Science Olympiad at MIT Invitational 2020

[poonicle]

What was with not being able to separate the test or image sheets? Dealing with the problems with the test taking environment was probably harder than the test.

AGREED I really really really wish we could have separated the test or image sheets ahaha but it was the standard Unome test (except *plot twist* no multiple choice?!?)

I was really happy to see the same Disease ES from MIT 2017, 2016, 2015, etc. as well. She writes such long tests which was nice.

[bromothymol]

Anatomy (15): a difficult, well-written test on bones, skin, and muscles! the questions were made to force you to go beyond what you've memmed which I appreciated. good balance of different topics. multiple-select questions were all or nothing, which was just wonderful

Botany (2): we were anticipating for the test to be a lot longer and cover a lot more content. instead the test writer seems to be an enthusiast of potato development and hydroponics. why. plants are fun though. hope to see it as a trial again next year!

Designer (7): as expected, a time crunch with an emphasis on problem-solving. the event supervisors actually gave us 55 minutes and even with that we weren't able to complete the test lol. if I had one item of criticism it might be that the focus on problem-solving came at the expense of not as much molecular genetics? might just be that I spent a lot more time on the problem-solving so it weighs more heavily in my memory now though. good test pogba

Disease (9): this isn't one of my main events but the test definitely kept us busy writing for the whole time. I enjoyed the wide breadth of topics covered but it kinda sucked that the tables were so tiny for holding so many pages of test

XD (1): not much to say except that it was nice to see a topic that hasn't been beaten to death. event was run smoothly and we were given 5 extra minutes for cleanup at the end which was appreciated. the smell of the barbecue sauce and ketchup though, please never again

overall a very fun experience, met some cool people, took some hard tests, went to a boba shop that looked and felt like a cursed image 10/10

[Unome]

What was with not being able to separate the test or image sheets? Dealing with the problems with the test taking environment was probably harder than the test.

AGREED I really really really wish we could have separated the test or image sheets ahaha but it was the standard Unome test (except *plot twist* no multiple choice?!?)

I was really happy to see the same Disease ES from MIT 2017, 2016, 2015, etc. as well. She writes such long tests which was nice.

It definitely would have been nice, but unfortunately printing costs tend to put that out of reach, especially for a long test at a large tournament.

I intentionally didn't put any multiple choice, for several reasons. First, multiple choice takes a very long time for me to write. Second, I wanted to challenge myself to get a good variety of question difficulties without using MC as a crutch for easy questions. Third, I wanted to avoid a situation where teams were actively avoided the free response sections because they didn't think it was worth it (which seems to have worked, I had relatively few teams skipping sections vs. last year).

A lot of conceptual stuff tends to get repetitive when I write. Standalone complex conceptual questions are difficult for me to write, which is why I tend to push them into being subsets of other problems. Incidentally, this is probably why I'm so much faster at writing ID event tests than any other event, since the ID event model strongly favors integrated question sets.

[pb5754]

What was with not being able to separate the test or image sheets? Dealing with the problems with the test taking environment was probably harder than the test.

AGREED I really really really wish we could have separated the test or image sheets ahaha but it was the standard Unome test (except *plot twist* no multiple choice?!?)

I was really happy to see the same Disease ES from MIT 2017, 2016, 2015, etc. as well. She writes such long tests which was nice.

It definitely would have been nice, but unfortunately printing costs tend to put that out of reach, especially for a long test at a large tournament.

on the other hand, the astro test was longer and we could write on it

(will write reviews later)

[LiteralRhinoceros]

What was with not being able to separate the test or image sheets? Dealing with the problems with the test taking environment was probably harder than the test.

AGREED I really really really wish we could have separated the test or image sheets ahaha but it was the standard Unome test (except *plot twist* no multiple choice?!?)

I was really happy to see the same Disease ES from MIT 2017, 2016, 2015, etc. as well. She writes such long tests which was nice.

It definitely would have been nice, but unfortunately printing costs tend to put that out of reach, especially for a long test at a large tournament.

Re-stapling should not be a problem considering you were stapling the answer sheets.

[windu34]

Event Supervisor Review: Circuit Lab
First off, I would like to say that I had a great time meeting many of you competing yesterday. The test and labs were very challenging and Asher and I were very happy to see competitors successfully solve some of the harder problems on the test. I hope that all that competed and those of you that will take the exam in the future will use it as a study resource and a tool to help you discover some important engineering skills as well as some interesting circuit-related concepts and applications!

Written Exam: Over the course of the past several months, Asher, Megan, and I have been working together to write several circuit lab exams for different competitions. Often times, we would find that we wrote a problem or two that we felt were too hard to be appropriate for most competitions. We saved those problems and put them on the MIT exam. The exam that you all took was definitely one of the most challenging exams in terms of time constraint and problem difficulty that I would expect to see this season. There were 1-2 pages too many on that exam and in most cases, I would have stopped earlier: but this exam was for MIT, and we wanted our exam to live up to the reputation. That said, I don't think that any of the problems were a huge stretch from the rules and probably represent the most difficult problems that could be written for each of the topics on the rules.
As always, here is the test and key: MIT 2020 Circuit Lab Test and Key

Labs: We spent a lot of time thinking about what an interesting, yet feasible lab would look like. The classic battery and resistor labs with "black box" measurements seemed too boring, as did the "draw a circuit diagram of the battery-resistor circuit". We wanted our labs to be more interesting and representative of what you might see taking a university electrical engineering course. The labs ended up being a "see, then do" type of format with competitors first analyzing a circuit that contains the fundamental aspects on a breadboard (op-amp circuit and logic circuit), drawing the given circuit (simply looking at what was built on a breadboard and drawing a schematic), and then building (or adding on) an additional circuit representing a slightly more advanced concept (Lab 1: Adding a non-inverting op-amp to increase voltage output, Lab 2: OR gate observation, building a 3-input NOR gate). We also wanted to minimize the movement of students without sacrificing the quality of the lab experience. We decided the best option would be to make the labs portable and purchased DC breadboard supplies with 9V battery inputs for circuits to be built on individual breadboards. I was very happy with the way the lab setup worked logistically, and if we had had more volunteers, the resetup time of 10 minutes would have been perfect. In the future, I think I would try to decrease the amount of activities in the lab section while maintaining the weighting. I think it is important for students to have more of an ability to work with real components and not just solve theoretical problems on paper.

Observations: Asher and I have worried that teams would excel easily at many components of our exam since we have written and released our past exams. This didnt seem to be the case and despite seeing teams with UT Austin and Boca/NT tryout exams in their binders (which is totally legal btw), we didnt see any teams feeling the need to "refer" to them to help solve problems on the MIT test, which we have striven for and have done our best to avoid repetition (other than in format). A handful of teams found a loophole through Lab 2 that didnt require transistors to build the circuit using transistors as we had intended.

Scores: While I was slightly disappointed to not see any teams breaking 200 pts, I wasn't very surprised given the difficulty of the exam and time in the season. While the scores werent as high as I had hoped, the distribution was still pretty good and I think teams were separated fairly well. Stats: High: 153, Mean: 62, St. Dev: 39

ScoreGraphs.PNG

[xiangyu]

Anyone know the Wright Stuff and Boomilever Top Scores?

[pepperonipi]

Detector (32nd): Test was pretty good in my opinion. Good length, and the questions were reasonably hard. The setup of the calibration water was pretty well done and the final testing area was also well done. No major complaints for sure.

Ornithology (5th): Long test as usual for an MIT event. I thought the questions covered a good range of topics in the event and nothing was out of the ordinary. The setup maybe could have been a little nicer (moving between stations was sometimes a little challenging with the amount of chairs and how close the tables were), but besides that it was great event!

Overall (4th): Great tournament as usual for MIT! Really enjoyed seeing some of our school's alumni and meeting OSS! Thanks to all who put in so much work to make the tournament great!

[Unome]

AGREED I really really really wish we could have separated the test or image sheets ahaha but it was the standard Unome test (except *plot twist* no multiple choice?!?)

I was really happy to see the same Disease ES from MIT 2017, 2016, 2015, etc. as well. She writes such long tests which was nice.

It definitely would have been nice, but unfortunately printing costs tend to put that out of reach, especially for a long test at a large tournament.

Re-stapling should not be a problem considering you were stapling the answer sheets.

The majority of the answer sheets were heavily damaged from this already. I already had to replace several image sheets and test packets after they became too broken down to read some of the questions or images in necessary detail. I don't understand how teams treat their papers so badly, but it does happen to an alarming extent. I've tried paperclips in the past but that hasn't worked out too well - usually we lose a bunch of them.

Regarding whoever mentioned Astronomy, their Image Sheet was a single page, while mine was 7 pages - such is the difficulty of a mapping-based event. I've considered printing test packets for every team, although I don't know how much that would change things since my test packets are generally very short anyway.

If anyone does have suggestions I'd definitely like to hear them - this has been an ongoing problem that I haven't found a good solution to.

[sciolyperson1]

Boomilever (14th) - Pretty standard. Bernard supervised the event and the event was run smoothly and efficiently. They had two autoloaders (vs. 1 from last year). (10/10)

Gravity Vehicle (1st) - The supervisors were slow to check people in, but the event was run without any major error. The track had a good smooth surface and a great location away from any major moving traffic. They should've enforced the no taking pictures rule a lot of people have vids and pics of my device now rip. (9/10)

PPP (12th) - the setup was excellent, best I've seen all season. There were two launchers set up, one for practice, another for official launches. Since we've never had the opportunity to test in any height taller than 18 ft, it gave us a chance to see if our parachute would actually deploy in a 55 ft ceiling (it didn't, rip, we used our LISO backup parachute). (10/10)

LMMM (13th) - The test was setup in stations and 15 minutes were given to complete the "on-site meme". The proctor was the one of the two MIT event directors and their team ran the event smoothly and efficiently. Test was really hard lol, although we made a 18k word cheatsheet just for the event we only used it for 4 of the stations of the test. (whatever meme number /10)

WIDI (9th) - All the desks were arranged in a circle and facing outwards, so that nobody could look at one another's structures. The structure was long and not just difficult to interpret, but difficult to build (a lot of balancing involved). (9/10)

Overall (3rd) - It was great to meet many other people from other schools. Those water bottles that they gave out were really nice lol. There weren't any delays with awards or events, which was great. I'm surprised as to how well we did (our team was supposed to be an unstack according to captains, but it ended up being a 50% stack). Excited to come back next year - hopefully top 10 teams get 2 teams again just like they did for teams this year. (10/10)