Med School Interview Questions

The questions are no big deal, but what I find interesting is that medical school do personal interviews at all. No place where I’ve ever worked has interviewed grad school applicants. It’s hard for me to see what you get from it, that it would be worth the cost. I guess there must be quite a bit of psychology literature on this question.

22 thoughts on “Med School Interview Questions

  1. I get the impression that a lot of biology (and related) grad programs do interviews. My wife had interviews with a bunch of neuroscience departments.

  2. I just started up as a Ph.D. student in neuroscience after applying to many different programs and types of programs (psychology, cognitive science, neuroscience, speech and hearing biosciences). Every single program did interviews. The interview process was less rigid however, as most of these specific questions had already been addressed in the personal statements.

    Indeed, it seems that for the biomedical sciences grad programs, the vast majority of interviewees are ultimately accepted and what I thought was an "interview" is called "recruitment" now that I'm in the program.

  3. This is not unique to med school. Top MBA programs require interviews since networking skills are so important, and they're hard to determine from a resume.

    There's also the idea that if you can interview well for school, you can interview well for a job, so maybe it helps them in the long-term to have a higher job placement rate from their program.

    Assuming the interviewer is good at their job, there are many intangibles that can be picked up through interviewing.

  4. In Sweden, which also have interviews, is the idea to get students that are highly motivated and suitable to become a doctor. I guess that since education and health care is funded by the state in Sweden, these things becomes more important compared to countries where you pay for it yourself.

  5. I am a former student of Robyn Dawes (who knows quite a lot about how terrible personal interviews are for providing information about how to pick from a pool of applicants). I am also a 4th year medical student currently applying for residency programs (as an aside, residency programs are the next step after medical school and both basically require personal interviews).

    As far as I know, there is no evidence that personal interviews for medical school or residency are at all effective. Of course, it depends on what you consider effective . . . passing the next step of boards? interacting well with patients? interacting well with your peers? not dropping out?

    The strongest evidence, in my opinion (this is second hand from Robyn Dawes, I haven't read the original studies), against in-person interviews comes from some time in Texas when the legislature demanded that the medical schools increase their class size after the classes for the year had been picked. The set of people who were interviewed was those who passed some sort of grades + test scores application cut off. Those who had been offered a position were those who had scored the best in the in-person interview part. The legislature then basically forced them to extend offers to those with lower in-person interview scores. At the end of medical school, there were no differences in the performance of the two groups.

    Of course, as a residency applicant this year, I get to spend lots of time and money doing all these interviews . . .

  6. One negative aspect of the interview process is that it's terribly expensive for people who are at a time when they have very little money. My wife went through it a few years ago, and had to travel all over the country doing interviews.

  7. As I understand it, there's considerable evidence that interviewing people always leads to making poorer decisions about their suitability. Of course, people who think otherwise are unlikely to be convinced by dry statistics, humans being what they are, but as far as I know the (extensive) research in this area hasn't really turned up any exceptions; even if you think you're interviewing people for something where their people skills are important, apparently how well people can fake it during a one-time, high-pressure encounter isn't much of a predictor of how they'll do in any other context.

  8. In psychology it has long been pretty standard to interview clinical psych applicants, as a way of assessing "clinical skills" (whatever that means) and filtering out people who were smart enough to have good academic records but have personal issues that would be obstacles as clinicians. Minnesota used to be somewhat notorious as an exception — they required a professionally administered MMPI instead of an interview. (I think they now do interviews, though I could be wrong.)
    But in the past decade or so there has been a trend toward doing interviews for non-clinicians too (i.e., just-research PhDs). My department just started doing that a few years ago. As Janel points out, there's relatively little evidence that they actually contribute useful information in selection. But people have a hard time believing that their interview impressions don't matter.
    However, we've started to suspect that there is a recruiting advantage. When we conduct interviews we get to show people around and introduce them to the people in our department — in effect, letting them interview us. More speculatively, there may be some kind of signaling going on. Departments that don't conduct interviews may come across as less selective, and therefore less attractive to the best candidates. Sample sizes are too small to distinguish from chance, but I think our acceptance rate has gone up a smidge since we started interviewing.

  9. Isn't is a good thing that at the end of a program that poor interviewees and good interviewees have the same outcome? Doesn't it just mean that the poor interviewees have learnt the skills they lacked on entry? It would be interesting to know how much extra work (and cost) has gone into bringing them up to standard.

    In New Zealand there are interviews. There used to be none and entry into medical school was based solely on exam results but that used to result in getting "book swots" with zero people-skills and (more recently) poor English.

  10. I'm a physician and health policy researcher. Medical schools have a much stronger sense of "the person" than PhD programs. They care more about extracurriculars, volunteer work, and think they need to assess the individual's personality.

    The med ed literature proves that what they're doing is wrong. The testing characteristics of 2 half-hour interviews says more about the interviewers than applicants and leads to worse decision-making, but no one can move away from it because you're seen as not caring.

    The smart solution was from McMaster University in Ontario, who developed a series of 10 brief interview stations that are designed to elicit medical skills – delivering bad news, etc – and, more importantly, providing a more stable sample size for more accurate assessment. They're finding better reliability and, so far, better validity of predicting residency achievement.

  11. Not only do they use it, they rely heavily on it. For my institution in 2007 7787 applied, 749 interviewed, 301 were accepted, 112 enrolled (or so claims the internet). This statement on the admission FAQ made water come out my nose when I read it: "Applying to medical school is not like playing the lottery; chance does not play a role." Saying more about why they rely on the interview so much would be impolite. The MD/PhD program takes the opposite approach: almost everyone who they bring in for an interview they are ready to admit; the interview is recruitment and detailed lifestyle and compatibility checking.

    Medical school applicants tend to apply to a largeish number of schools, which results in the huge reduction requirements from a mostly qualified pool (half of applicants will get in somewhere). The cost of the interview process is mostly born by the applicants and hidden in the institution; I don't know that departments are billed for the faculty time that gets consumed. Arbitrary and grueling admissions processes make the school more selective (appearing at least).

  12. The statistics department at NYU also interviews its incoming doctoral students. I do not know the official reasons, but it may be because it is in a business school or because it is a particularly small program.

    Of the statistics and economics departments to which I applied, it was the only one that interviewed me before deciding.

  13. Dear god, we just had an interminable faculty meeting about exactly this topic. In psychology, I think the idea is that for some programs, we have a relatively small number of applicants for the PhD program and the training is very much of the apprenticeship type. There definitely is literature suggesting relatively poor predictive validity of interview performance for later success. But, many of us would argue that we are less attempting to score people on a continuous scale but rather that we have almost a binary measure known as the 'two headed test'. That is, we are looking to ensure that the person is not a complete freak and that the particular advisor can at least at the outset, imagine working with this student. A great deal of noise in this measure as well but I think the extremes of the distribution are better picked up in interviews than the middle portion of the distribution.

  14. Sackett and Lieven's 2008 Annual Review of Psychology article provides a good overview of the personnel selection literature from an I/O psychology perspective.
    <a>A pdf of the article is available online.

  15. All three (stats) PhD positions I applied for as I completed my undergrad involved interviews. I don't see how (i) supervisor can judge personal chemistry with student, (ii) student can judge personal chemistry with supervisor, (iii) student can judge atmosphere of department, (iv) department can justify using what originally was tax payers' money—without interviewing.

    This was in the UK, though, where things differ.

  16. My father gave medical boards, did some admissions, taught in med school, etc. If he were alive, he'd say it's a carryover from the old days that he could summarize by saying, "You'd better answer the why question by saying that you want to help humanity." He used to say they tried to weed out the nut cases. BTW, to connect this to some statistical interest, he gave radiology boards and would ask a number of basic questions because he said, "You'd be amazed at the number of people who can identify the most obscure things but can't diagnose the basics like diabetes." So would that be a better way of examining? Is the low bar of weeding out whackos effective – and I knew a brilliant guy who couldn't get into medical school because he was obviously nuts.

  17. Independent of the evidence, med schools and residencies require very intimate working conditions. We may be poor judges of character, but we still like to have in-person interaction with people we'll be spending so much damned time with.

  18. In Linguistics Ohio State invites people to on-site recruiting visits after we have made an offer, but Stanford (whom we compete with, fairly successfully, for Ph.D students) does on-site interviews before deciding about offers. I suspect that the experience of the on-site visit is similar, except that at Stanford people are told that decisions have not yet been made. In practice, it seems probable that the faculty already have a pretty good idea who will be selected, and that the Stanford visit is mainly about ratification and/or sanity checks.

    I've heard that many interviews and aptitude tests have benefits that come from their positive effect on the people selected. Marines (or people who programming aptitude tests in the 70s) feel privileged to be part of the selected group and this has good effects that are separate from any reliable ability the test might have in choosing the best people out of the pool. [while I do think the above is plausible, I can't back it up properly. I felt bad about the vagueness of the "I've heard" above. I suggest that anyone who really cares should do a more thorough job on following up than I did, including

    Mayer, D. B. & Stalnaker, A. W. (1968) Selection and evaluation of computer personnel – the
    research history of SIG/CPR. In the proceedings of the 1968 ACM National Conference (23rd
    ACM National Conference), 657-670.

    and I think that studies using the term 'face validity' may yield good perspectives]
    In the case of subjects like Linguistics, where graduate students are typically not paying their own way, it is also very beneficial to give potential recruits a sense of what the departmental culture and peer group is like. This helps to ensure that we spend the university's money (which ultimately comes from paying students and state taxpayers) on people who look to us as if they will fit in and themselves feel that they will succeed in the program. Given my suspicions that the admission decisions at Stanford are often pretty much a done deal at point of 'personal interview', I wonder whether Ohio State should pretend to still be in doubt, simply in order to enhance the experience of the visit weekend. But you didn't hear me say that …

  19. Michael,

    The Bob Sutton blog posting you link to argues, based on a meta-analysis, that structured interviews are good at predicting job performance and that unstructured interviews are a weaker predictor. The best predictors were work samples and some measure of the general factor in intelligence. I have vague recollections of reading the original paper and that seems a good summary.

    Sutton has written a delightful book on the workplace environment entitled The No Asshole Rule. In there he reports the interviewing practice of an innovation company (pp. 64-65). References are important ("from people we trust"). Competence is checked before interview (presumably via application form or CV) so the focus can be on the applicant's "human qualities" in an interview – how well they'd fit into the work environment. They also argue that it's important to have people from a range of status levels at the interview: people who would be below, above, and alongside the candidate if they were to get the job.

    I know of no work which tests what affect this approach has on the general vibe in the workplace – which must be at least as important as the ability of individuals. I also have no idea if this could be applied to grad school. Seems like a good idea to try though, and see what happens.

    Andy

Comments are closed.