I don’t know what it is about college cost reporting, but it either attracts shoddy reporters, or makes good reporters shoddy.

Take the recent piece by Andrew Martin and Andrew Lehren. Here’s a paragraph that actually appeared in it:

Ninety-four percent of students who earn a bachelor’s degree borrow to pay for higher education — up from 45 percent in 1993, according to an analysis by The New York Times of the latest data from the Department of Education. This includes loans from the federal government, private lenders and relatives.

For all borrowers, the average debt in 2011 was $23,300, with 10 percent owing more than $54,000 and 3 percent more than $100,000, the Federal Reserve Bank of New York reports. Average debt for bachelor degree graduates who took out loans ranges from under $10,000 at elite schools like Princeton and Williams College, which have plenty of wealthy students and enormous endowments, to nearly $50,000 at some private colleges with less affluent students and less financial aid.

I honestly don’t know where to start.

• Ninety-four percent is an absurd number to assert for student borrowing if we take a standard measure of borrowing. Estimates vary, but about 1/3 of students do not graduate with outstanding loans. I’m sure there’s some convoluted reasoning behind the number (If my parents give me laundry money, is that “loans from relatives?”). But it’s not a number I’ve ever seen in 15 years of following this, and it is in direct contradiction with the sources they pull their other data from.
• Here’s a question — when do you use the median instead of the mean? There’s a simple answer — when your data is skewed or outlier-prone. They give the mean debt here ($23,300), but not the median debt ($12,800). Anyone even marginally statistically literate can see from those figures that the student debt problem is significantly a problem with the 10 percent owing over $54,000 and the 3 percent owing over $100,000.
• Who are those people with the massive debt? On average, they aren’t 23-year-olds. As the paper the article links to indicates, the highest debt is in the 30-39-year-old bracket, followed by the the 40-49 year old bracket.
• Which brings us to another issue. The big borrowers tend to be medical and law students going for graduate degrees….

I could keep going — but why bother? No one gives a flying fig about the reality of the debt situation apart from furthering their own agenda, so most people will forward it if it supports their worldview and ignore it if it doesn’t.

They’ll be shocked that 94% graduate with loans, and when they find out it’s 66% or whatever,  they’ll not be shocked that it is lower than they thought. They’ll still be outraged.

When they find out the median student debt is half what they thought it was, they’ll be…. well, outraged! When they learn that 30-somethings have the highest debt, they’ll be — well, you get the picture.

All this is a shame, because if we’d stop the forwarding of these stories (and shame on you if you forwarded this article) we’d see that the problem of student debt is very different from what the NY Times would have us believe.

For instance, if we understand that student debt is not driven by 94% of students graduating with moderately large loans, but is partially driven  by 10% students graduating with insanely large loans, that piece seems a pretty easy one to fix. Figure out the reasonable maximum cost for an undergraduate education (say, $50,000 including housing), and don’t guarantee loans over that. Let loans over that be subject to bankruptcy, etc.

Take the student in the beginning of the article, who went to Ohio Northern. That’s near $50,000 a year list for that school (an outrageous school for the Times to pick for an anecdote, but there you go). But that price is not indicative of what college costs to anyone who is marginally price-conscious.  Ohio University has a list price, tuition and board,  of $20,000 (in-state) and $28,000 a year (out-of-state). Tuition at the University of Wyoming will cost you $11,000 or so (out-of-state), and the average aid package there is about $12,000.

While there may be some closing of these gap with tuition reductions, why in the world would we underwrite loans for Ohio Northern for no demonstrated additional benefit? Isn’t that (and not widespread loan in smaller amounts) really a bigger problem?

This is just one example. We could also talk about law school loans at a time when the lawyer surplus makes law school a path to poverty. We could look at what is actually going on with the high level of debt for people in their 30s — which could indicate both advanced degrees and long defaults accumulating interest.  These are the sort of questions you get to pretty quickly when you actually dig into the numbers. Unfortunately you’ll get to none of them through reading the New York Times.