ACT announces changes in reports to colleges that could drive admissions decisions

About a year ago, the ACT organization announced what appeared to be subtle changes for 2015-16, mainly in the Writing section of the test.

In a carefully worded news release, ACT described changes starting with the September 2015 test as “designed to improve readiness and help students plan for the future in areas important to success after high school.”  In general, ACT proposed to tweak the optional Writing Test in small—possibly unnoticeable—ways.

And while the 1-to-36 scale would remain the same, ACT indicated students would also be evaluated in new areas of writing competency, including ideas and analysis, development and support, organization, and language use.

But it wasn’t until ACT recently announced changes in reporting documents provided to both students and colleges that the full story came clearer. 

In draft versions of score reports planned for schools and students, it’s evident that ACT not only wants to provide information on student test performance in five core sections of the test (including the optional Writing section), but also wants to chop and dice it into a series of 11 sub- or “domain” scores, including everything from “rhetorical skills” to “ideas and analysis,” all of which scored on a scale of 2 to 12.

In addition, ACT will generate two new hybrid scores in English Language Arts (ELA) and Science, Technology, Engineering, and Mathematics (STEM), based on various combinations of English, Reading and Writing or Science and Math scores.  The new report will also provide terse one- or two-word written assessments of “understanding complex texts.” 

And for colleges correlating career readiness with retention and completion, there’s a little bronze-to-gold level rating “certifying” skills critical to future education and career success.

But of somewhat greater concern are assessments provided to approximately 450 institutional participants in ACT Research Services of “Overall GPA Chances of Success” in various general categories of majors including education, business administration, liberal arts, and engineering, as well as “Specific Course Chances of Success” in broad areas such as freshman English, college algebra, history, chemistry, psychology etc.

Chances of success are made in terms of those students likely to receive a “B” or better in these areas or those students likely to receive a “C” or better.  And they are nowhere to be found on the ACT report provided to students and families.

According to information provided by ACT, chances of success are calculated by using

•  Student-reported information gathered as part of the registration process, including high school GPA and specific course grades earned
• Performance on the student’s ACT test
• Data provided by participating colleges/universities about the previous year’s enrolled students including the college grade average and course grades achieved by first-year students

During registration, students are asked to voluntarily report grades in core academic courses.  These grades are converted by ACT to an unweighted GPA on a 4.0 scale. None of this data comes from the high school and there is no obvious mechanism for verifying its accuracy, although students are clearly warned, “The information you give may be verified by college personnel.”

While these kinds of assessments aren’t exactly new, the intense interest in marketing chances of success to unnamed colleges purchasing a service that estimates student potential based on information reported by the test-taker computed together with scores and historical data provided by the institution is troublesome.

In other words, through the college score report forms, ACT effectively gets more actively involved in the college admissions process by projecting for admissions readers how likely it is that an applicant would not only succeed at their institution but also in their chosen field of study.

With the score report in hand, an admissions officer noting that a particular applicant has indicated a desire to major in business on their application will be able to see how likely ACT thinks it will be that the student will actually succeed in his or her major.  And admissions could choose to admit, deny or recommend another major based on this speculation.

But students are left completely in the dark, as nothing appears on documents they receive that would reveal what ACT is suggesting about their chances of success at a specific institution.  Chances of success do not appear on the ACT Student Score Report because, according to ACT, “the college owns the information” and “by sending their test scores to a college, students give the college permission to use the data as they see fit.”

Yes, nearly all colleges already have enrollment management software that does something similar.  But ACT should be in the business of writing and administering tests—not getting in the middle of college admissions decisions.

It’s one thing for standardized tests to be important factors in admissions, but now ACT proposes to pass judgment on chances of success in ways that are patently unfair to individual students. And these kinds of projections have no place in reports forwarded to colleges unless they are also provided to the person who paid for the test—the test-taker. 

ACT disagrees. In an email provided in response to a series of questions concerning the new reports, ACT says,

“The recent enhancements we have made to the ACT Student Score Report reveal more data than we’ve provided before.  The chance of success belongs to the institution, but if they provide permission, we will share that data on the ACT High School Score Report and in turn, a counselor or administrator may share it with the student.”

When registering for both the SAT and the ACT, students are asked to provide a good deal of demographic information including everything from zip code to grades in individual classes.  Assuming that GPA information is factored into “chances of success”  along with scores, it may make sense for some students to simply leave this information blank and thereby avoid the possibility of being labeled as a potential failure before even being given a chance to succeed.

In the era of big data, these kinds of intrusions and assessments are bound to become increasingly problematic.  A partial solution is to be aware of what’s going on and play the game accordingly.