It depends on the college, but for most colleges that do interviews, I don’t think this is accurate. One of the few colleges for which we have specific numbers is Harvard, due to the lawsuit analysis.
One of Harvard’s experts in the lawsuit created a model that was able to explain 64% of the variance in admission decisions based on ratings of applicants received in a variety of categories, and a variety of controls. The percentage of predictive ability lost by excluding ratings from one category were as follows. LORs and alumni interviews were grouped together in this model. The isolated pseudo r^2 numbers suggest LORs have ~1.5x the predictive ability of alumni interviews. Using this ratio results in the following approximate loss in predictive ability when excluding different parts of the application.
Excluding LOR Ratings: Lose 33-40% of predictive ability.
Excluding Personal Rating: Lose 19% of predictive ability
Excluding Interview Ratings: Lose 16-20% of predictive ability.
Excluding Academic Rating: Lose 17% of predictive ability
Excluding Extracurricular Rating: Lose 13% of predictive ability
This model suggests while interviews may not be tremendously predictive of decisions in isolation, they often add something to the admissions decision that is not duplicated by the other components of the application. The model suggests that Harvard admission decisions would be more likely to change if they stopped considering alumni interviews as part of their admission process than if they stopped considering test scores (information gained by test scores appears to be largely duplicated by other sections of the application). Consistent with this, the applicant files sometimes include comments about the “alumni IV” and suggest that it is influencing the decision.
As mentioned above, there is a good amount of variation by college. For example, some of Cornell schools state that alumni interviews are only for informative purposes, not influencing decisions. Others Cornell schools state that interviews are mandatory for nearly all admits. Statements on MIT’s website and CDS suggest MIT interviews are likely more influential than at Harvard. Different colleges use different admission systems.
I have been a Stanford alumni interviewer for the past 6 years, and have interviewed about 30 kids. None of them were accepted, including a few kids that I thought were terrific and of whom I wrote rave reviews. Granted, I did not have any access to their stats, LORs, essays, but a few of these kids seemed likely admits, for example a URM person who would also be first gen college, had a parent in jail, did MITES at MIT, worked 20 hours a week supporting his family, and wrote a novel. It is dispiriting when so many great kids that you root for get rejected. I am no longer doing it because I felt that I was just leading on these kids.
I agree, that one of the main purposes of the alumni interview is to keep the alumni connected. One other purpose may to try to increase yield among those who are accepted. I have been a Stanford interviewer, and there are explicit instructions to reach out to any of your interviewees who have been accepted.
I think we have to remember that the bulk of applicants to an elite are rejected, lots of great kids. This concern among interviewers, that “their” candidates don’t get an admit is like the concerns about admissions, in general. It’s often tough to grasp that it’s not enough to be a stellar student (and/or human being.) And not only that, but there are those dang institutional needs that play. The cold reality.
Many kids on CC get confused and think just getting an interview represents being a finalist or semi-finalist. Likewise, some past interviewer comments on other threads seem to show some interviewers feel they should have more authority or weight in admit decisions. Ime, the meet report can add so much to a picture of the applicant. In ways, an alum is like a scout, “eyes-on.”
The report can turn a so-so reviewer reaction to more. Or lend to the certainty some kid is not a contender. But it’s one piece of the vaster puzzle.
It doesn’t matter, imo, why alums are asked to interview, to keep them connected or pass on info. Bottom line is, when there’s a report, it’s the only face-to-face input a reviewer gets, unless a rep happened to meet a kid during a hs or community visit. Kids can unexpectedly shine or blunder, promote their case or not. That’s informative.
For those colleges that require or recommend alumni interviews, it isn’t hard to discern their motivations. It’s primarily a tool to keep alumni engaged. It’s only ancillary and highly unexpected that a few of these interviews may help shed some light on a few applicants, either postively or negatively.
You’ve posted this opinion before, yet have not provided any evidence. The Harvard analysis makes it clear that interviews are influential in decisions at Harvard (or at least were at the time of lawsuit collection). Other highly selected private colleges that provide specific numbers also suggest an influence. For example, prior to COVID, MIT’s website used to state that students who were offered an optional interview and chose to wave it, only had a <1% acceptance rate. MIT also marks interview as “important” in the CDS and has an especially long and detailed alumni interview report.
You mentioned “require” interviews in the quoted text. The few highly selective private colleges that require interviews are often unique. For example, Cornell Art used to require interviews prior to COVID, and those interviews involved discussing their portfolio with an Art alumni or Cornell faculty. It seems counterintuitive that a college would require interviews, if they rarely have any influence on decisions.
Is the evidence the anecdotal reports that some interviewers for highly selective colleges with <10% admit rates see the vast majority of the kids they interview rejected, including ones for which they strongly recommend? How would that differ from any other section of the application? For example, I’m sure some teachers who write LORs see that every student from their HS seems to get rejected from a particular <10% admit rate college year after year, including the kids for which they write exceptional LORs… This does not mean LORs are primarily a tool for maintaining teachers relationship with the college. This only means that there are other considerations in the admission decision besides just LOR or just interview.
I expect that interviews have different degrees of influence at different colleges, and it’s not accurate to make a sweeping statement. At at least some highly selective colleges, interviews have a noteworthy influence on decisions, like other sections of the application.
My daughter had an excellent interview with a Dartmouth alum a couple days ago. It was pretty clear she was going to strongly recommend my daughter for admission … I wonder how much these interviews really matter. I was trying to temper my daughter’s enthusiasm afterwards, by telling her the alumni don’t really have much say in the matter!
Your “evidence” always seems to be some regression analysis in the Harvard lawsuit. Regression analyses are never conclusive for anything and they can be used to tell any side of a story, as is the case in the Harvard lawsuit. Both sides of the lawsuit used similar regression analyses to reach very different conclusions.
I was a Wesleyan alumni interviewer for years. All my interviewees were great kids. I nearly always gave them great scores. How could I not? We were specifically told not to evaluate them academically; that wasn’t our job. I saw it more as weeding out kids who clearly had no social skills or perhaps signal which ones for whom Wesleyan was a first-choice. If I had five interviews a year, I was lucky if one gained admission. Maybe, in some global measure that I am unaware of, my recommendations might have had a correlation with happy outcomes. But, I’d be hard put to read more into it than that.
I listed 3 different schools in my post – Harvard, MIT, and Cornell – and listed 3 different reasons for why interviews appeared to be influential at each of the schools, only one of which was regression analysis. While this evidence may not be perfect, some evidence is better than no evidence. And I have still have not seen any evidence for your alternative claim.
Regarding the Harvard lawsuit analysis; the respective models for the two sides of the lawsuit actually were actually extremely similar and showed extremely similar conclusions, including in regards to how different sections of the application appeared to be correlated with the decision. The hundreds of pages of debate and rebuttals more relates to whether a borderline factor reached statistical significance or not. The factor was close enough to the borderline that small differences in assumptions and controls changed whether it reached the statistical significance threshold or not. Having one of dozens analyzed factors in analysis being borderline statistical significance does not mean that everything else in the analysis is garbage. particularly for conclusions that are far beyond any type of statistical significance threshold, such as interviews influencing the decision.
Yeah S19 had a fabulous Dartmouth interview too. And was invited to an alumni holiday luncheon and got a ton of attention from his AO - and he was denied. Don’t count your chickens before they hatch!
Based on your previous posts, you seem to be too confident of your value in interviewing applicants. All of us are limited by our own experiences. Here’s an article in The Atlantic a few years ago discussing the futility and irrelevance of alumni interviews.
I agree. We jumped through so many hoops to get S19 to alumni interviews. Two of them were an hour away and in snowstorms and he was denied by one and waitlisted at the other and both went LONG - over two hours - and got his hopes up. D21 hasn’t had any alumni interviews, just some with AOs and I think those are likely more worth it.
If a meet report exists, it’s got the potential to be informative.
Bari Norman. (Most recent link.) Another “former adcom.” Left Barnard 16 years ago, if you count the precollege. Otherwise ended her AO role in 2000. Gets to write something for Atlantic, collects a check. Yeah-huh.
I interview for Stanford. Note that my earlier post did not say Stanford interviews were especially influential. Instead I listed 3 other schools. My earlier posts actually implied the opposite – interviews were less influential at Stanford than most other highly selective private colleges. This partially relates to their unique history, including not starting interviewing applicants until a few years ago.
Skimming through the linked article, it doesn’t appear to state anything I wasn’t already familiar with, nor does it mention your claim about, “For those colleges that require or recommend alumni interviews, it isn’t hard to discern their motivations. It’s primarily a tool to keep alumni engaged.” Instead it highlights MIT as a school where interviews have weight, just as I did stating,
“Certainly there are schools that give weight to the interviews and make them available for virtually every applicant, and that accessibility is crucial if the practice is to have any sort of impact on an application. Take the Massachusetts Institute of Technology, for example.,”
This is similar to my repeated position, which is requoted, “I expect that interviews have different degrees of influence at different colleges, and it’s not accurate to make a sweeping statement. At at least some highly selective colleges, interviews have a noteworthy influence on decisions, like other sections of the application.”
I believe you’ve said previously that you can predict admission outcomes from your interviews. That seems to indicate that you’re confident of your value in the interviews. I agree that the relevance (or irrelevance) of alumni interviews varies by college and I’m well aware that MIT is the most invested in the intervew process among the colleges where the interviews are a regular part of their admission process. But even for them, the value is marginal.