Science Olympiad at MIT Invitational 2019

[mastersuperfan]

I suppose I might as well do some event reviews.

Anatomy and Physiology: 1/5. I think this is the event that wins "mostly poorly run of the day." The numbering was consistently off (presumably because the whole test was thrown together last minute) and numbering changes were being announced throughout the block. This was even worse at the three-minute stations—anyone else who did A&P will know how painful that kidney model was when you had to refer to a separate sheet to figure out which numbers on the model matched up with which spots on the answer sheet. We also couldn't split up or write on the test, so partners were forced to do the test together. Having ~70 multiple choice that are all "select all that apply" with on partial credit is incredibly frustrating and results in multiple choice scores more often being decided by lucky guesses than on actual skill—plus the fact that many teams didn't even realize they were select-all-that-apply in the first place. The grading was also inconsistent between both our teams, and points were added up wrongly on numerous occasions. I didn't think you could get worse from last year's A&P test, but I guess I was wrong.

Designer Genes: 4/5. An easy test. There was nothing wrong with it (I rather enjoyed some of the more creative questions), but it didn't come with the difficulty that MIT tests are renowned for. The ES was considerate enough to give everyone two copies of the test packet, one for each partner. Wasn't prepared for the select-all-that-apply questions on inherited diseases.

Forensics: 5/5. Solid forensics fare. Nothing new or surprising (aside from the fact that I still have no clue whodunnit), but high-quality and run well. The very small quantities of materials got frustrating sometimes, but I think it was an appropriate challenge—although, is it possible to identify all the plastics with only three pieces? (I didn't even touch them during the competition.) My only big complaint is that one team brought all their bags of powder and fiber samples over to one of the Bunsen burners and basically set up camp there, refusing to let anyone else share it. Perhaps a rule against that could be enforced next year.

I didn't do Chem Lab this competition, but looking at the test, it seems pretty good. Though there isn't much math, a lot of the questions are topically very interesting. I do agree that the heavy emphasis on MO theory might have been a bit too far removed from the event description, but I've seen much, much worse.

I also heard Thermo was run poorly because one of our teams received water at a temperature at 55 degrees, which is outside the allowable range and screwed up the predictions.

[primitivepolonium]

IIRC the Designer Genes event supervisor will be the Nationals supervisor? (I think it's Designer Genes.)

As for MOs: yeah. I was in the unenviable position of writing for incredibly vague rules I'd never competed in. My rationale was that MOs are responsible for a lot of physical properties, but tbh I should have just stuck with flame test colors or something. I also figured that every other exam this season was going to be filled to the brim with a&b calculations so opted for a more qualitative route.

[Unome]

IIRC the Designer Genes event supervisor will be the Nationals supervisor? (I think it's Designer Genes.)

Ah, I thought I recognized the name. That's pretty interesting news.

[mastersuperfan]

I think one focus next year should be on making sure all the grading is correct and double-checked, even if it takes some extra time. After looking through some of our teams' tests, I've noticed an egregious number of instances where entire sections or pages weren't added in calculating the final score—particularly in A&P, Expedes, Forensics, and Water Quality. It's kind of discouraging to know that your placing is substantially lowered below your actual performance due to forces beyond your control...

[Unome]

I think one focus next year should be on making sure all the grading is correct and double-checked, even if it takes some extra time. After looking through some of our teams' tests, I've noticed an egregious number of instances where entire sections or pages weren't added in calculating the final score—particularly in A&P, Expedes, Forensics, and Water Quality. It's kind of discouraging to know that your placing is substantially lowered below your actual performance due to forces beyond your control...

I definitely made an effort to double-check summation on my test after seeing several such errors over the last few years at MIT.

[nejanimb]

I think one focus next year should be on making sure all the grading is correct and double-checked, even if it takes some extra time. After looking through some of our teams' tests, I've noticed an egregious number of instances where entire sections or pages weren't added in calculating the final score—particularly in A&P, Expedes, Forensics, and Water Quality. It's kind of discouraging to know that your placing is substantially lowered below your actual performance due to forces beyond your control...

Was the issue with ED a math error in adding up points on the rubric, or what you think was points someone should have gotten credit for but didn’t? The point totaling is always tougher than you’d think it should be, but this year we had two people add up the points for every test to check it, plus an additional random selection audit during score submissions — hard to know what more to really be done. But, apologies for what I would imagine is a non-zero number of errors that still slip through! Feel free to email me if any questions/errors on that.

If it’s more about points not given for content in the lab, I’d wager it’s more because there’s a difference of perspective in what should count in scoring!

(and yes, giving teams large kitchen knives for this event was an insane idea.)

[Alex-RCHS]

IIRC the Designer Genes event supervisor will be the Nationals supervisor? (I think it's Designer Genes.)

Ah, I thought I recognized the name. That's pretty interesting news.

If this is public info, could someone tell me who this is?

[windu34]

If this is public info, could someone tell me who this is?

Im not sure if Im allowed to release that info, but all the ES names and contact info was emailed to teams that attended MIT Invite.

[mastersuperfan]

I think one focus next year should be on making sure all the grading is correct and double-checked, even if it takes some extra time. After looking through some of our teams' tests, I've noticed an egregious number of instances where entire sections or pages weren't added in calculating the final score—particularly in A&P, Expedes, Forensics, and Water Quality. It's kind of discouraging to know that your placing is substantially lowered below your actual performance due to forces beyond your control...

Was the issue with ED a math error in adding up points on the rubric, or what you think was points someone should have gotten credit for but didn’t?

It was a math error with adding up points on the rubric. In all fairness, though, the calculation mistakes on Expedes were minor, involving only 2 or 3 points (to my recollection). Not nearly as bad as having a 36-point page missed on Forensics...

Hey everyone,

I've posted the MIT Chemistry Lab exam https://drive.google.com/drive/folders/ ... bKiJHNPHt_.

All comments welcome!

I was intrigued by the 15-point math problem on page 11 and decided to try it myself. I'm getting an answer of 0.018 M, but the key says 0.00638 M. I think there may be an error in the first step of the key when [C6H5CO2-] is calculated, because I think it should be 0.290 M instead of 0.00333 M. Although I did the problem in a very different way, I am also getting the key's answer when I use 0.00333 M, so I don't think I'm making a mistake here. Could you check this? (On another note, I noticed that the key assigns point values to solving for [H+], but my solution doesn't require you to find it at all.)

[dvegadvol]

Where will the MIT tests be posted? Anybody know when? Thanks,