Statistical Malpractice: ACT-Tests Edition

by James A. Bacon

Here’s the smiley-face spin on the latest ACT college-readiness scores by Virginia students, straight from the Virginia Department of Education press release:

Virginia students again outperformed their peers nationwide by significant margins this year on the ACT college-admissions test. The performance of Virginia high school graduates improved in all subjects assessed by the four-part test, while the achievement of graduates nationwide declined in English and mathematics, and was flat in reading and science.

Here’s a frowny-face interpretation of the latest data: The percentage of Virginia high school graduates taking the ACT exams declined again in 2020: by two percentage points. Of the 19.3% of grads (not quite one in five) who took the exams, fewer than half (48%) rated college-ready in all four of the English, reading, math, and science benchmarks.

Technically, both versions are accurate. Which spin do you find more believable?

Similar to the more widely used SAT exams, the ACT exams test the college readiness of high school graduates. Virginia has among the highest average test scores of any state in the country. But the Old Dominion has one of the lowest percentage of students taking the exam. Fifteen states require 100% of high school graduates to take the exam. It’s an objective measure of a high school diploma’s worth.

It is reasonable to suggest that the smaller the percentage of students taking the exams, the better the average scores will be. Why? Because college-bound students are more likely to be academically prepared than non-college-bound students. Students who have no intention of going to college have no reason to take the exams, so their lower-than-average scores are not included.

In 2018, 24% of all Virginia grads took the exam, according to ACT data. In 2019, the percentage declined to 21%. In 2020, the percentage declined to 19% — or, according to the more precise VDOE data, 19.3%.

Colleges and universities are becoming increasingly skeptical of the value of standardized test scores, so many college-bound students did not take the tests. However, it is reasonable to assume that no one in Virginia, where testing is voluntary, takes the test unless they aspire to attend college. I find it frightening to think that half the test takers would need remedial work in one or more subject areas.

Bacon’s bottom line: In my admittedly jaundiced analysis, the Northam administration and its progressive allies are presiding over the greatest collapse in K-12 educational achievement in modern Virginia history. Under the circumstances, I would predict that administrators at all levels of the educational bureaucracy to obscure their failure by eliminating standardized tests when possible and spinning the results outrageously when not possible. Bacon’s Rebellion will stay on top of this ongoing story….

There are currently no comments highlighted.

13 responses to “Statistical Malpractice: ACT-Tests Edition

  1. The declining percentage of students taking the exam is probably the other side of the coin of the increasing number of students “graduating” because those standards are changing, as well. The ACT has always been focused on the college-bound, and those seeking to attend schools that seek that test result, thus a subset of the most prepared.

    The cultural winds forcing down educational achievement (put digital screens at the top of the list) are far too deep to blame on one governor or one political party. Reading and writing skills are eroding. Reason is replaced with cant. That letter from the UVA student embarrassing UVA was Exhibit A. Impenetrable.

    • In a sea of stupidity the smart fish eat well. We need more school choice in Virginia. The more dumbed down public education becomes, the more successful those who refuse to accept a dumbed down education become.

      It may be time to start a private version of Thomas Jefferson High School with loans for the students who pass the older, harder entrance exam. I’m sure the watered down TJ will be able to beat the new private TJ at football but I can hear the students from the new private TJ in the stands now …

      “That’s alright. That’s OK. You’re going to work for me one day.”

    • It’s not just reading and writing skills that are declining. People can’t seem to speak these days without saying “like” in every sentence.

      What’s um, like, ya know, up with that?

    • The fact that only a little over half of those taking the test were “college ready” in all areas is appalling. But, I agree with Steve, you can’t blame this on the Northam administration. Northam has been in office for less than three years; the kids taking the tests were already in high school when he took office. Their already inadequate base had been set and I am sure that the teaching available to juniors and seniors when Northam took office are still available now.

  2. … and all the children are above average.

  3. Jim says:

    “Colleges and universities are becoming increasingly skeptical of the value of standardized test scores, so many college-bound students did not take the tests.”

    I disagree. I think the reverse is true, namely, that colleges and universities are becoming increasingly afraid of the great value of standardized test scores in predicting college success, because that success and accuracy will limit their ability to admit under prepared kids into their colleges for their own selfish reasons that are increasingly being found illegal and/or bad practices that harm all kids, and most harshly harms disadvantaged kids.

    In fairness to Jim, his last paragraph limits what I consider his earlier misinterpretation. That last paragraph says:

    “Bacon’s bottom line: In my admittedly jaundiced analysis, the Northam administration and its progressive allies are presiding over the greatest collapse in K-12 educational achievement in modern Virginia history. Under the circumstances, I would predict that administrators at all levels of the educational bureaucracy to obscure their failure by eliminating standardized tests when possible and spinning the results outrageously when not possible. Bacon’s Rebellion will stay on top of this ongoing story….”

    But here again, as Steve suggests, this collapse has been going on for at least four decades. Although the collapse has been greatly accelerating over past decade, concurrent with the collapse of our culture, and the pernicious affects of high tech that offers so many ways to amuse, without learning. Hence reading and writing, social, and emotional skills are evaporating among our youth at an alarming rate, even among those most talented and advantaged as seen in our elite colleges, such as UVa., that often acerbate these horrible trends, instead of ameliorating them.

  4. James Wyatt Whitehead V

    Alabama waived the SAT/ACT requirement for 2021. 3.0 GPA automatically gets you in a number of Alabama colleges. The College Board must be having a bad case of hemorrhoids over this. $1.1 billion profits for the CB in 2017. I suspect the same might be coming to Virginia. No SAT/ACT required.

    https://mynbc15.com/news/local/pandemic-causes-alabama-colleges-to-wave-actsat-requirements-for-2021

  5. I agree with those who say we can’t blame the decline entirely on the Northam administration. The Northam team inherited a lot of policies from previous administrations. But the Northamites have done nothing to reverse the decline and, in fact, have accelerated it with their social-justice rhetoric. Their goal is not to improve overall test scores as much as to close the black-white gap. And even there, they are unsuccessful. Their policies are not working by any measure.

    • The NAEP measurements of proficiency have been stagnant for a long time and not just in Virginia but nationally.

      Further, the stock “advice” from many critics is to provide vouchers to private schools – but without the same level of transparency and accountability that we hold public schools to.

      Which makes one wonder what the critics of public schools really want.

      I think as soon as Jim points to Northam, he basically invalidates his entire post.

      The best kids from US schools are better or equal to the best students from other countries. Where we fall down is on the percent of proficient. We have maybe 1/3 and other developed countries do better but to characterize the public school system as a “failure” is just plain bogus when we are ALSO producing some of the best students in the world.

      The picture is not a fail unless one considers anything other than perfect as a failure, It’s more complicated and “gray”.

      This country would be nowhere without public education. We would not be a world leader in anything.

      • The best students in the world will be successful no matter how good or bad their schools and teachers are. The measure of success or failure for our public schools should be how well they educate the next tier of students, the “not-quite-the-best students” in the world.

  6. The best students in the world come from countries that pretty much provide universal education to their citizens.

    Once you recognize that it’s public education that makes the difference, it becomes an issue of which countries do the better job of producing BOTH the top tier students as well as students that are “proficient” in terms of meeting the standards that are defined as “proficient” – which is a construct of public education – PISA.

    Public Education is an inherently socialist concept – not a capitalist concept. It transfers wealth from people to be used to pay to provide universal education for all kids regardless of their income or wealth.

    Libertarians and wannabes – claim that it’s not the job of the State to provide universal education. There actually was no such thing in any country on earth when Jefferson was alive.

  7. Private-school students perform better on standardized tests than do public-school students (https://www.jstor.org/stable/20642346?seq=1). According to the report, when corrections are made for student, family, and school characteristics, the public-school students do somewhat better. In other words, the hard data show that private-school students do better, but when the data is modified by soft factors, public-school students do better. I wonder how much class discipline and order is contained in the adjustment factor “school characteristics.”

    • Does it depend on who you ask:

      “No, private schools aren’t better at educating kids than public schools. Why this new study matters.”

      Despite evidence showing otherwise, it remains conventional wisdom in many parts of the education world that private schools do a better job of educating students, with superior standardized test scores and outcomes. It is one of the claims that some supporters of school choice make in arguing that the public should pay for private school education.

      The only problem? It isn’t true, a new study confirms.

      https://www.washingtonpost.com/news/answer-sheet/wp/2018/07/26/no-private-schools-arent-better-at-educating-kids-than-public-schools-why-this-new-study-matters/

      See, this is the problem. If private schools want to make that claim then they should willingly disclose their results publically every year and let the public decide and if it’s true then build support for public funding for disadvantaged kids – which I would support.

      What I don’t support is no transparency while citing “studies”. Get it all out in the open and let the public decide on a true apple-to-apple basis.

Leave a Reply