I have been trying to get The New York Times to fix basic factual errors in Laura Pappano’s article on first-generation college students, which appeared in the paper on Wednesday. The Times quickly corrected a minor error. The most important factual error I reported was ultimately corrected, but only after a series of e-mails with the paper’s Education Life Editor, Jane Karr, during which she resorted to a personal insult that I find very telling.
I am providing our full correspondence. But I’ll warn you, it’s mostly a detailed discussion of statistics. The summaries I wrote below each e-mail offer a more concise account of the issues.
In the original version of Laura Pappano’s article, which I discussed in this post, there was this sentence about Washington University in St. Louis.
“Washington University in St. Louis, the least economically diverse top school, last year vowed to more than double enrollment of Pell recipients in five years, to just 13 percent.”
That wasn’t accurate. I wrote to Louis Lucero II, an assistant in the office of the Times‘s Senior Editor for Standards. He and I had corresponded about a previous NYT error (essentially the same error) that appeared in David Leonhardt’s article about Wash. U. and low-income students in January. Our correspondence was very polite, but the error wasn’t fixed. So I wrote the following message this time.
We corresponded last week regarding an error in David Leonhardt’s article on low-income students at Washington University in St. Louis. I’d like to point out that the error I reported in that article is repeated in today’s article by Laura Pappano:
Ms. Pappano writes: “Washington University in St. Louis, the least economically diverse top school, last year vowed to more than double enrollment of Pell recipients in five years, to just 13 percent.”
As much as I like the use of the word “just” in that sentence, there are two errors.
1. Washington University’s announcement of its new goal regarding Pell numbers came in January, not last year.
2. Wash. U.’s goal of 13% Pell Grant recipients applies only to the 2020-21 freshman class, not the total undergraduate population. Therefore it is not true that the university has “vowed to more than double enrollment of Pell recipients in five years.”
Pasted below is my e-mail from last week with a more detailed account of the distinction I made in Point 2. [See the blog post linked above for that.]
Best, Chris Pepus
Summary: Just read the numbered points above.
Jane Karr, NYT Education Life Editor, wrote back.
Thank you for alerting us to the error in our first-gen story about the timing of WashU’s announcement. You are correct that it was mid-January, not last year, and we have corrected the copy.
On your second point, however, I don’t agree that this merits amending. The entire paragraph is about freshmen, which is how Pell numbers are reported in this article. I don’t think the reader will assume we mean total undergraduate enrollment.
Thanks again for reading.
The proportion of freshmen at elite campuses who are first generation — 11 percent at Dartmouth, 12 percent at Princeton, 14 percent at Yale, 15 percent at Amherst, 16 percent at Cornell, 17 percent at Brown — nearly matches that of their low-income Pell grant recipients. Washington University in St. Louis, the least economically diverse top school, recently vowed to more than double enrollment of Pell recipients in five years, to just 13 percent.
Summary: We fixed the error on the date, but since the previous sentence refers to freshman enrollment, readers will understand that the sentence on Wash. U. does too.
Dear Ms. Karr:
Thanks for replying and for straightening out the error regarding the date of Washington University’s announcement. I see your point about the article’s reference to “the proportion of freshman” [whoops, “freshmen”] in the first sentence of that paragraph. But the second sentence mentions only “enrollment of Pell recipients” at Washington University in St. Louis. That seems to refer to general enrollments rather than freshman classes, especially since Washington University is not one of the colleges mentioned in the discussion of freshman classes in the first sentence.
But even if all statistics in that paragraph are read as pertaining to freshman classes rather than full undergraduate enrollments, the sentence on Washington University is still wrong. An increase in Pell enrollment to 13% by the 2020-21 freshman class would not constitute an increase of “more than double,” unless you have reliable information differing from that reported by both the U.S. government and Washington University administrators.
The U.S. Department of Education’s published data on Pell recipients at each U.S. college do not break down the numbers by year in school. The figure of 6% for Wash. U., which Ms. Pappano appears to be using, applies to the entire enrollment, as of the most recent reported academic year, 2012-13.
The Education Department’s list of Pell Grant recipients by institution is available as an Excel file on this page:
Federal data on enrollments are available at this link, by entering the name of the institution in the search field, clicking “Reported Data” and going on to enrollment data. (It took me a couple tries to figure that out.)
Washington University reports that its percentage of Pell students in the current freshman class is 8%. Here is the relevant quote from the university’s January announcement, with link.
“The percentage of Pell-eligible students in the 2014-15 freshman class has grown in recent years to 8 percent, from 6 percent in 2013-14 and 5 percent in 2012-13.”
So, barring another, better, source, Wash. U.’s percentage of Pell students in the freshman class is either unknown (according to published federal data) or 8% (according to Wash. U.). Neither supports a claim that Wash. U. will “more than double” the percentage of Pell students in its freshman class between now and 2020-21, simply by reaching its goal of 13%.
I would ask that you specify that Wash. U.’s goal of increased Pell recipients in 2020-21 refers to the percentage of freshmen, because I believe that sentence is misleading as it stands. In any case, the “more than double” quote in that sentence is contradicted by the best available sources.
Best, Chris Pepus
Summary: I don’t agree that it is clear that the word “freshmen” in the previous sentence demonstrates that the sentence on Wash. U. is about freshmen rather than students in general. The previous sentence did not mention that university. But even if you read it that way, it is not true that Wash. U. will more than double its freshman Pell enrollment in five years. Wash. U.’s goal is 13% by 2020. Its current freshman enrollment doesn’t appear in federal statistics and is reported by Wash. U. to be 8%.
Ms. Karr replied.
However fun it is to slog though ipeds data, please make your point by honing in on the numbers. Make a screen grab of them if you will. That’s the first full year prior to 2014-15 (I doubt this year is in the federal tables yet). So go back to what they actually have. Provide the raw numbers and conversion.
I see that as one sentence, total. Not a series of links and instructions.
Summary: If you can’t explain all this in a sentence, don’t bother me.
I understand Ms. Karr’s frustration with the federal government’s Integrated Postsecondary Education Data System (IPEDS) site, which is hard to navigate and requires scrolling through a lot of long documents displayed inside a small box on the page. But that’s not my fault.
I didn’t send a screen grab of the IPEDS data because it is it at the end of a very long document and you have to be able to look at more than one part of the document to see what the data refer to.
Here’s the numerical issue in a nutshell. The assertion in the article’s sentence on Wash. U., based on your construction of that paragraph, is that the university will increase Pell enrollment among freshman by “more than double” between now and 2020. That would only be true if Wash. U.’s enrollment of Pell recipients in the latest freshman class were below 6.5% of total students in that class. The latest federal Pell data (2012-13) give only the total enrollment for each institution, not enrollment by class. I’ve attached an Excel file of that data. Wash. U. is on Line 993. If you don’t want to open an attachment, my last message has a link to that Excel file on the Department of Education’s website.
I already sent you the statement in which Wash. U. officials claim that 8% of the current freshman class are Pell Grant recipients. For the “more than double” quote to be correct, you must have proof that Wash. U.’s freshman Pell enrollment is at least 1.6 percentage points below the number the university reports. That proof does not exist in any federal Pell data that I have seen.
If you have such proof, please share it with me, because I’d be very interested.
Sorry. That’s more than a sentence.
Best, Chris Pepus
Summary: You can’t say that an institution will “more than double” a number in five years unless you can prove that the starting number is less than half the ending number. You can’t, can you? Ms. Karr replied.
We used the latest ipeds data available in the college navigator. Under the financial aid tab ipeds offers data on both “all undergraduates” and “full-time beginning undergraduate students.” The “all undergraduate” data is 6 percent — and the “full-time beginning” is 5 percent. “Full-time beginning” is a worthy stand-in for “freshmen” and that going from 5 percent to 13 percent is easily “more than double.” Good luck with any new venture. We see you’ve quit your job over this statistic, but we are comfortable with what we’ve written.
Summary: Ms. Karr doesn’t say it in the message, but the IPEDS statistics we are discussing come from the latest academic year available from that source: 2012-13 Therefore, this is Ms. Karr’s argument: The sentence that says that Wash. U. promised in 2015 to “more than double” its percentage of Pell Grant recipients in 5 years is true because, if Wash. U. meets its goal, it will have more than doubled its percentage of freshman Pell Grant recipients between 2012 and 2020.
So 8 years equals 5 years, or, to put it more simply, 8=5. Of course, I recall reading in 1984 that O’Brien demanded Winston agree that 4=5. But asking someone you just met to agree that 8=5 seems to me like real hubris. On the other hand, if Ms. Karr were able to establish that 8=5, the problem of 13 not being “more than double” 8 would disappear. So there is that. I got this message from her next.
We are correcting it. See online.
The new version of the sentence on Wash. U. reads this way:
“Washington University in St. Louis, the least economically diverse top school, in January vowed to increase freshman enrollment of Pell recipients from 8 percent to 13 percent by 2020.”
So, as you can see, the friendly editors of The New York Times are glad to fix mistakes and misleading language when readers make them aware of the problem. I’m not thrilled about the Times using the 8% figure, since the Wash. U. administration is most likely the only source for it. But I have little doubt the figure is correct, and we’ll be able to make sure when the federal government releases its data for this academic year.
There is, however, the matter of the last two sentences of Ms. Karr’s penultimate e-mail in this exchange. “Good luck with any new venture. We see you’ve quit your job over this statistic, but we are comfortable with what we’ve written.”
You didn’t think I would let that slide, did you? See my next post.