<p>
[quote]
At Bucknell, the inaccurate data resulted from the college leaving some students' scores out of test averages. In a few cases, the omitted students had scores higher than those reported. But most of the excluded students had lower scores, so the result of leaving them out was to inflate Bucknell's averages. "[D]uring each of those seven years, the scores of 13 to 47 students were omitted from the SAT calculation, with the result being that our mean scores were reported to be 7 to 25 points higher than they actually were on the 1600-point scale," said a letter sent to the campus from John C. Bravman, the president. "During those seven years of misreported data, on average 32 students per year were omitted from the reports and our mean SAT scores were on average reported to be 16 points higher than they actually were."</p>
<p>To inform individual investors, publicly-listed companies report earnings quarterly, and their annual reports are audited. The Wikipedia article on Sarbanes-Oxley states that “as a result of SOX, top management must now individually certify the accuracy of financial information.”</p>
<p>Many students and their parents are investing far more in a college education than they will in any single company. They are spending not only their own money but taxpayer money (Pell grants). Maybe university presidents should be required to affirm the accuracy of information their schools provide, and individuals who lie should face jail time, just as dishonest CEOs do.</p>
Does the mean score matter for US News? Isn’t what they report in the rankings the 25% and 75%iles? How are those affected? Or do they base the ranking on the mean?</p>
<p>Just curious.</p>
<p>Likely there’s some relationship, just can’t be sure what it is.</p>
<p>The next sentence [of the quoted piece] after the SAT numbers refers to ACT scores actually being higher than what was reported:</p>
<p>“Meanwhile, for several of the years in which errant SAT data were reported, the University reported ACT scores for the entering classes that were actually one point lower than the correct figures.”</p>
<p>While I am disappointed that this occurred, the enrollment manager is no longer there and I am pleased with how Bravman is handling the situation.</p>
<p>^^^ @surfcity -
I’m not worried about it. My question is generic, not about Bucknell or any other school. I am actually just wondering why these articles often talk about the “average” SAT scores. I wasn’t aware schools even reported the averages. Frankly, it’s probably a less meaningful metric than the percentiles.</p>
<p>I think many parents and applicants use average SAT scores, or percentile ranges, as rough guides to decide whether to investigate further, or to decide if a school’s a reach or a safety. Many online college search tools allow users to set SAT/ACT parameters for the search. </p>
<p>Increasing the average, even slightly, may serve to discourage the lowest-scoring applicants, and to put the school on the list of “acceptable back-up schools” for students with higher scores.</p>
<p>How were the scores omitted? Were they from the wait list, or from late entering students? Was it intentional or just a pain to get those scores as they were not in the same place?</p>
<p>I could be wrong, but I don’t think I’ve ever seen the “average” test scores on the US News site (only the 25th and 75% scores). It’s obviously possible to have the same average scores and very different 25th and 75th percentiles. I was wondering how inflating the average would benefit the schools in the rankings, as opposed to inflating say the 25th or 75th percentile level. </p>
<p>The USNWR “premium” site reports:
SAT Critical Reading average score (e.g. Swarthmore’s is 721)
SAT Critical Reading 25th-75th percentile range (Swarthmore’s is 680-770)
SAT Math average score (Haverford’s is 700)
SAT Math 25th-75th percentile range (Haverford’s is 650-750)</p>
<p>From what I read, not all of the scores that were not included were below average. The average of the scores excluded are below average. There is a difference. I’m interested in what the reason was for the scores to be excluded. Were they SAT scores of kids who had better ACT scores so the ACT scores were used for admission purposes? In that case, some kids simply did not send lower SAT scores and the school might have just not included those scores since they were not using them for the admissions decision, but they were on file and available. Were they SAT scores of WL applicants, that were just overlooked. Was this deliberate or not.</p>
<p>A school I know (not a college) because it did not include standardized test scores of those kids in special services. It was the school designated in the district to take those children needing special support. Though mainstreamed to as much as they felt was possible, a number of those students were not anywhere near grade level, some non funcitoning whatsoever academically. To include those test scores, would have artificially deflated teh school’s profile since the other schools even in the same district had zero of such students since they were all sent there. But where they drew the line as to whose scores of the extra support needed kids came into question. </p>
<p>So it’s not always a diabolical plot to cheat on the stats.</p>
<p>Some fund managers get to “vaporize” hundreds of millions of dollars in segregated accounts but the Federal government takes no action against them. I’d imagine that we’d see the same protection for the connected at work here. </p>
<p>So how will this affect admissions for this year? Maybe Bucknell will accept more in anticipation that its yield rate will be substantially lower? I’m only wondering because I fall kinda in the middle range for Bucknell, and honestly I don’t care about them falsely reporting data. That doesn’t change the quality of education there.</p>
<p>How did last year’s false reporting hurt Claremont Mckenna and Emory? Or did those findings come out after all the acceptance and what not.</p>
<p>Except for the fact that Sarbanes-Oxley is a federal law. Hopefully the analogy does not extend that far. Because the notion of potentially prosecuting college administrators for submitting incorrect SAT information to a magazine strikes me as completely absurd. Hopefully their punishment will be a ban from the rankings and public exposure of their dishonesty (if it is not merely an innocent mistake).</p>
<p>I think USNEWS has to get much tougher with institutions that fudge data. If USNEWS wants to use certain criteria, then if schools fudge, they should be penalized.</p>
<p>I find such admissions reassuring. If many colleges are trying to nudge the SAT range a little higher, the cumulative effect is the impression that every candidate must walk on water in order to get in. If many colleges exaggerate even a little bit about their SAT scores, it’s as if they’re inventing hundreds of very high scoring applicants out of whole cloth.</p>
<p>If you scored at the college’s (official) 75th percentile, you may really be at the 80th percentile. If you think you’re at their 30th percentile, you may really be at the 45th percentile. In short, your score may be more impressive than you realize.</p>