DFW rate stands for the proportion of students in a class or group of classes who earn a D, an F, or Withdraw. This has become a trendy metric in recent years due, in part, to its negative correlation with retention. A search for DFW on the Chronicle of Higher Ed site, for example, returns 10 hits since 2018 - several of which mention retention as well. As a professor of mathematics at a small, public university in the US with a focus on the liberal arts, I've found myself thinking about our DFW rate and found that I don't really know the answer to the question:

What is a healthy DFW rate?

Such an answer would certainly be multifaceted (balancing faculty expectations with those of the administration) and would also depend on the student body. Since this is not a discussion site, though, let me be clear that I'm specifically wondering:

  1. Are there any authoritative publications on the question?
  2. Is there any publicly available data that would allow us to compute or estimate DFW rates for specific classes at a wide variety of schools?

Note that I have found some papers (like those in the Chronicle) that discuss DFW rate. They all tend to be very isolated, though, dealing with one or two anecdotal cases without mentioning any specific goals for DFW rates.

With regard to question 2, I should mention that I'm quite adept at dealing with IPEDS and, while there's plenty of data surrounding grades and retention, there doesn't appear to be any data of sufficient granularity to compute, say, the DFW rate for Introductory Calculus.

  • 1
    I think this question is about the content of your research and should be closed. You should be looking in higher ed research journals, not the Chronicle. There is no simple answer to this question as there are many variables which relate to student success. Jan 26 at 19:30
  • 5
    @AnonymousPhysicist I am a mathematician so this question is definitely does not concern my research; rather, it concerns my experience in academia. I do agree that just one metric is insufficient to measure overall student success. This just happens to be the one I'm curious about with regard to this specific question. Jan 26 at 19:45
  • 4
    The funny thing about this question is that it is hugely culturally variable. Various European countries seem to regard as optimal DFW rates (say 60%) that would be considered completely unacceptable in the US, and this is despite having strict qualifiers for who should go to university. Jan 26 at 20:12
  • 1
    Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking.
    – Community Bot
    Jan 26 at 20:15
  • 3
    Healthy for what purposes? What goals are being considered? For example, if the course entrance selection criteria were perfect the DFW rate should be zero.
    – Dan
    Jan 26 at 20:38

1 Answer 1


Determining a healthy DWF (or DFW or DWFI, adding Incomplete) rate is a policy question, since it depends on how your university would define "healthy." For instance, is "healthy" a matter of sustainable enrollment delivering tuition dollars that keep the university budget sound; is "healthy" a matter of outcomes for students who want to graduate in 4 or 6 years; is "healthy" a matter for departments who want to ensure they have enough students progressing into and through their majors and minors; is "healthy" a comparative measure between your institution and institutions like it (small liberal arts colleges, R1s, community colleges); is "healthy" a question of programmatic rigor? It's likely that an academic affairs unit will consider several factors at once if they are determining a target measure.

As a result, the literature usually doesn't refer to a single numerical "healthy" rate. Indeed, I have seldom seen "healthy" used with DWF. Instead, I usually see "high DWF" or "low DWF," which are set relative to expectations within a specific context. For instance, how is one course's DWF rate as compared to expectation or to an average of similar courses, or how is one university's DWF rate compared to its peers? For example:

  1. Mutanyatta-Comar & Mooring. (2019). Evaluation of a peer-led team learning-flipped classroom reform in large enrollment organic chemistry courses. From general to organic chemistry: Courses and curricula to enhance student retention. American Chemical Society:

The course presents conceptual challenges for many students in these large sections resulting in a high DWF rate (approximately a 30% DWF rate).

  1. Ferri et al. (2019). Three models for blending classes in a multisection course. Blended Learning in Practice : A Guide for Practitioners and Researchers. The MIT Press.

The DFW rates, another department-centered metric for the course, falls in the range of 12% to 14%, which is comparable to other required junior-level courses in the department.

  1. Tallarico & Wisniewski, M. A. (2015). A better way to teach principles of microeconomics? Evidence from the classroom. Journal of the Academy of Business Education, 16, 21–38.

The method of assessment had a clear impact on the DWF rate, increasing from 10 percent when using the quiz method to 56.67 percent when using the exam method.

So beyond the general definition of what the measure looks at (failure, withdrawal), there isn't a standard way to interpret the rate.

Finally, it's possible that program chairs and other leaders have a gut idea of what a good DWF rate looks like in their context. For instance, in mine (large public university with an access mission in the United States), lower than 10% is low / good and higher than 30% is high / bad. But if I had to argue one was high or low in a research or assessment context, I would always carefully define compared to what because those terms don't mean much by themselves.

  • You are correct to emphasize that none of these citations talk about a healthy rate. Just an observed rate.
    – Buffy
    Jan 26 at 21:12
  • @TaliesinMerlin Yes, you're exactly right - it's policy question. I was motivated to ask the question because we (our math dept) have been asked by our administration to lower our DFW rate without clear policy on what a reasonable rate might be - let alone data support such a rate. Thanks! Jan 26 at 21:18
  • @MarkMcClure: You might want to consider one of my observations about universities: the political and economic context pushes most universities in the direction of becoming a diploma mill. Jan 26 at 21:53

Not the answer you're looking for? Browse other questions tagged .