Your Argument Isn't, Cannot, and Will Never Be "Right"
SATs and the social origins of flawed reasoning
In the spring of 2018, I wrote an op-ed for my college newspaper, The Daily Californian, in which I argued against UC Berkeley’s continued consideration of standardized tests like the SAT and ACT in its admissions decisions. My argument essentially went like this: since Black, Latino, and poor students perform categorically worse on the SAT than white students from families in higher tax brackets, UC Berkeley (featuring a student body underrepresented by Black, Latino, and poor students relative to the state population) should ignore SAT scores in admissions decisions, therefore giving prospective students a fairer chance of admission.
If you’ve spent any time studying the testing issue, these points won’t surprise you—this was a pretty mainstream liberal stance at the time.1 As such, I felt really good about the article after its publication; I received wide praise from family and friends, and got a ton of positive feedback from relative strangers online. I even puffed my chest out a little when, in 2020, the University of California announced it was permanently phasing out its standardized testing requirements.
Today, I consider my SAT op-ed to be among the most regrettable things I’ve publicly written. I’d like to explain why.
My article successfully presents data to demonstrate the relationship between test scores, race, and income, and I think it frames the issue pretty well; going to college is slightly overrated, but it’s still a big deal, especially for students from underrepresented backgrounds.
Beyond that, though, the argumentative foundation of the piece crumbles on a number of fronts.
First, it fails to meaningfully consider the consequences revoking the test from admissions decisions would bring. “De-emphasizing test scores and investing resources into studying applicant essays, transcripts and an increased focus on teacher recommendations is the first step toward more equitable admissions decisions,” I write. If I had examined any of those alternatives more closely, I would have discovered a massive hole in my argument: all of them, including GPA, are at least as impacted by systemic inequality as SAT scores are. Education writer Freddie DeBoer has equated blaming testing for inequitable admissions practices to faulting a thermometer for climate change; test scores are merely indicators, not drivers, of deeper, society-wide inequalities which the SAT simply measures, not exacerbates.2
Neglecting this point was a product of a more structural reasoning flaw endemic to the piece. Instead of starting my research from scratch, digging deeply for all available data points, and proceeding with an argument, I took the supposition that the SAT is harmful as an unquestioned truth and worked backwards to gather evidence in support of a conclusion I had already reached. While my article demonstrates that SAT scores are related to race and income, it fails to establish why the relationship between scores and demographics exists, and, further, why the reason it exists may (or may not) be unfair.
Why did I do this? I can think of a few interrelated reasons.
First, as I alluded to above, I was universally praised within my social network for publishing this opinion. That didn’t exactly come as a surprise to me; as I said, this was a mainstream progressive viewpoint which a number of popular outlets (The Daily Beast, Huffington Post, Brookings Institution) had already covered extensively. That doesn’t mean I shouldn’t have written the story—as a then-college student at a public school uniquely suited to impact wider policy decisions, I had relevant perspective to add—but it does mean that I knew I wasn’t making many points which readers within my social circle were likely to disagree with. A desire to be praised by my peers undoubtedly incentivized me to take the angle I did.
Equally influential as the carrot of praise was the stick of potential condemnation, particularly in the university environment I occupied.
More specifically, I was definitely afraid of being called racist in college. I just was. This was true despite my certainty that any suggestion I might have made on the testing issue would have been rooted in a good-faith, sincere effort to aid low-income, minority communities. At Berkeley, though, there was very little safe rhetorical wiggle room outside the viewpoints of the typical identity politics liberal and the insufferable right wing section guy. Even if I had been interested in filling the “lefty white guy who wants to uplift the working class and minorities as much as you do but has a very slightly different strategy for how to do it” gap in my classes, I wasn’t at all confident enough in my own eloquence to navigate such treacherous linguistic terrain effectively enough to not piss people off.
That was especially true whilst handling topics like standardized testing and college admissions, both extremely complex, personal issues which directly impacted every student on campus. Being called out, in any form, for expressing a more nuanced opinion on the testing issue would have had social ramifications unpleasant enough to me to discourage a more thorough investigation. The value proposition of seeking and expressing such views wasn’t worth it to me in the face of the potential social fallout.
It was far easier and safer, both in person and in writing, to demonstrate agreement with the ends (upon which, realistically, most Berkeley students were roughly aligned) without touching the finer points of the means. In the case of the SAT article, that meant not just ignoring but neglecting to search for evidence which might have even slightly disrupted the consensus progressive notion that banishing standardized testing (the means) would automatically lead to better outcomes for low-income Black and Hispanic students (the ends).3
Look, nothing actually harmful would have happened to me had I more thoroughly examined this issue and published a slightly consensus-defying analysis; very few people were paying attention to the student paper, and even if some were no one was chasing me with a club or expelling me from a student org because they disagreed with what I wrote in the Daily Cal. I know I didn’t write the SAT article, or any other work, solely to boost my campus credibility or to avoid being cancelled or whatever—I had higher standards than that (although this was definitely something other students did).
But the thought of being called racist in itself, even without any further “real life” fallout, was ominous enough to shape my reasoning process. Just as I am certain that winning the favor of my peers incentivized me to write with a certain slant, the fear of even a mildly unpleasant reaction disincentivized me from digging into the facts in this instance as deeply as I should have.
Ultimately, I failed to examine the SAT issue critically because I took for granted that no counterpoint was worthy of critical examination. Neither I nor my editor ever mentioned inserting more counter-evidence into the piece because I am certain that we both, because of the intellectual and social environments we inhabited, believed my argument was “right.”
Which is, of course, nonsensical. The very idea of an argument is predicated on the notion that it is being argued—it is up for debate. The “right” and true stuff, the facts, are the building blocks of an argument, the ammo used to make a point. Then, if there’s enough support behind an argument, it might be tested in real life—the widespread adoption of test-free and optional admissions practices at American colleges in recent years is an example of an argument put into practice. We’ll see how it goes! The only guaranteed outcome is that the result, decisive as it may appear, will spawn an entirely new genus of arguments, all of which will compete to out-convince one another until the end of time. Those arguments might be convincing, flawed, thorough, narrow in scope, preposterous, satisfying. But they will never be right.
Whilst at Berkeley, the only individuals I knew about who pushed back against my thesis were commentators on the Daily Cal website; I dismissed them then as loony alt-righters looking to muck things up on a college message board (in fairness, the comments suggest at least a couple of them probably were).4 I failed to consider, though, not just their extreme points of view but any other points of view, in my work—even those sincerely rooted in a desire to reach the same goal (fairer outcomes for low-income students) by prescribing a different strategy. I took the “correctness” of my argument as fact, then blindly poached evidence to match my presumption, oblivious to any details which might have undermined a dogmatic thesis.
In the end, I think SAT advocates like DeBoer (who supports a return to mandatory testing) probably take their point slightly too far. I don’t think the relationship between the SAT and societal inequality is quite analogous to that of a thermometer and climate change; a thermometer is a more useful tool to measure temperature than the SAT is to diagnose inequality, and I’m not sure the relatively minor benefit of accruing additional scoring data justifies continually putting kids through the meat grinder of year-round test prep cycles.5
My opinion on the matter isn’t entirely fixed, and, anyways, I have no interest in rehashing the whole SAT debate here. It’s entirely reasonable that, having weighed all available evidence, one might still reach the independent conclusion that the test should be banned; popular views are by no means necessarily poorly reasoned pitfalls of groupthink. Nor do I intend to criticize anyone for holding an especially resolute and disciplined opinion on this or any similar subject; passionate argument rooted in reason is an essential feature of any dynamic, problem-solving environment. I simply mean to demonstrate that to meaningfully reason, write, think, one must acknowledge the presence of an argument at all.
I sense it still is.
There are even some cases, as DeBoer has noted, where grinding for the SAT might actually serve as the only admissions lifeline for an otherwise disadvantaged student whose family can’t afford, say, resume-boosting piano lessons or a Habitat for Humanity trip between sophomore and junior year.
My article entirely ignores data from Asian students (and its implications).
I only discovered DeBoer’s work after I graduated.
A test-optional system giving disadvantaged students the opportunity to all-out study for a high score while still allowing anyone to opt out of a draining adolescent pastime seems most logical to me.