2020 Princeton Invitational

[gillio]

Multiple tests weren't returned after the tournament Is it worth asking for them?

[bernard]

Event Supervisor Review: Detector Building

Devices: Almost all teams competing presented Devices, and most Devices were without construction violations. Common construction violations were not having LEDs and having sensor & wires/cables shorter than 30.0 cm. A few teams had completely concealed their sensors in the process of waterproofing, making them unable to show on their Device that their choice of sensor was legal.

Design Logs: I scored Design Logs very leniently, with half of the teams present to receive all points; however many had "minor" errors or omissions that at other tournaments may earn fewer points. The Design Log should not be overlooked for it contributes a fifth of the total possible points, so I focused on providing useful feedback to each team that presented one.

Issues included (and most were not penalized): only including drawings of the Device (not a photograph); labeling the photograph but not detailing functions; using values that were clearly not voltages (e.g. "7000 V", using the raw analogue input value); plots missing axis labels, titles, and/or units; printing a graph so small or in such poor quality that it could not be read; not defining variables (e.g. using "x" without specifying that it represents a voltage in Volts); defining variables but not specifying units for the values they represent; writing an "equation" that is really an expression (e.g. "123 * V + 456" on its own is not an equation for temperature); and providing a printout of code but not highlighting sections requested in the rules.

Test: As pepperonipi mentioned, I wrote the test to focus on what I thought makes Detector Building a unique event: engineering hardware devices to gather data for developing models. I tested whether students were familiar with using components used for constructing devices (e.g. questions 1-3, 13, 17), understood the science that allows for temperature-sensing with these components (e.g. questions 7-8), and knew how to work with data effectively (e.g. questions 10, 19-24). Historically it is unusual for a Technology & Engineering Committee event to have a written test, but I appreciate that the test is a way of checking that "teams must be able to answer questions regarding the design, construction, programming, and operation of the Device per the Building Policy found at www.soinc.org;" students that did not know how to build their Device would not have performed well on the test (e.g. question 4 is a calculation that teams should have completed when adding LEDs to their Devices).

Most teams left many questions unanswered, and the most commonly unanswered questions were between questions 19-24. Perhaps they were skipped because linearizing data on a calculator can be time consuming or teams were not familiar with linearization (as it is not required for building the Device but still a part of the topics for the Written Test), but these questions were collectively worth many points, and the one team that answered them almost entirely correctly had a much higher Written Test score than the second highest score. Files for the test can be found at scioly.org/tests by searching my username.

Scores: Twenty-five teams earned all possible Design Log points and all possible LED color points. The highest Temperature Accuracy Score was 59.4/60, and the high score for the Written Test was 22.6/30. Earning almost all points for the Design Log and LED colors were necessary for performing well overall. However, neither Temperature Accuracy nor the Written Test appeared to be more important than the other; teams could compensate for a slightly lower Temperature Accuracy Score by earning more points on the Written Test, and vice versa.

Overall: I was impressed with all the Devices I saw and very happy to see almost all teams with Devices. I absolutely love the idea behind the Detector Building event, and it was my pleasure to supervise it for PUSO. I appreciate Chirag from PUSO for working with me on this event leading up to the tournament and acquiring all the materials we used; all the volunteers who helped with evaluating devices and grading tests (including pikachu4919!); and the flexibility of the PUSO organizers with changing the event room the night before (from one with only chair-mounted desks to the room we used) and for being so responsive to our volunteer needs throughout the day.

detectorbuilding_2020_c_princeton_scores.png (52.1 KiB) Viewed 3617 times

[pb5754]

better late than never ig

event reviews:
astro (1): basically a (much) easier version of mit w/o js9. dso section definitely could have been longer/harder tho.
fossils (8): good test, good difficulty. didn't realize that there were 1/2 point deductions for incorrect spelling though oops
geo mapping (8): felt pretty math-heavy. there probably could have been a better difficulty spread but whatever.

[Anonymous15]

Hey everyone!
Looking through this forum, I'm seeing that a lot of teams unstacked (like LM) and others stacked (like WWPN). Why did different teams decide to stack and others decide to unstack-obviously both LM and WWPN are great teams, so I was just wondering the benefits of stacking vs unstacking at a rather competitive invitational like Princeton.

Thanks!

[juicet1n]

Hey everyone!
Looking through this forum, I'm seeing that a lot of teams unstacked (like LM) and others stacked (like WWPN). Why did different teams decide to stack and others decide to unstack-obviously both LM and WWPN are great teams, so I was just wondering the benefits of stacking vs unstacking at a rather competitive invitational like Princeton.

Thanks!

For my school, we decide pretty arbitrarily whether or not to stack or unstack. At invitationals, we don't really mind how the team places as a whole, as we only want to get better as we prepare for states. We are allowed to take 18 people to the state competition (rather than 15), so by unstacking, we have more flexibility with trying out new states pairings (as it's often difficult to fit everyone on a single stacked team).

[pb5754]

When are tests/histograms expected to be released?

[bernard]

When are tests/histograms expected to be released?

Supervisors had until last Friday to submit edits to exams so I'd estimate sometime this week if they heard back from everyone.

[dakota71]

Thank you for sharing the tests. Will the astronomy exams be posted also?

[Adi1008]

Thank you for sharing the tests. Will the astronomy exams be posted also?

After March 7th.

[bernard]

Link to Princeton tests, keys, and histograms.