Does CASPer predict MMI/Interview performance?

[sorrynotsorry]

Hi All, 

Do you think the CASPer result will predict interview performance? Unfortunately I did not meet the CASPer cutoff for Dalhousie OOP. This got me really concerned. 

 I am a slow typer and tend to have typos when I try to speed up   

 

[corgi321]

When did you get that CASPer cutoff e-mail from Dalhousie? And to answer your question, no. CASPer has a really strict format and a massive time constraint. The biggest thing about MMIs is that they are assessing oral communication whereas CASPer doesn't. MMIs involve actually talking to someone, not frantically typing on a keyboard to answer 3 open-ended questions in under 5 mins. As long as you're comfortable talking to interviewers and ready to discuss anything they might throw at you, I wouldn't let your CASPer score be of concern with respect to the MMI or a regular panel interview.

[sorrynotsorry]

6 minutes ago, corgi321 said:

When did you get that CASPer cutoff e-mail from Dalhousie? And to answer your question, no. CASPer has a really strict format and a massive time constraint. The biggest thing about MMIs is that they are assessing oral communication whereas CASPer doesn't. MMIs involve actually talking to someone, not frantically typing on a keyboard to answer 3 open-ended questions in under 5 mins. As long as you're comfortable talking to interviewers and ready to discuss anything they might throw at you, I wouldn't let your CASPer score be of concern with respect to the MMI or a regular panel interview.

I got the email a couple weeks ago and thank you for the assurance. 

[marsaturn20]

I also think that CASPER is NOT a good predictor for the MMI. At least for MMI, you have more time to explain your responses since you have about 8 minutes to talk if I get that timing correctly. So, you definitely have more room to present your ideas. In addition, MMI also has a wide range of topics, whereas for CASPER, it mostly focuses on ethics. 

I know that not meeting the cut-off for Dalhousie was a disappointing news, but OP, it should NOT shake your confidence. Once invited, do a lot of interview practice and immerse yourself in a wide range of topics.

[DrOtter]

If you asked McMaster before the lottery fiasco this year, they'd probs say that CASPer is a GREAT predictor for the MMI/interview, hence its invention and implementation.

But then they actually did statistical analysis last cycle due to the lack of MMI and they ended up just lotterizing most of their incoming class. Goes to show how CASPer is nowhere near as determinant as MMI or interview is. 

[IMislove]

Speaking of the lottery, this opinion article just came out in Medical Teacher https://www.tandfonline.com/doi/abs/10.1080/0142159X.2020.1832206.

[bearded frog]

The original point of Casper was to correlate with MMI, and I think they showed it did when they originally introduced it. You could probably find it on pubmed.

On 10/18/2020 at 8:52 AM, DrOtter said:

If you asked McMaster before the lottery fiasco this year, they'd probs say that CASPer is a GREAT predictor for the MMI/interview, hence its invention and implementation.

But then they actually did statistical analysis last cycle due to the lack of MMI and they ended up just lotterizing most of their incoming class. Goes to show how CASPer is nowhere near as determinant as MMI or interview is. 

I'm not sure how you came to this conclusion. They found that their pre-interview score method, including casper, predicted admissions success so well that everyone does similar on interview and that it was not statically different to assign spots by lottery. Cleary the implication is that if they didn't preselect so well, including casper, there would be much higher interview outcome variability, so basically the opposite of what you claim here....

[DrOtter]

20 hours ago, bearded frog said:

The original point of Casper was to correlate with MMI, and I think they showed it did when they originally introduced it. You could probably find it on pubmed.

I'm not sure how you came to this conclusion. They found that their pre-interview score method, including casper, predicted admissions success so well that everyone does similar on interview and that it was not statically different to assign spots by lottery. Cleary the implication is that if they didn't preselect so well, including casper, there would be much higher interview outcome variability, so basically the opposite of what you claim here....

OP is asking about their interview performance specifically, not the probability of being selected for an interview. So in my reasoning, provided that you're selected for the interview, which depends on GPA and CARS as well as CASPer, the CASPer performance will not be so predictive of MMI performance.

Because if it was strongly predictive, McMaster would have re-adjusted the CASPer weight in their post-invite formula, since MMI would have accounted for 70% of the post-interview score anyway. The fact that they didn't and chose to go with the lottery showed that CASPer is not that predictive of MMI performance, or at least not any more than GPA and CARS anyway. 

[Psych]

On 10/24/2020 at 6:37 PM, bearded frog said:

The original point of Casper was to correlate with MMI, and I think they showed it did when they originally introduced it. You could probably find it on pubmed.

I'm not sure how you came to this conclusion. They found that their pre-interview score method, including casper, predicted admissions success so well that everyone does similar on interview and that it was not statically different to assign spots by lottery. Cleary the implication is that if they didn't preselect so well, including casper, there would be much higher interview outcome variability, so basically the opposite of what you claim here....

McMaster sent out an FAQ to applicants this year where they directly stated that CASPer correlation with the MMI is weak (CASPer correlates with non-cognitive skills but not MMI performance)