Does MMI really correlate with med school performance?

[CavsFan2016]

---

Members do not see this ad.

Just curious, I was wondering if anyone had the data for this.

 

[CharlieBillings]

I was told it correlates well with Step 2 scores, but then again I've also been told Step 2 is not very important compared to Step 1.

 

[NotASerialKiller]

I'm pretty sure the goal is to produce empathetic med students who are comfortable and capable when interacting with other human beings, not to produce med students with higher scores.

 

[Gurkhali]

MMI performance and even regular interview performance is affected by too many factors, a couple being experience and maturity, to say there is any meaningful relation, imo. If you are aware of the issue, you can fix it and it should have no bearing on your eventual performance.

 

[redsox93]

I hope not I'm terrible at MMIs. But I think the majority of that had to do with interview nerves because I felt like I did just as terribly in the rest of my interviews haha

 

[KrombopulosMichael]

My understanding is that MMIs were implemented less for gauging how well a prospective student would perform in medical school and more on whether they have the interpersonal and critical reasoning skills required of a doctor.

 

[efle]

I can't find it, but I once saw a table that showed MMI correlated very well with clinical year grades, significantly better than traditional interviews.

Neither of the two correlated well with preclinical grades or step 1, which as you'd expect were best predicted by MCAT and sGPA

 

[aldol16]

MMIs are correlated with ****tiness of interviews. No, I don't have a reference.

 

[Lucca]

Here are 3 examples from the literature that study MMI as a predictor of academic or professional success in medicine.

From JAMA, a study on McMaster MD students in Canada after MMI implemented:
http://jamanetwork.com/journals/jama/fullarticle/1475189

The 12th reference in the JAMA article above, from Medical Education, discussing relationship between MMI and clerkship and licensing exam performance.
http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2929.2007.02709.x/epdf

The 16th reference from the JAMA article above, from Medical Education, discussing validity of MMI as a predictive tool in med admissions
http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2923.2009.03407.x/epdf


Haven't read them -- only skimmed -- just posting here for those interested.

 

[efle]

^ These appear to all look at performance on the Canadian exams, which also correlate best with the Verbal section of the MCAT iirc. For the US Step exams (again iirc) it's the Bio section of MCAT and there isn't much correlation with any type of interview

 

[Lannister]

I can't find it, but I once saw a table that showed MMI correlated very well with clinical year grades, significantly better than traditional interviews.

Neither of the two correlated well with preclinical grades or step 1, which as you'd expect were best predicted by MCAT and sGPA

I go to a school that does MMIs and I'm pretty sure I saw the same stats.
I'd love to know how I did on my MMIs, I wish my school would tell me.

 

[droidz]

imho - and really just my opinion - McMaster designed the MMI and CASPER exam to make themselves seem more special and elite than they really are.

 

[CavsFan2016]

So is there any evidence from American MD schools that supports MMI performance?


Sent from my iPhone using SDN mobile

 

[Lucca]

So is there any evidence from American MD schools that supports MMI performance?


Sent from my iPhone using SDN mobile

I saw some info on this from OSU in a presentation that was posted on YouTube and a friend shared on FB. I haven't taken the time to search for it, but I remember seeing MMI performance correlated better with clerkship grades than McAT sections and the mcat sections predicted Step scores better

 

[Goro]

The data I've seen is that it's a good predictor of empathy. Only the MCAT is a predictor of med school performance, and a weak one at that.

Just curious, I was wondering if anyone had the data for this.

 

[bluesailing]

around 14:30
high correlation is seen with 3rd-4th year clinical performance, but not with step 1/2.

 

[danimal22]

I think that MMIs have the potential to be pretty reliable if the prompts are written well. I have only had one MMI and I thought the questions they used were both clever and made sense to see how well I can communicate and look at situations from multiple perspectives. Maybe it's just my personality, but I had fun at my MMI. It felt kind of like a game and I thought I gave good responses. I can see how some personality types might have a difficult time with an MMI and may do better in a traditional interview though.

 

[Skiiiiiing]

A lot of the unstructured interviews are a total waste of time; I strongly prefer the MMI because of rater bias and eccentricity.

I had one who wanted to argue with me about the name of the winner of the Nobel in economics was this year. She was wrong, but I didn't call attention to it. Had another where we didn't even talk about anything related to school or medicine or motivation - I have no idea what that guy is going to even going to write.

 

[YeaJeets2]

MMIs are correlated with ****tiness of interviews. No, I don't have a reference.

Just burst out laughing at work