Skip to content

Dhruva's Dumb Website

MIT Invitational 2019

Science Olympiad2 min read

action

This year I had the pleasure of supervising at the renowned MIT invitational (along with Donna and Aditya), where we got to see many talented teams from across the country compete. Here is the exam and key we used, with minor corrections. (I've posted another helpful document at the end of this post).

  • Exam
  • Key
  • Image Sheet

This year's exam was, in my opinion, at least an order of magnitude harder than last year's. This resulted in generally low scores across the board. In addition to the objective difficulty, the exam was also long. This meant that performing well was a matter of good (lucky?) time management as much as it was about actual skill. In my opinion, that's not how an exam should be, and I take responsibility for underestimating the difficulty and length of the exam. It's also only mid-season, so teams are not as prepared as they will be when they compete at state/nationals.

Here are some statistics about the scores of the 71 teams that competed. The exam totalled 126 points, out of which 64 came from section A, 26 came from section B, and 36 came from section C.

Section ASection BSection COverall
Mean34.0%29.6%22.0%30.2%
Standard Dev.17.1%20.8%14.3%14.6%

The score distribution was unimodal, with a very shallow, linear dropoff on the low-scoring end, and a steeper dropoff on the high end. I think this is a strong indication that there were too few easy questions to differentiate less experienced teams. This is definitely something I have been keeping in mind for my UT exams, but clearly there is still room for improvement. On a feedback form that was given (and incentivized with bonus points), many teams expressed sadness/frustration with the difficulty of the exam, which corroborates the conclusion.

Some teams also were disappointed by an apparent lack of diversity in the exam questions. On this, I disagree. The questions actually covered quite a broad range of topics, including many opportunities to showcase both knowledge of theory and prowess in mathematical problem-solving. I suspect that teams were occupied by section A for most of the exam period, which perhaps made it seem like the content areas were badly covered. This boils back down to the exam being too long.

I do think that this exam can be used as a good guide to studying, both for teams that attended and teams that didn't. However, I know that finding good resources and helpful information can be hard. I've written a walkthrough for section B (the section that I wrote) which explains each answer in depth, and refers you to links which contain more information. Aditya may also contribute a more detailed writeup of his section C solutions as well, so be on the lookout for that. I hope you find this helpful and that it can guide your studying a little.

scratch

© 2020 by Dhruva's Dumb Website. All rights reserved.
Theme by LekoArts