Jump to content
Premed 101 Forums

High tech medical exam cheats busted


Skim

Recommended Posts

In the extreme, you may be theoretically correct, however, for all practical purposes, it is a definite improvement. And it levels and improves the playing field in the selection process. Mac did their homework, just as they did with the MMI, before implementation.

 

Perfection does not exist.

Link to comment
Share on other sites

  • Replies 87
  • Created
  • Last Reply

As someone who spent time designing animal cognition studies and spent way too much time taking study design courses, this is always what kills you: what trait are you looking to predict, why is it beneficial, how you define it qualitatively, what is the construct validity of your definition, how you objectively measure the construct, even if there is construct validity to your definition of trait x, is correlation not caused by some other co-correlatory predictive factor that you may or may not be taking into account.

 

The more psychology I've read in my life the less respect I've had for the objectification of certain aspects of psychology, in the end there are simply some things that don't have variables in a linear regression model with high predictive power, there's also problems with co-predictivity of variables (which overweighs certain variables or even tests), the omitted variable bias (leaving out an important predictive variable increasing the weight of variable x), and a whole bunch of other problems with tests like these.

 

In the end, I think they should just have a criterion, take these classes, get above this GPA in all or a certain number of years, meet these MCAT cutoffs, interview cutoffs, and then we'll draw your name out of a hat, because in real life, that's how accurate some psych tests are... anyhoo!

 

 

My answer to that is to then use some standardized test (possibly the GRE) as well as looking at GPA- preferably one that is 1) correlated with some desired applicant trait or outcome

Link to comment
Share on other sites

The great things about stuff like the MMPI (Minnesota Multiphasic Personality Inventory) and I'm assuming CASPER is you don't know what the optimal answers are, there are often so many questions that have no obvious ties to what you think the characteristic the question your answering is predictive off (certain visceral pain for anxiety etc.), plus validity scales, that you can't really cheat, plus, you really don't know what "kind" of person they're looking for anyways.

 

I haven't taken CASPER but I believe it's possible to have someone else sit and do it for you? If so, how does the implementation of CASPER ensure authorship authenticity?
Link to comment
Share on other sites

I haven't taken CASPER but I believe it's possible to have someone else sit and do it for you? If so, how does the implementation of CASPER ensure authorship authenticity?

 

I've heard it's possible to track the pattern of your keystrokes, which is like a fingerprint - everyone's keystroke pattern is unique. If the pattern differed from question to question, they would know it's not the same person...

Link to comment
Share on other sites

if it's online there's definitely validity (congruency) scales to make sure only one person is doing the test.

 

I've heard it's possible to track the pattern of your keystrokes, which is like a fingerprint - everyone's keystroke pattern is unique. If the pattern differed from question to question, they would know it's not the same person...
Link to comment
Share on other sites

I've heard it's possible to track the pattern of your keystrokes, which is like a fingerprint - everyone's keystroke pattern is unique. If the pattern differed from question to question, they would know it's not the same person...

 

if it's online there's definitely validity (congruency) scales to make sure only one person is doing the test.

 

Nevertheless, it is possible to have, say, a med student do the entire CASPER for you and no one would know. When CASPER was first announced, I expected Mac to make everyone write it in a controlled environment like the MCAT and was surprised at the relatively lax regulations around an item that constitutes 46% of your admission score.

Link to comment
Share on other sites

The great things about stuff like the MMPI (Minnesota Multiphasic Personality Inventory) and I'm assuming CASPER is you don't know what the optimal answers are, there are often so many questions that have no obvious ties to what you think the characteristic the question your answering is predictive off (certain visceral pain for anxiety etc.), plus validity scales, that you can't really cheat, plus, you really don't know what "kind" of person they're looking for anyways.

 

the MMPI for med school admissions - I love it!!

 

If they could use the MMPI to measure something like "Skills in Baseball" using questions such as "are you a bathtaker or shower-taker?" (which they do) they should develop a "Skills in Medicine" scale as well. That would be more fun than studying Claisen condensations and refraction rules for lenses for the MCAT.

Link to comment
Share on other sites

Nevertheless, it is possible to have, say, a med student do the entire CASPER for you and no one would know. When CASPER was first announced, I expected Mac to make everyone write it in a controlled environment like the MCAT and was surprised at the relatively lax regulations around an item that constitutes 46% of your admission score.
Well, to be fair, Mac was planning on testing applicants' keystroke "fingerprint" during the interview (like as an MMI station), but that didn't happen. I wouldn't be surprised if a computer MMI station was introduced to the interview process in the future.
Link to comment
Share on other sites

Nevertheless, it is possible to have, say, a med student do the entire CASPER for you and no one would know. When CASPER was first announced, I expected Mac to make everyone write it in a controlled environment like the MCAT and was surprised at the relatively lax regulations around an item that constitutes 46% of your admission score.

 

I've always assumed that in an ideal world, Mac would invite all applicants to the MMI. Due to time, space, and monetary constraints of administering any sort of evaluation to the thousands of applicants, they decided to use CASPER as a sort of "do-it-at-home MMI" as an initial screening tool for applicants.

Link to comment
Share on other sites

Well, to be fair, Mac was planning on testing applicants' keystroke "fingerprint" during the interview (like as an MMI station), but that didn't happen. I wouldn't be surprised if a computer MMI station was introduced to the interview process in the future.

 

Yeah... I'm pretty sure that was just a scare tactic.

Link to comment
Share on other sites

Well, to be fair, Mac was planning on testing applicants' keystroke "fingerprint" during the interview (like as an MMI station), but that didn't happen. I wouldn't be surprised if a computer MMI station was introduced to the interview process in the future.

 

What an excellent idea! And fool proof.

Link to comment
Share on other sites

Yeah... I'm pretty sure that was just a scare tactic.
I was talking to Wendy on interview day and she said that it was all planned, but they basically couldn't get their act together since it was the 1st year of CASPer. She was the one who told me that it may happen in the next few year. *shrugs*
Link to comment
Share on other sites

What an excellent idea! And fool proof.

 

Sarcasm?

 

Unless you are using the same keyboard I assume there would be pattern difference as your fingers wouldn't be used to a different layout. For example, I can type well on a toshiba PC, not as well on another brand, and Terrible on a MacBook.

 

Is there any studies to back this idea up across different computer models?

Link to comment
Share on other sites

Sarcasm?

 

Unless you are using the same keyboard I assume there would be pattern difference as your fingers wouldn't be used to a different layout. For example, I can type well on a toshiba PC, not as well on another brand, and Terrible on a MacBook.

 

Is there any studies to back this idea up across different computer models?

 

Not sarcasm. If CASPER becomes an MMI station, there is no possibility of cheating - it will be absolutely fool proof. Only one known applicant in the room with a computer and a Mac representative.

Link to comment
Share on other sites

Not sarcasm. If CASPER becomes an MMI station, there is no possibility of cheating - it will be absolutely fool proof. Only one known applicant in the room with a computer and a Mac representative.
I think you may be misunderstanding the point of having a CASPer-like station as part of the MMI. It's not that they're checking to see that you're thinking on your own (as the regular MMI does that). It would be to test your typing signature. And that's not foolproof, as someone else mentioned, there are a lot of factors like keyboard, etc. to consider.
Link to comment
Share on other sites

Sorry but I am still missing the point. Mac wants to ensure that the applicant is actually responding, correct? Well, if the applicant responds on Mac premises, say as the last MMI station, answering the Casper questions, then Mac knows absolutely that a named applicant has answered Casper and in isolation. The keyboard signature than becomes irrelevant. That is why I said they would have devised a fool proof system (in other words, they would not then be ascertaining your signature, rather Mac would be witnessing in real time you answer Casper without any aid whatsoever).

Link to comment
Share on other sites

Sorry but I am still missing the point. Mac wants to ensure that the applicant is actually responding, correct? Well, if the applicant responds on Mac premises, say as the last MMI station, answering the Casper questions, then Mac knows absolutely that a named applicant has answered Casper and in isolation. The keyboard signature than becomes irrelevant. That is why I said they would have devised a fool proof system (in other words, they would not then be ascertaining your signature, rather Mac would be witnessing in real time you answer Casper without any aid whatsoever).
The point of CASPer is to filter out people pre-interview. It's like an MCAT or GPA cutoff. CASPer allows them to evaluate applicants and pick the top 500 (or however many) to interview.

 

At the interview stage, Mac doesn't need to see you answering a CASPer question without any aid. They do that by evaluating you doing a bunch of verbal MMIs. At this point, the only thing to gain from doing a CASPer question is watching how you type, not what you type. The "what" component has been resolved with the MMI. The "how" component is seen through the keyboard signature which would then be compared with the pre-interview CASPer to determine if you were really the one who answered the questions. The keyboard signature absolutely cannot be irrelevant, as that's the entire point of the typing station and is the aspect that Mac is trying to test.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...