You won't be surprised to learn that the US government has been tracking the level of student loan debt incurred by graduate students on a per program basis.  That recently released data is in the College Scorecard here. I've created a few charts for you to explore the data further.

As I understand the data, the College Scorecard matches lists of people who have graduated from each program (provided by schools) with student loan data on the same people held by the Treasury department.

What we end up with is the average student loan debt of graduates by program. Ostensibly, this could be useful for prospective students who are researching their graduate school options. As you'll see below from my comments, I don't think the data are quite ready for comparison shopping. But they are interesting to sort through.

Explore My Dynamic Plots

The graph above is one of a few ways to look at this data.  But it is static.

I've also created three different interactive graph views, that you can view here, here and here. These are much easier to explore, but you'll probably need a laptop.

Some Warnings About the Data

There are few issues to bear in mind. The biggest issue is that there is a lot of missing information.

The Scorecard seems to use a lower cutoff of 10 graduates per program in deciding whether to make public an average loan value, due to privacy concerns. Thus, average loan values are not available for smaller programs. That's a big deal since most programs are small and most graduate students are in small programs.  The data set is biased to larger programs, which could be a meaningful bias.

The values in these charts correspond to about 14,000 graduate degree programs. We figure there are about 60,000 graduate programs out there that anybody might consider attending.  Chances are that data is missing for 3 out of the 4 programs you might wish to compare.

Finally, these numbers represent total student loan debt. We already know that most student loan debt is incurred in the graduate school years. Nevertheless, the amount of debt attributed to the undergrad years is not parsed out.

Some Key Take Away's

  1. Students earning professional doctorates can have much higher loan burdens than others. That shouldn't come as a big shocker. That's best seen and explored here. I'm talking about degree holders in the healthcare fields, psychologists, lawyers and education doctorates.
  2. Within most academic fields, the higher end outliers for debt burden are programs at private nonprofit universities. You get a general sense that the debt burden of public university graduates tends to be lower. That's best seen here.
  3. In most fields, people who earned a masters have lower debt than those earning doctorates. For example, a doctorate in the health fields will incur more debt than most master's. But have a closer look, in many fields, doctorate students are about the same as masters, debt-wise. Stipends and scholarships associated with some research doctorate programs can lead to lower debt burdens than seen for many masters programs even though the time in program can be twice or three times as long!  
  4. This isn't a sticker price comparison. The more you explore and think about the data, the more hidden variables you can imagine that might explain how the debt held by graduates differs between any two programs in the same field. YMMV.
  5. Most of the stories have been and will be written about the outliers. For a reality check look carefully at a distribution of the program means. As you can see, most programs graduate people with debts less than $100,000. The average graduate program produces graduates with an average mean student loan debt of around $50,000.