Like all faculty members, I’ve written lots of reference letters. I’ve written letters for great students and for those with lots of ‘potential.’ I’ve written letters to public and private employers; to scholarship, fellowship, and awards committees; to graduate and summer research programs. You get the idea. I also review applications for our graduate program and write letters for undergraduates, which means I’ve even written letters to myself: “Dear future Leidy, Freddy is an excellent student and person, I am certain he will succeed in graduate school. You will thank me for recommending him (and for my healthy living). Sincerely, present Leidy.”
Over the years I’ve become more efficient in my letter writing; letter requests are rarely so unique that I need to start from scratch. Still, I easily spend 20 hours a year writing reference letters. If the other 1.2 million college instructors in the U.S. are having similar experiences, then reference letters are well over a $1 billion per year industry (to put this number in perspective, I like to think of my salary, and add $1 billion dollars). The $1 billion figure is conservative, and it completely ignores staff time to handle, file, and store the letters, not to mention the students’ scavenger hunt for willing writers.
I should be clear that I really don’t mind writing letters. Reflecting on each student’s distinctive qualities and how they showed up in our interactions is quite rewarding for me. So, if one of my students ends up reading this, please don’t be scared to ask me for a letter, you are not the problem. At fault are those of us who ask for the letters.
The vast majority of the time, requiring a reference letter is stupid. When Nicole has a 3.9 engineering GPA, scored perfectly on the GREs, and brings water to children in Haiti in her free time, what can I possibly say that makes any difference? She should be evaluating me. Reference letters are just as pointless for the opposite types of students. Even the worst products on amazon have a few good reviews. Fortunately for us, the seller isn’t able to pick just three to show you. Letters can also be stupid times two, three, or even five. Who decided that three should be the default number of required references? I envision an historic committee meeting centuries ago at some prestigious institution where an esteemed faculty member successfully argued; “you know, many applicants are able to come up with two good letters, but that third letter really makes all the difference.”
Of course there are exceptional cases where a reference letter matters. But, what percentage of reference letters do you read in your application review processes? What percentage of letters matter? Any unread letter is a waste; one that is read but makes no difference is even worse since it takes both the writer’s and the reader’s time. Fortunately, since this problem is pretty much all our fault, we can actually do something about it. A simple option is to just adopt what is already done for many faculty searches and only ask for letters in cases where it is going to make a difference. Or, better yet, just call the references in these cases. Widespread adoption of this simple change would save hundreds of millions of dollars worth of our time.
We’ll continue wasting our time writing letters until we’re all willing to spend a minute of our time to save someone else an hour of theirs. Then, we can all use the saved time to get to know students better. It will help us write better letters.