Sports, Test Scores, and the Difference Between Science and Journalism

Inside Higher Ed has an article on athletics and admissions based on an investigative report from the Atlanta Journal-Constitution. The report compares the SAT scores of football and basketball players to those of other students, but what it really highlights is the difference between science and journalism.

The basis of the report is pretty simple: the paper got the test score reports for 55 major colleges and universities, from data that they are required to file with the NCAA. They compared the average scores for football and basketball players to the scores of other athletes and students in general, and found all sorts of splashy results:

  • All 53 schools for which football SAT scores were available had at least an 88-point gap between team members’ average score and the average for the student body.
  • Schools with the highest admissions standards, such as Georgia Tech; the University of Virginia; the University of California, Berkeley; UCLA; and the University of North Carolina at Chapel Hill, had the biggest gaps between the SAT averages for athletes and the overall student body.
  • Football players performed 115 points worse on the SAT than male athletes in other sports.
  • The differences between athletes’ and non-athletes’ SAT scores was less than half as big for women (73 points) as for men (170).

Of course, there’s a screamingly obvious objection to their method. I’ve blogged about it before, and it’s even raised in the AJC story, by Georgia Tech basketball coach Paul Hewitt:

But Georgia Tech men’s basketball coach Paul Hewitt says he and other coaches are able to go beyond test scores to find recruits who can succeed in school while also having the talent to play at a high level.

Hewitt says the only fair comparison is between athletes and other students with similar socioeconomic backgrounds. Seen that way, he argues, athletics programs perform very well. Black athletes, for example, graduate at higher rates than black students as a whole.

“To insinuate that athletics has caused this problem of poor graduation rates [among black students] is wrong,” Hewitt said.

And here is where we see the difference between science– even social science– and journalism. A scientist would anticipate this objection, and attempt to address it in the report. For example, you could try to find some data on the socioeconomic backgrounds of the athletes, and attempt to control for those effects by comparing the athletes’ scores to their demographic peers, to see if the gap remained, or narrowed, or even expanded.

A journalist just reports the objection in a half-dozen sentences buried in the middle of the story, and considers his job complete.

That’s the difference between somebody who is attempting to learn something useful about a situation, and somebody who is just trying to sell newspapers.

Don’t get me wrong, here– I’m not just saying this because I’m a college basketball fan, and think this will make college hoops players look better. It probably would narrow the gap, but it’s not hard to imagine plausible ways it could make things look worse, if you did a good job of controlling for both race and class. African-American basketball players from poor economic backgrounds at a place like Georgia Tech might well find themselves being compared to students who are better than average. I imagine there aren’t a whole lot of poor African-American students waltzing into Georgia Tech on family connections or to drum up donations, so it’s conceivable that other students from backgrounds similar to some of the basketball team might need to be really strong to end up there in the first place.

But that’s the thing– we don’t know, because nobody’s got the data. And given that the story doesn’t even offer a token excuse about how that information isn’t available, it doesn’t seem like they’re in a hurry to get it. Without that comparison, though, this whole story is worthless.

Of course, if they had done the actual work– or even the trivial amount of work required to check out Hewitt’s claim about graduation rates (see my old post)– they might need to address the way that low-income and minority students systematically underperform. Which might lead people to wonder what that says about the job we do educating children from poor and urban communities in this country.

And, God knows, we can’t have that.

22 thoughts on “Sports, Test Scores, and the Difference Between Science and Journalism

  1. I believe you are dancing around trying to protect a deeply corrupt system.

    Even this;
    But Georgia Tech men’s basketball coach Paul Hewitt says he and other coaches are able to go beyond test scores to find recruits who can succeed in school while also having the talent to play at a high level.
    suggests that students of lower potential are brought into the system primarily for their sports ability, those similar students without sports ability are ignored or would never qualify.

    It is also interesting that you implicitly equate sports to race, or economics.

    College is not (or should not) be about SPORTS. It’s about intellectual development. Here’s a novel thought: how about selecting students on quantifiable qualifcations (whether test scores or other) without dragging the totally irrelevant factor of physical sports skill into the equation

    Additionally college should be about accomplishment, not babysitting for students that are ill prepared.

  2. Having participated in college athletics and understanding the efforts put together by universities through their Student-Athlete Academic support programs, I actually have to say that even when accounting for socio-economic factors the graduation rate are pretty appalling. Student athletes in the major sports get free one-on-one tutoring through these programs for any and all classes they need help in and are required to use it if they at any time are below a threshold GPA. Study hall is mandatory when an individuals GPA takes a nose dive. The general college population does not have the benefit of these programs yet still manage to graduate at higher rates. While one might argue that these services are necessary due to the schedules of the student-athlete I am sure that many non-athletes also have jobs, some might have families or other obligations (with this being probably more true for those in a similar socio-economic background) and therefore face equal disadvantages yet they still manage to graduate a higher rate. Wit

  3. Why should college not be about sports? For the university, big name sports programs are a profitable side business and help drive future revenue from alumni. For student spectators, these programs provide something fun to do — particularly important for certain schools where there is literally nothing else to do in town on a weekend. And the pro leagues get a cheap farm system. Everybody wins.

  4. Academic achievement requires intelligence, more or less. Athletic achievement does not – but try doing it without somatic hypertrophy. Your car will not display powered flight, nor should it. A linesman need not contemplate Plato’s cave (a profoundly awful silopsistic sump – less two hours during the final). KILL BUBBA KILL!

    90% of the population is not in the top 10% of intelligence. Get over it, then redirect purchase of the future toward the few Gifted minds that can achieve it.
    depp=true

  5. I am not sure what is the point of the objection. If you are trying to demonstrate the failure of the student-athlete model of education, this is relevant and sufficient data. For that purpose it does not matter precisely *why* the test scores of student-athletes are significantly lower. I agree that for other purposes this data is insufficient, and also that the questions of class and race are much more significant, neither one of those points makes the data worthless.

    (and on the question of class and race, think about what you could do with only a small fraction of the money paid to professional coaches and other auxiliary personnel, if college sports was limited and was not trying to compete with professional sports for TV rights).

  6. I am not sure what is the point of the objection. If you are trying to demonstrate the failure of the student-athlete model of education, this is relevant and sufficient data.

    I disagree. If you want to know whether the student-athlete model is a failure, you need to look at how the student-athletes compare to other, comparable students, or you run the risk of measuring nothing but confounding factors.

    This is incredibly bad practice. You would never run a physics experiment without a control, and you should never analyze social-science data without trying to control for socioeconomic factors. Particularly in issues associate with education– comparing the test scores of students from poor families in lousy school districts to those of wealthy suburbanites with money for test-prep classes tells you nothing about the abilities of the students themselves.

    I get really irritated about this because it’s so stupidly obvious what the comparison ought to be, and yet it’s never done properly. On top of that, this is sort of a perfect storm of irritating commentary, from racists like “Uncle Al” on the right, and stuffy academics on the left who disapprove of any time students spend outside the classroom or the library.

    If I were starting from scratch, I wouldn’t design anything like the unholy mess that is college athletics. But we’re stuck with the system that has evolved, and if we’re going to try to improve it, we need to start by measuring it properly. What we have here isn’t anywhere close to a legitimate attempt to measure it.

  7. I’m a journalist (and a scientist) and Chad is obviously correct to call shenanigans on this. The author toyed around with raw data without any controls. Furthermore, it is completely irrelevant that college sports may or may not be corrupt. That’s injecting morality into the debate and should be considered a diversion.

  8. “For the university, big name sports programs are a profitable side business ”

    It is a rare school that makes money on its athletic program – many of the big ones who go to bowl games lose money on the venture.

    I’m not a fan of university athletics – they are too often merely vanity plates on the name of the school. However, given that they exist, I’m not sure why there should be a reason to dumb down entrance requirements and academic standards for the athletes – if they are in the school, they should have to perform as the other students.

  9. “It is a rare school that makes money on its athletic program – many of the big ones who go to bowl games lose money on the venture.”

    Does that take into account donations from alumni?

  10. Well, the fact that is being demonstrated here is that student-athletes would not be admitted to the university without being uniquely useful to its economic mission. We can argue whether this fact can be understood better and put in a wider context, but I think this data demonstrates this fact very clearly.

    As for the unholy mess, I think it could be controlled fairly easily by limiting the amount of money flowing in college sports, and subjecting it to rational cost-benefit analysis. I am not sure about all stuffy academics, but I for one would not object to having amateur sports teams on campus. On the other hand, I would not feel comfortable functioning in an academic environment that is viewed by its administrators as being an appendage to a professional sports team.

  11. I understand there is insufficient data for a comprehensive report on the effects of race and class and gender, but may we safely say that [college] sports is an intellectual plague on male society? Why do we care so much about sports that we have to get into hypothetical arguments about various social controls? Isn’t any disparity in scores a cause for alarm and not a time for excuses?

  12. Well, the fact that is being demonstrated here is that student-athletes would not be admitted to the university without being uniquely useful to its economic mission. We can argue whether this fact can be understood better and put in a wider context, but I think this data demonstrates this fact very clearly.

    How can you conclude that without doing the check that Chad suggests? Maybe the university has generally lower admissions standards for students who are socioeconomically in worse shape.

  13. Sorry, the data has to do with admission criteria, the question how the athletes do while in school is entirely different. The data demonstrates clearly that admission criteria for athletes are lower than the admission criteria for students who cannot be used to generate income. I do not see how any other set of data can negate that fact. It may well be that student-athletes go on to perform as well as others from the same racial and economic background, I do not see why this is relevant.

  14. When I arrived at Caltech at age sixteen and tried out for various sports, my hand grip was stronger than any freshman NOT on the football team. So I tried pole vaulting and javelin, not well enough, and resigned myself to managing the soccer and fencing teams.

    At the time they joked that our football team was the only one of any college in America where the IQs were higher than the weights.

    And then I broke both wrists in a freak motorcycle accident in Pasadena, and my sports career became more limited.

    But, strangely enough, there was a long history of Caltech football, including a season in 1944 when the team was not only undefeated, by the likes of USC and UCLA, but unscored upon. This has never happened since in the NCAA.

    There’s a simple explanation. In 1944, Navy men swelled the ranks of the student body as part of the search for an atomic bomb.

    For that matter (see the fine documentary film “Quantum Hoops”) the 2006 basketball team had more high school valedictorians than varsity players.

    Also, although extreme physical rigor is demanded, along with academic excellence, it’s not an Olympic sport yet, though India and China have been hedging their bets, but the last man to walk on the moon is a Caltech alum — “a streak that remains to this day.”

  15. Sorry, the data has to do with admission criteria, the question how the athletes do while in school is entirely different. The data demonstrates clearly that admission criteria for athletes are lower than the admission criteria for students who cannot be used to generate income.

    Do they have worse test scores than students from the same socioeconomic background? That’s the relevant comparison, here, and it hasn’t been done.

    It’s conceivable that athletes are drawn disproportionately from groups that, for one reason or another, have lower test scores as a general matter. In which case what we’re seeing is less an athletic preference than a demographic preference. Or, perhaps, the athletic preference is outweighing other factors that would tend to lead to students from those groups not being admitted.

    Is it really likely? Probably not. But without data, it’s entirely possible.

  16. I am still not quite understanding: then thesis is that athlete’s SAT scores is representative of their racial and economic peers among high school students, or among students admitted to the university? it was my impression that it is the former. In other words if you are from an under-privileged background, your only way in is to be of economic value to the privileged class.

    In any event, as you note, the idea that the students in question are admitted due to some demographic preference, mysteriously in effect only in those progressive division one schools, is ludicrous. We can safely have a discussion about the real world without eliminating absolutely all logical possibilities first. Otherwise, it would look suspiciously like nitpicking your way out of an unpleasant fact.

  17. I wouldn’t go as far as “ludicrous.” I would say that the chance that this is the only factor leading to the score gap is very low, but I do think it’s a major contributing factor. I suspect that the gap would shrink dramatically if you compared athletes at major public universities with their demographic peers at the same universities, in the same way that the graduation rate gaps cited in the older post do.

    I suspect we’re in violent agreement about the core issue, here– that the race and class disparities in our pre-college educational system are shameful. We differ only in what we think that means for the ritual denunciation of college athletics.

    To some degree, I’m bothered by this because the ostentatious hand-wringing about the state of college athletics is pre-empting a conversation about the larger issue. And worse yet, in some cases, providing rhetorical cover for outright racists who would be happier if the groups in question weren’t present at all.

  18. We are probably in agreement on the core issue. I think there is absolutely a class problem in major American universities as I experienced them. I think college sports is one way to mask this problem, by artificially enhancing the numbers admitted from the under-privileged class, while not providing them with anything remotely resembling a useful education.

  19. Do they have worse test scores than students from the same socioeconomic background? That’s the relevant comparison, here, and it hasn’t been done.

    That is NOT relevant. That is a red herring. Why should the ‘socio-economic background’ (however defined) be different for sports players than non sports players (and why are you assuming it is different– is not that assumption racist at some level?)

    If ‘socio-economic background’ is systematically different for sports players than general student body, then there seems to be a significant problem. Students are being brought into the school primarily because of their sports ability rather than academic qualifications. Even if you accept the weak argument that the coaches look for disadvantaged students with hidden potential (yeah, right), that says that disadvantaged students with hidden potential who are not sports stars are being ignored.

  20. James Smith Bush was Yale class of 1844. I don’t know his sports activities while matriculating.

    Then Robert E. Sheldon Jr., Yale class of 1904, again I don’t know sports involvement.

    Then the grandson of James Smith Bush and nephew of Robert E. Sheldon Jr., namely Prescott Sheldon Bush, a native of Columbus, Ohio, entered Yale University in 1913, after five years in St. George’s Episcopal preparatory school in Newport, Rhode Island. Besides playing the Great Game with classmates Percy A. Rockefeller, Averell “Bunny” Harriman (class of 1913), et al., as Wall Street executive banker, WW I munitions controller of contractors, and United States Senator representing Connecticut, he played varsity golf, football, and baseball, and was president of the Yale Glee Club.

    His son, George Herbert Walker Bush, before serving as forty-first President of the United States, attended Phillips Academy in Andover, Massachusetts, where he was president of the senior class and secretary of the student council, president of the community fund-raising group, a member of the editorial board of the school newspaper, and captain of both the varsity baseball and soccer teams. George Herbert Walker Bush had been accepted to Yale University, then decided to fight in World War II instead of going to college, but matriculated after his discharge and marriage, and was enrolled in an accelerated program that allowed him to graduate in two and a half years, rather than four. He was a member of the Delta Kappa Epsilon fraternity and was elected president, and also, as was his father, in Skull and Bones. He also captained the Yale baseball team, and as a left-handed first baseman, played in the first two College World Series. As the team captain, Bush met Babe Ruth before a game during his senior year.

    George W. Bush, however, had a rather less distinguished prep school and college athletic career (i.e. cheerleader) before being a quite profitable Chairman of the Board of the Texas Rangers baseball team. He wanted to be Commissioner of Baseball. There are wonderful alternate worlds where his wish was granted, and most of the world was a much better place as a result.

    I don’t know about SAT scores for these members of the Bush family. My point is that when the college athleticism declined, academic and real world skills also plumitted.

    There is a traditional connection in the angloamerican empire between school sports and academics and leadership (i.e. wars won On the Playing Fields of Eton, the Rhodes Scholarship, and so forth).

    Hence this is a traditionally important linkage, whatever the correlation coefficients may be.

  21. If Hewitt had a clue, which he probably doesn’t, he would compare the success rates of athletes to those of other categories of students that are admitted based on “other” factors. A common example is a major in music or some other fine or performing art, where an audition plays a much greater role in admission than SAT scores. I recall discussing admission “exceptions” with a top admin at one institution and was told that there were more (total, not by percentage of students) issued in academic areas than in sports. An interesting measure might be to look at grad rates of students admitted specifically as “exceptions”, both athletes and others. After all, not all athletes are in the special admit category. It could be that ALL of the sub-par athletes fail to graduate and most of the sub-par musicians do graduate. I’m even willing to consider leaving early to go pro as “graduation” provided the player actually makes the playing roster of an NFL or NBA team.

    It is interesting to compare black athletes and non-athletes, although it would be a better comparison if the non-athletes got the same level of academic support.

    You don’t usually find a typical student getting a free personal tutor just because they are working 20 hours (or more) per week to pay for school. What would the grad rate of the non-athletes look like if they got that kind of support?

  22. “It is a rare school that makes money on its athletic program – many of the big ones who go to bowl games lose money on the venture.”

    Does that take into account donations from alumni?

    Yes, but it might not count involuntary transfers from student fees. That is where they often make up the difference, because they can’t use state funds.

    There was an extensive study of the finances of the public universities (all of them) in the state of Michigan done by the Detroit Free Press or News some years back. There are schools that lose money on major college football.

Comments are closed.