Interpreting the ‘new’ PSAT/NMSQT® score reports

Students lucky enough to navigate the College Board website and successfully retrieve PSAT/NMSQ test scores from October are probably wondering what to make of the dizzying number of scores. As part of the redesign and repackaging of SAT-related products, the College Board has spun off no less than 16 separate scores to consider and stress over.  Among these are a total score, a math score, an evidence-based reading and writing (ERW) score, “Nationally Representative Sample Percentile(s),” three test scores, two cross-test scores, seven subtest scores, and a National Merit® Scholarship Corporation Selection Index. 

And each of these sets of scores has a different score range. The total score ranges from 320 to 1520; math and ERW scores range from 160 to 760; test scores and cross-test scores range from 8 to 38; subscores range from 1 to 15; and the NMSC Selection Index ranges from 48 to 228.

Families anxious to make decisions about future testing and those curious about how the new NMSC Selection Index might translate into future merit scholarship designations are naturally asking which scores are most important and what exactly they mean.

“Unfortunately, these overhauled reports feel overwrought and confusing. While they offer a vast array of measures—some helpful, some less so—and newly created subscores, they fall short of providing clear takeaways most students are seeking,” explained Bruce Reed, of Compass Education Group.  “Aside from strongly encouraging students to now practice on Khan Academy, the new PSAT reports are not as obviously actionable as users need them to be.”

So what can students take away from their PSAT experience?

First, be assured that no college will ever see these test results. They are for your use only. So relax. 

Next, be careful about assuming too much about the relationship between the new PSAT scores and the old PSAT or SAT. For a number of different reasons, the College Board is making the new PSAT scores look like SAT scores by using three-digits instead of two.

According to Bruce Reed, “While the two new tests share a common scale, they occupy slightly different ranges on that single scale.”

And that’s where confusion sets in.  SAT will continue to theoretically span the 200-800 range, while PSAT is staggered down to 160 to 760, to reflect the test’s slightly lower difficulty level. It’s not that some test-takers are incapable of exceeding a 760, it’s just that the PSAT doesn’t claim to verify that.

But, the shifting down of the scale may lead to the wrong assumption that the scores themselves must also shift down.  In fact, the exact opposite is more likely to be true.

“Interestingly (and conveniently for the College Board), nearly all PSAT scores are ‘up’ this year,” explains Reed. “But this has nothing to do with a jump in overall student achievement. It doesn’t necessarily mean scores are ‘better’ than they were last year, and it certainly doesn’t mean they are ‘better’ than ACT scores despite their appearance.”

Reed describes the phenomenon in terms of a “statistical tailwind,” which is occurring in the order of roughly 20 to 30 points per section for most students. This has occurred partly because the test is new and partly because of the elimination of the “guessing penalty.”  Think in terms of the effect random guessing might have on the very lowest end of the scale.

Next, despite suggestions to the contrary, the scores in and of themselves do not provide clear guides as to which test to take—ACT or SAT—to fulfill college entrance requirements. Out of concern for already-announced scoring issues and timelines as well as uneasiness related to the general unknowns of the new SAT, most counselors and test prep experts are still suggesting students take the ACT.  Since the first scores from the new SAT won’t be available until late May, the most useful guide—a concordance table comparing scores from the two tests—won’t be published until after the first two administrations of the new SAT. So that won’t help. 

The College Board has also prepared a series of concordance tables comparing the 2014 PSAT with the 2015 PSAT.  While the comparisons may be interesting, these tables are not without their limits, as they provide no direct comparison to the ACT.

To approximate a comparison and provide a general guide, Compass Prep analyzed all available research and used internal data to create a table comparing 2015 PSAT and ACT scores. While not a perfect predictor of success on either ACT or new SAT, the table does provide a sense of which students might do better on which test. Note that while most students will likely find their PSAT and ACT scores intersect somewhere in the purple “Judgment Call” band, some may discover that one test is actually better suited to them.

Finally, in absence of comparable test scales and relevant concordance tables, students may derive some information as to their performance from “percentiles.”  Keep in mind, however, percentile scores from the PSAT cannot be directly compared to ACT percentiles, so their usefulness is limited.

“The College Board now provides two types of percentiles: National Representative Sample and Test User. Without going into details, the National Sample represents a form of percentile-inflation,” said Reed. “The former will always be higher than the latter, but it’s the latter, which students can only access online and after considerable effort, that should be used when comparing to actual test-takers.”

Again, percentiles simply provide a fast glimpse into how well a student did relative to other PSAT test-takers and are limited in terms of how well the project success on other tests or the likelihood of becoming a National Merit Scholarship semifinalist.

These are difficult issues to sort out—even for the experts. And so far, the College Board has driven the conversation.  Many thanks to Bruce Reed, of Compass Education Group, for cutting through some of the College Board marketing information and putting the new PSAT score reports into perspective. 

Monday:  Will my child be a National Merit Scholarship finalist?