Welcome Guest ( Log In | Register )

Reply to this topicStart new topic
> Relay Scores, how I analyzed the old Super Quiz
post Apr 14 2012, 09:58 PM
Post #1


I've been tinkering with various ways to depict my team's Super Quiz success at Regionals '12, with each method focusing on either determining our strength in social studies as a subject or seeing how we do with relay. All of these methods assume a 40-question written SQ test worth 600 points and 5 relay questions worth 80 points each. I'll refer to the written score as W and the relay score as R.

1) Scaled-up written test. This measure converts the written-only portion of Super Quiz to a 1000-point event.

2) Relay-to-written ratio. This measure is the ratio of the number of questions a competitor answered correctly on relay to the number answered correctly on the written portion.
(R/240)/(W/600) or 2.5R/W
A value greater than one suggests that the student is good at relay (and/or got lucky with easy questions).

3) Equalized Super Quiz score. This measure makes each question from the 45 total in the Super Quiz event worth the same number of points for a total of up to 1000 points.
Note that this method may be considered slightly unfair to students in higher categories if students in lower categories were intentionally given easier relay questions.

4) Percent written or percent relay. This measure gives the percentage of a competitor's total Super Quiz points that were earned in written or relay, respectively.
Percent written=W/(W+R)*100%
Percent relay=R/(W+R)*100%
Interpretation: The written portion accounts for 60% of the event's total possible points, while the relay accounts for 40%. This measure does not account for the high weight of relay questions, so if you like that relay questions are weighted, great. If not, multiply the relay-to-written ratio by 100%, and boom, you have this measure with equally weighted questions.
When evaluating these results, consider that a student with a percent written ≥60% performed better in written than in relay.

I'll edit this post with more measures as I devise them, but until then, happy crunching. Hope this helps!
Go to the top of the page
+Quote Post
post Apr 14 2012, 10:35 PM
Post #2


Fancy fancy wink.gif
Go to the top of the page
+Quote Post
post Apr 15 2012, 12:14 AM
Post #3


On a somewhat related note, the average of art, music, econ, science, social sci, and lang and lit test scores probably serves as a good indicator of a Decathlete's success on the new Super Quiz. I plan to use the scaled-up written portion score for social sci, and I'll probably end up multiplying the entire average by the relay-to-written ratio. If you plan to use this method, keep in mind that one very small mistake can make the difference between a 0.89 and a 1.33 relay-to-written ratio (those are my H2 and H3, who got the same written score). I would suggest finding relay-to-written ratios from multiple competitions and averaging those ratios to get a broader representation of Decathletes' relay capabilities.

So, 5) my New Super Quiz predictor value is given by [art+music+econ+sci+L&L+(5/3)*W]*(2.5R/W). This number could be very high or low depending on the relay-to-written ratio, but keep in mind that up to 1000 points are possible before the ratio.
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:


Lo-Fi Version Time is now: 21st May 2018 - 10:13 PM