Quantum Country second repetition session accuracy measures

These are the accuracy % scores of users’ first review sessions on Quantum Country questions. These numbers are limited to traces with at least 6 months of data, i.e. to skip non-serious users (a quick glance at the unconditioned data showed fairly small differences: lower accuracies overall, slightly different order but same shape of distribution).

https://docs.google.com/spreadsheets/d/1jVM73UQosMUjuuqSOyfxBp4I0oqhYZqXK0w1KIMjgGU/edit#gid=1240834815

Accuracies seem to follow a power law: 0.599x^0.113 (R^2=0.966)

See also:

• Quantum Country in-essay question accuracy measures
• Quantum Country first repetition session accuracy measures

Bottom ten questions, QCVC

0.3379790941
Who has made progress on using quantum
computers to simulate quantum field theory?
0.4432132964
What is the matrix representation of
the $Z$ gate?
0.4819277108
What’s an
example where the target state input to a CNOT gate is
left unchanged by the gate, but the control state changes?
0.5186335404
What is the length of $|\psi\rangle$ in
the Dirac notation?
0.5195530726
What are
three common names for the $\dagger$ operation?
0.5321428571
The net effect of a quantum circuit of CNOT
and single-qubit gates is to effect a ___ operation on the state
space of the qubits.
0.5977961433
What is the matrix representation of
the $Y$ gate?
0.5983827493
What entry
appears in the bottom left corner of the $Y$ gate?
0.6218274112
What is the
matrix representation of the Hadamard gate?
0.6226415094
What’s a quantum circuit that can
compute the NAND of two classical bits, $x$ and $y$?

Compared to first repetition

Nth worst in second repetition was Mth worst in first repetition (bottom 10, QCVC only):

• 1 -> 1
• 2 -> 2
• 3 -> 4
• 4 -> 5
• 5 -> 9
• 6 -> 3
• 7 -> 7
• 8 -> 10
• 9 -> 6
• 10 -> 8

So the second repetition’s worst 10 are the same as the first repetition’s worst 10, except a bit shuffled.

Query

WITH

withTotalTimestamp AS (SELECT *, COUNTIF(sessionID IS NOT NULL AND isRetry IS NOT TRUE) OVER (PARTITION BY cardID, userID ORDER BY timestamp ASC) AS sessionNumber, TIMESTAMP_DIFF(MAX(timestamp) OVER (PARTITION BY cardID, userID), MIN(timestamp) OVER (PARTITION BY cardID, userID), DAY) AS traceDays FROM `logs.reviews` WHERE isRetry IS NOT TRUE),

traces AS (SELECT * FROM withTotalTimestamp WHERE traceDays >= 180 and sessionNumber = 2)
SELECT ANY_VALUE(essayName) AS essayName, cardID, COUNTIF(reviewMarking="remembered")/COUNT(*) AS accuracy, COUNT(*) AS N, ANY_VALUE(question) AS question FROM traces JOIN `logs.latestEssaysCards` USING (cardID) GROUP BY cardID ORDER BY essayName, accuracy