked by 1 person

      • Steven,

        My point is that mathematics standardized exam scores do a good job predicting how well students will do in economics classes. It is simply incorrect to say that the scores are uninformative about a students ability to do mathematics. Perhaps it is more obvious at my institution because of the relatively open admission of students where standardized test scores play no role (well over 90% of applicants are accepted). It is common that students would have difficulty solving Y = cY+I+G for the variable Y.

        Do I have enough time to teach economics and algebra to each of the 600 students I will have in the fall semester? Probably not. Do students have enough time to learn economics and algebra in my class, along with the other courses on their schedule next semester? Again, I think not. That is why it is important to admit students who have a good chance to succeed rather then take money from students who have little chance at graduating. Don’t you agree with this last point?


      • TE, we’ve gone over this ground so many times. First, you have not proven that standardized test scores do a better job of predicting who will do well in college nor have you proven they do a better job than just letting in the wealthiest applicants or even a better job than high school GPAs. Second, schools of all stripes should meet students where they are. That means providing the teacher with the resources she needs to meet those needs, too. If you have so many students who need remediation, that should be provided and you should not be solely responsible for doing it. These are the problems educators are encountering at all levels. When you realize that and join the fight, you will do more to improve education than continually championing exclusion and statistics over justice and equity.


      • Steven,

        I am not arguing that standardized exam scores do a better job than teacher assigned grades. I am arguing that standardized exam scores in combination with teacher assigned grades do a better job than teacher assigned grades alone. If you have two students, each with a high school gpa of 3.2, and one has an SAT score of 1530 and ACT score of 36 and the other student has an SAT score of 975 and ACT score of 18, you would not say each student is equally well prepared to attend college.

        The most unjust thing a university can do is to accept a student for enrollment, accept the student’s payments, knowing full well that the chances of the student graduating are vanishingly small. That is what worries me most about my universities relatively open enrollment.


      • TE, so college admissions should be dependent on grades AND parental income? Because at best economic factors are all the additional information these test scores will give you. You have not proven that standardized test scores tell us one additional thing about students. Look – I agree that students need to be provided with the resources to succeed but that doesn’t have to mean economically elitist and classist enrollment policies.


      • Steven,

        SAT and ACT scores are not as closely correlated with income as your post suggests. Georgetown’s Center on Education and the Workforce did an interesting study looking asking how admission to the 200 most selective colleges and universities would change if the only admission criteria was SAT score.

        If you were correct, that SAT scores are simply a measure of family income, then using only SAT scores for admission should result in admitting far larger proportion of students from wealthy households. The Georgetown study does suggest an increase, but only a small one. Under the holistic admission used by these schools, 60% of the incoming class were from the most affluent 25% of families. Under an SAT only admission system, 63% of admitted students would come from the most affluent 25% of families.

        Using SAT scores as the only admission criteria would lead to different students being admitted. The study found that 53% of students actually admitted to the most selective colleges and universities would not have been admitted under an SAT only admission process. The majority of students that would fail to gain admission under an SAT only admission standard would also be from the most affluent 25% of families.

        Many interesting points here. You can read it here:


      • TE, did we read the same study? It seems to support my argument more than yours. The study concluded that basing admittance entirely on test scores would fill the 200 most selective colleges with MORE white and advantaged students. Researchers found a test-only admissions policy would raise those colleges’ share of white students from 66% to 75%, while diminishing the share of black and Latino students from 19% to 11% and Asian students from 11% to 10%.
        More than half (53%) of the students attending such schools would be replaced if they switched to test-only admissions policies. Of that group, roughly half would come from families in the top income quartile.

        Yes, some of the most affluent students score high but not the highest. So what? They still score high enough to get in – higher than most impoverished students.

        Researchers concluded that this study offers evidence to rely less on test scores for admissions.

        “In the wake of the college admissions scandal, our thought experiment tested whether removing legacy and social capital from the admissions equation would have a more equitable outcome,” said Anthony Carnevale, Georgetown University’s Center on Education and the Workforce (CEW) director and the report’s lead author, said in a statement. “But a test-only admissions policy would only further privilege in the higher education system.”

        The authors contend standardized tests have had too much weight in admissions and may not be the best predictor of an applicant’s success in college. Again this seems like support for my argument, not yours.


      • Steven,

        I did read it. Using only the SAT score would result in a 3% increase of affluent students, from 60% to 63%. If you used family income to determine placement, I would expect to see a 40% increase, from 60% to 100% of students would be affluent. Would you expect to see a smaller increase? That is the difference between using SAT exam scores and family income as admission requirements.

        Of course, I do not advocate only using the SAT score for admission, but argue that it should be one of a set of things universities look at before they ask students to pay many thousands of dollars to the university.

        In a world where the average student admitted to UC Berkeley had a weighted high school gpa of 4.45, how do you think low income students, who attend high schools without access to honors or AP courses, will do if you only consider high school gpa for admission? I would be surprised if only 63% of students would come from affluent families.


      • Oh my goodness, TE. You cannot see the forest for the trees. You ignore the conclusions of the study you, yourself, cited to cherrypick one statistic within it. This is troubling for many reasons. First of all, you’re only talking about “affluent” students – kids from families with a median annual income of at least $122,000. They would only increase by 3 percent if we used just test based admissions. Great. I’m not in that income bracket. Neither are most of my students. But the more parents earn, the more likely their kids are to score highly on standardized tests. That’s my point – not that all kids whose parents make more than $120K ace the tests. Second, the Georgetown study was not conducted to assess the correlation between economics and test scores. It was conducted to discover how admissions at 200 of the top colleges would be effected if SAT and ACT scores were the only criteria. They only looked at kids who applied to these schools. That matters. All impoverished kids going to college didn’t apply – only those who thought they had the best chance of getting in. Moreover, researchers didn’t look at everyone who applied. Once they filled 300,000 seats, they stopped. I wonder how many applications they didn’t even examine.

        Moreover, I am not saying GPA is perfect. I am saying it is better than standardized test scores. We need to increase resources at impoverished schools so that all kids have the same opportunities. We need to do the same for low income and struggling students at colleges and universities. Standardized testing is no part of the answer. We’ve been using these tests for decades. Time to try something that actually works.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.