Cut Scores on Standardized Tests Are Arbitrary; Pearson’s GED Recalibration Proves It

It is hard to remember that standardized test scores have no real, objective meaning.  How easily we forget this truth.

The standardized tests that are used to rate and rank public schools—the ones that became ubiquitous after the passage of No Child Left Behind, and which, by the way, the Every Student Succeeds Act will leave in place—are graded with cut scores.  Somebody in your state education department decides on a cut score that will mean “passing” and other cut scores that will mean “proficient” or “outstanding.”  The people who set those cut scores have a lot of power because the scores on the tests confer value judgments on children and on their teachers and on their schools. The scores can shape the self concept of a child or determine whether a teacher gets a raise or even whether a teacher gets fired. Even if a third grader, who began the school year as a non-reader, learns how to read and makes a lot of progress, that child and that child’s teacher won’t get a passing mark unless the child makes the 3rd grade reading proficiency cut score.

But the cut scores are relatively arbitrary.  And sometimes they even get changed.  Here is a commentary from US News and World Report about Ohio last November:  “Last month, Ohio released initial results from its PARCC test, a standardized test used by a group of states, showing that nearly 65 percent of students were proficient in math and English language arts. These numbers were a whole lot better than those of other states on the PARCC, so is Ohio leading the way? Politically, yes. Academically, no.  Ohio got these results by rejecting PARCC’s recommended cut scores to designate proficiency and adopting more generous ones. Ohio claims that 69 percent of its fourth graders are proficient in English, but if the state were using PARCC’s cut scores, the results would suggest that just 37 percent of Ohio students are on track for college or career.”  Although U.S. News criticizes Ohio for changing the cut scores to ensure that additional children passed the tests in this controversial year when Ohio implemented the Common Core and its harder tests, someone in Ohio had realized that failing too many children would only make everybody look bad—schools, teachers, and children—so the powers that be in Ohio just fixed the scores.

The latest example of the adjustment of cut scores comes from the corporate education giant, Pearson, a British company with global reach that has been losing money because a lot of states in the U.S.  have been dropping the PARCC tests, which it produces.  Last week, it was reported by the BBC that Pearson is cutting 4,000 workers worldwide—10 percent of its workforce.

The other division of Pearson that has been struggling is the one that runs the GED—the General Education Diploma—in the United States.  This is the test young adults who lack a high school diploma can take to earn high school equivalency.  You may have forgotten that in 2011, in collaboration with the American Council on Education, Pearson took over the GED.  And in January of 2014, Pearson began using a revised GED exam—a new version of the GED that has to be taken entirely on-line and which employs the new Common Core standards, which emphasize analysis, higher-order thinking, and more advanced math.  It has been reported that passage rates dropped precipitously, in some places by 90 percent.  And for the young adults, the cost of taking the test jumped from $40 to over $100.

Since 2014, a lot of states have switched to tests produced by different companies as a way to measure high school equivalency.  And on Monday of this week, the Austin American-Statesman reported that the Texas State Board of Education is meeting this week “to discuss and possibly choose one or more testing providers for the GED following criticism that the current test is too difficult and expensive.  Vendors will give presentations to the Board Wednesday with hopes of replacing testing giant Pearson Education which rolled out a new GED…exam in 2014.  Since then, about half as many Texans have been signing up to take the test.  Critics have said that the cost, $135 for the test series, is too high for people who need it…. In 2013, which was the last year with the old test, nearly three-quarters of Texas’ 59,000 test-takers passed and earned a high school equivalency certificate.  There were fewer than 20,000 Texans who took the new exam in 2014, and 51.5 percent passed the computer-based version.”

And so it is not surprising that on Tuesday, Pearson and its GED Testing Service put out a press release to announce that the passing score on Pearson’s GED test is being recalibrated: “Today GED Testing Service announces a recalibration to the GED test passing score… In most states the passing score for high school equivalency is moving from 150 to 145… The scoring enhancement is driven by a detailed analysis of educational outcomes of GED program graduates compared to high school graduates over the past 18 months.”

Politico Morning Education explains Pearson’s scoring recalibration more clearly: “The GED Testing Service is changing the minimum passing score for the widely used high school equivalency test.  Students can score five points lower on the test and still pass…. And it’s recommending that states apply the changes retroactively, so students who took the GED anytime after January 2014 and didn’t quite make the cut could now have a passing score.”

The recalibration of the passing cut score in this case means that Pearson was going broke by being really tough with its cut score.  A lot of states had dropped Pearson or were threatening to drop Pearson as the GED provider. So its GED Testing Service has now made the GED easier to pass. Manipulation of the passing level doesn’t really say much at all about the preparation of the people who are paying a lot of money to take GED, but it says a lot about the arbitrary nature of the cut scores on standardized tests.

Advertisements

3 thoughts on “Cut Scores on Standardized Tests Are Arbitrary; Pearson’s GED Recalibration Proves It

  1. “The scoring enhancement is driven by a detailed analysis of educational outcomes of GED program graduates compared to high school graduates over the past 18 months.”
    I guess there are a few people who believe this !!!

  2. I want to scream. And weep.
    We live in a world that can’t resolve its problems w science (psychological, statistical, etc …),obsessed w ‘objectivity’ that does not exist. We dress up our (subjective, political) decisions in ‘research’ that is NOT generalizable, so circumscribed that we can’t really replicate it.
    We then perform meta-analyses that dress up non-generalizable samples and variables that fail equally.

    The fact is that every cut-score is arbitrary. The ‘powers to which we delegate power’ want winners and losers. (Perhaps, because we think it’s a game like sports … not aware of human lives so effected. reduced to a score/number.)
    So, they (politicians, testing companies, etc) find a place with an acceptable # of losers. It is arbitrary! Not objective. There is no objectivity to testing. But the testing obscures the subjectivity w a lot of numbers and complex analyses.

    And i would argue that statistically, based on the error of measurement, that even when a cut score is set .. that a score below it is within the error of measurement. That means: wherever the ‘cut-score’ is set, there is no ‘difference’ between the passing cut-score and the failing cut-score just below it, due to the error of measurement. NO DIFFERENCE.

    What would it mean to tear off the veil of this pseudo-science?
    To unmask the ‘smoke and mirrors’ of (typical) social science (educational research, testing) we accept because we are not experts, we trust the authorities, we ourselves were subjected to all this testing, educated to believe it, and somehow survived?
    Social and educational ‘science,’ so understood, is primarily a rhetoric, not substance, not fact. It is marshalled for purposes, political, functional, etc. That dismantles the appearance of ‘objectivity’ in its uses. Actually reviewing the research in detail that is cited for the purposes so used reveals the profoundly limited domain of relevance of that research.

    my thoughts.
    mike

  3. Pingback: Pearson’s Operation of the GED: Kicking Down the Very People Who Need a Leg Up | janresseger

Leave a Reply

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

WordPress.com Logo

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

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s