“Failed” biology assessment; very sure I scored above 80%

Alignerr
1

Hi,

My name’s Dr. Alex Gener. I am new to Alignerr. I recently took the Biology assessment, which included pretty basic bread and butter type questions like what you might come across for AP Bio/MCAT/USMLE-type questions. I have a PhD in biomed and recently sat for the new MCAT, so all of the material was fresh in my mind. I also have a background tutoring for USMLE, and have edited review material and am on the editorial board for a journal in my field. As I was taking the Biology assessment, there was a poorly worded question (TestGorilla’s fault). During the test, I raised my concern with them through their chat bot, which funneled the concern through their support system. From TestGorilla’s app side, it shows I completed the assessment, but that they can’t display the results. They suggest reaching out to Alignerr. When I go to my Alignerr app, it says I “failed.” I am VERY sure I didn’t. I have been trying to talk to a actual human from Alignerr, but I haven’t been able to yet; the Alignerr support chat bot has been gate-keeping, which has been frustrating. It finally at least pointed to this link.

So, what gives?

2

Hi @itspronouncedhenner ,

Let’s clarify that you can access your result outside Alignerr via our exam provider TestGorilla, if not as you already did please contact them directly they should be able to provide this information to you.
I checked myself, now, and although it takes time to display your overall assessment score, it did, I won’t disclose your score here but the Alignerr app was correct in the status.

Thank you.

3

Hi,

I can see it says “failed” in Alignerr app, but I’m sure that’s not the case if passing is 80%.

I mentioned I already spoke with TestGorilla folks, and they don’t share results:

IMG_8688
IMG_86881290×2796 218 KB

If I could see the item by item scoring, I could identify errors. From my notes I might’ve gotten one wrong.

4

From what I saw, and again 80% is subjective given this can change anytime.
You were significantly under that.

5

Which doesn’t make sense; I’d like to review the actual questions.

I’ve worked with AI companies before who used incorrect items for their training and lacked appropriate expert oversight. I’ve also corrected editors’ errors and noted a typo on my med licensure exam. So, I am skeptical when my predictions don’t match the expected result.