College Board: "Adversity" Concept Was "Wrong"


I like it when someone has the class to admit that they were wrong about something. In the case of the so-called “adversity score" rolled out with great fanfare this past spring by the College Board, the majority of reactions were immediate and forceful that it was a bad idea. The College Board must have been listening, as two prominent sources report:

The Washington Post reports: College Board retreats on plan for single 'adversity' rating to go with SAT score.

The College Board is dropping a controversial plan to send colleges a single numeric rating of the adversity students faced in their communities as they took an SAT admission test, opting instead to provide separate measures to describe their high schools and neighborhoods.

Last spring, the revealing of plans to develop an “overall disadvantage level" for each SAT taker, on a scale of 1 to 100, prompted an uproar. Many dubbed it an “adversity score." Critics said it would be vulnerable to manipulation and could unfairly taint how an actual SAT score is perceived. The test itself gauges math and evidence-based reading and writing, with a widely recognized maximum score of 1600. That scoring scale remains the same

The CB's 180 on this may have had some people thinking that the whole initiative was a trial balloon to see which way the wind was blowing. Or, it could have been just inadequately explained when revealed. Whatever the reason for the blowback, CB wasted no time drawing up a new game plan to counter an idea the College Board's chief executive now says was a mistake.

Amid growing scrutiny of the role wealth plays in college admissions and in hopes of appearing "woke" and "politically correct", two years ago the College Board introduced its Environmental Context Dashboard to provide context for a student's performance on the test and help schools identify those who have done more with less.

In short, it was the SAT's way to implement affirmative action in test results. Skeptics said it was a terrible idea and it now appears they were right.

The version used by about 50 institutions in a pilot program involved a formula that combined school and neighborhood factors like advanced course offerings and the crime rate to produce a single number. But critics called it an overreach for the College Board to score adversity the way it does academics. David Coleman, CEO of the College Board said some also wrongly worried the tool would alter the SAT results.

“The idea of a single score was wrong," he said, quoted by the AP. "It was confusing and created the misperception that the indicators are specific to an individual student." ...

I'm wondering how carefully CB tested this concept before deploying it. In my view, the “idea of a single score" is obviously not precise enough to capture the complex cache of factors that comprise socioeconomic hardship. I'm also wondering which aspect of overall resistance to the idea tilted the scales toward retreat. We'll never know, I'm sure, but it would be interesting to get a peek into CB's marketing research reports to discover the tipping point.

Survey Results Echo Public Sentiment

I received an email yesterday from Russell Schaffer, Senior Communications Manager for Kaplan Test Prep. He noted that Kaplan is in the middle of conducting its annual survey of college admissions officers, something the firm does every year to provide its students and parents with the most up-to-date and accurate information on the college admissions process. Several of the questions Kaplan is asking are based on the adversity score issue.

Here's a sampling of those questions and the response results, as provided to me by Russell:

How strongly do you and your institution support or oppose this new "adversity score," as many are calling it?*

- Strongly support: 14%

- Somewhat support: 24%

- Somewhat oppose : 4%

- Strongly oppose: 2%

- Don't know: 56%

*295 admissions officers from the nation's top national, regional and liberal arts colleges and universities – as compiled from U.S. News & World Report – were polled by telephone between July and August 2019.

Based upon what you know today do you plan on using the adversity score to help you make admissions decisions?*

- Definitely yes: 3%

- Probably yes: 15%

- Probably not: 17%

- Definitely not: 13%

- Don't know/Too soon to tell: 52%

*259 admissions officers from the nation's top national, regional and liberal arts colleges and universities – as compiled from U.S. News & World Report – were polled by telephone between July and August 2019.

The College Board currently says that only colleges will see test takers' adversity scores, although they say they may decide to share the score with test takers too. How strongly do you support or oppose the College Board's current plan to not share this score with test takers?*

- Strongly support: 6%

- Somewhat support: 14%

- Somewhat oppose : 18%

- Strongly oppose: 13%

- Don't know: 49%

*264 admissions officers from the nation's top national, regional and liberal arts colleges and universities – as compiled from U.S. News & World Report – were polled by telephone between July and August 2019.

Course Reverse Not Surprising

It's not hard, then, to understand why College Board reversed course. At least according to these survey results, admissions officers were less than immediately enthusiastic about the idea, even after CB's enthusiastic, high-profile unveiling. As for CB's rewrite, Russell Schaffer notes, “These changes seem to make the process more transparent, which should be good for students, their parents, and colleges. And in the midst of the Varsity Blues scandal, this is really important."

What changes has College Board made? Here's the scoop:

The New York City-based College Board announced several changes to the tool Tuesday, including the decision to give students access to the information about their schools and neighborhood starting in the 2020-2021 school year.

Renamed “Landscape," the revised tool will provide data points from government sources and the College Board that are seen as affecting education. They include whether the student's school is rural, suburban or urban, the size of the school's senior class, the percentage of students eligible for free- and reduced-price lunch, and participation and performance in college-level Advanced Placement courses at the school. Between 100 and 150 institutions are expected to pilot the new tool this year before it becomes broadly available next year.

Admissions officers also will see a range of test scores at the school to show where the applicant's falls, as well as information like the median family income, education levels and crime rates in the student's neighborhood.

The imperfect tool's creation was an acknowledgment of persistent criticism of the use of admissions tests in an era of concern with unequal access to advanced coursework and high-priced tutors that further advantage those with the means to access them.

Changing the “adversity score" concept to a “Landscape" approach reminds me of something Mad Men's Don Draper once said: “If you don't like what's being said, change the conversation." In College Board's situation, they didn't like what they were hearing about their adversity score idea, so they changed the conversation to Landscape.

Wrapping it up:

Will the new version be better? Eddie Comeaux, vice chair of the University of California's Board of Admissions and Relations with Schools, said the improved clarity and transparency of Landscape addressed some of his initial concerns.

He still worries, though, about potential implicit bias among admissions officers, a problem that predates the context tool and is a focus of his work on the board, which regulates admissions practices.

“We want to look at implicit bias training and the ways in which certain indicators might signal a way in which (application) readers advantage or disadvantage certain applications and consciously or unconsciously not be aware of it," he said.

“I'm less concerned about Landscape," he said, “than I am about those who are making the decisions utilizing Landscape.

Because there is always "something" to "explain" away why someone doesn't get a perfect score on the SAT, and heaven forbid it has something to do with one's own personal qualities, ambitions and laziness.

As with most new approaches to solving ongoing complexities, time will tell whether or not Landscape will be the solution College Board is looking for. Stay tuned.