Simon Tonev

…That is to say, they must be so understaffed and so under pressure that they don’t even have the capacity to run the code before posting it.

I spent another few hours today correcting the SAS files for the 2023-24 final datasets that were just released. As with the preliminary datasets, the syntax errors are such that simply running the code would have revealed the issues…

Well I took the afternoon to go through the SAS code and found several more syntax errors. I offered to work with RTI on corrections and they (quickly) took me up on it. They must be SLAMMED.

Thanks so much for pointing this out. I’m currently going through the SAS code and am finding typos that are causing errors reading in the data file. So far, I’ve spent about 15min on HD2024 and and still not done. In previous years, I spent 0 minutes.

In addition, USNWR has added metrics that are *only* available from CSC, like % of college grads out-earning HS grads. Since those data weren’t available in time for the ranking last year, USNWR had to rely on the previous years’ data. It appears that they might have to do the same thing this year.

Ranking publications like US News have only recently made the (reasonable) decision to lean more heavily on IPEDS and Scorecard data over data provided directly by schools. Looks like they may have made the wrong gamble.

The FVT/GE reporting process has been an absolute nightmare for many of my colleagues. A great example of the admin “bloat” that is essential to complete new and ill-thought-out statutory reporting requirements.

You’d like to know 30-year trends for undergraduates graduating with English majors by sex and race/ethnicity?

Today: “I can get you something this afternoon.”

Not-to-distant future: “If I drop everything else, I *might* be able to get you something mediocre in 3 months?”

100%. Not sure how my colleagues feel, but I’ve had almost exclusively excellent experiences with IPEDS both from the collection, use, and support standpoint. Can’t say the same thing for my (likely underfunded) state ED.

*very* small potatoes compared to the other implications, but I haven’t seen anyone mention the data collection function of ED (e.g., IPEDS via the NCES). Not gonna love my job when I have to go to 50 different sources who collect data on 50 different systems using 50 different instruments.

And you’re absolutely CERTAIN that confirming 3 liberal judges in a trenchcoat is not a strategy worth trying?