One thing I think about is the way that other fields, where honest recommendation letters are not done, are much worse off in interviewing/hiring culture. The one I know best is software engineering, which for various reasons attempts to check if you can actually do the basics of the job in the in-person interview. I worry that giving up one of the true advantages that academic hiring has is worth it, even if we aren't sure exactly how it will go wrong.
One thought I've heard people express now and again is that it's not the content of letters of recommendation that matters as much as it is that you can get two or three people whose credentials are appropriate to what you're applying for to agree to write on your behalf (e.g., people who know about what it is you're applying for and who have a professional knowledge of the candidate). It's certainly true that if you look at an academic dossier and the recommenders are the candidate's mother, a supervisor from a past non-academic job, and the president of a local Rotary Club, you don't have to care about the content of the letters--that tells you enough. But this is really more like an argument for references over letters, and I'm not at all sure that references are any more preferable in terms of Schreier's concerns--they're just as intrusive to handle (maybe more so) for both referee and contact, they're not only prone to bias but when there is bias it's much harder to document, etc. It could be interesting to think about two levels of recommendation letters/communications: a simple one that is an attestation that a candidate can do X skills at a high level of competence and is prepared to undertake the work required and a more complicated one when the candidate is going to have to undertake some form of unusual or extraordinary work (some leadership role, working in a highly innovative lab or collaboration where they'll have to acquire new skills, etc.)
I think in many ways we have some of that already. For example, I was recently on the search committee for the new Dean of my school. There, we had no letters, and the search committee never heard anything from references. Instead, there was some process by which higher-ups got references, and I think in most cases talked with both named references and other people familiar with the candidates. So for leadership roles the standard letter of reference is already partially gone.
Yes, and the work of referencing is basically being hidden from people involved in the work of choosing. At best that just means that the process of referencing is "hey, to your knowledge, has this candidate murdered anybody or embezzled millions of dollars?"--it's not an evaluative tool, just a mechanism for limiting certain forms of liability.
My professional work involves reading and writing lots of recommendations for undergraduates and young alumni. I really appreciate the sentiments you express in today's piece. I do think those of us who rely on recommendations can do a better job guiding writers on what would be considered "added value" to an application through their letters.
One thing I think about is the way that other fields, where honest recommendation letters are not done, are much worse off in interviewing/hiring culture. The one I know best is software engineering, which for various reasons attempts to check if you can actually do the basics of the job in the in-person interview. I worry that giving up one of the true advantages that academic hiring has is worth it, even if we aren't sure exactly how it will go wrong.
One thought I've heard people express now and again is that it's not the content of letters of recommendation that matters as much as it is that you can get two or three people whose credentials are appropriate to what you're applying for to agree to write on your behalf (e.g., people who know about what it is you're applying for and who have a professional knowledge of the candidate). It's certainly true that if you look at an academic dossier and the recommenders are the candidate's mother, a supervisor from a past non-academic job, and the president of a local Rotary Club, you don't have to care about the content of the letters--that tells you enough. But this is really more like an argument for references over letters, and I'm not at all sure that references are any more preferable in terms of Schreier's concerns--they're just as intrusive to handle (maybe more so) for both referee and contact, they're not only prone to bias but when there is bias it's much harder to document, etc. It could be interesting to think about two levels of recommendation letters/communications: a simple one that is an attestation that a candidate can do X skills at a high level of competence and is prepared to undertake the work required and a more complicated one when the candidate is going to have to undertake some form of unusual or extraordinary work (some leadership role, working in a highly innovative lab or collaboration where they'll have to acquire new skills, etc.)
I think in many ways we have some of that already. For example, I was recently on the search committee for the new Dean of my school. There, we had no letters, and the search committee never heard anything from references. Instead, there was some process by which higher-ups got references, and I think in most cases talked with both named references and other people familiar with the candidates. So for leadership roles the standard letter of reference is already partially gone.
Yes, and the work of referencing is basically being hidden from people involved in the work of choosing. At best that just means that the process of referencing is "hey, to your knowledge, has this candidate murdered anybody or embezzled millions of dollars?"--it's not an evaluative tool, just a mechanism for limiting certain forms of liability.
My professional work involves reading and writing lots of recommendations for undergraduates and young alumni. I really appreciate the sentiments you express in today's piece. I do think those of us who rely on recommendations can do a better job guiding writers on what would be considered "added value" to an application through their letters.