I found an old article that was published by the AICPA and it stated this;
A critical component of
the CPA examination review process is
Preliminary Item Analysis.
During PIA, sub-committee
members review flagged simulations. These
include distributions of specific candidate
responses and the distribution of
answers across the total candidate
pool. We’re mostly concerned with
moderate and high performers, since
that’s where the determination is made
whether a candidate is qualified to
become a CPA. They also review
what are called “biserial correlations,”
which compare performance on simulations with how the
candidates did on the multiple-choice part of the exam. “It’s
expected that candidates who did well on the multiplechoice
questions will do well on simulations,” explained Mr.
Dutz. “During the review, the SMEs view the actual simulation
and the statistics of how candidates responded to the
individual measurement opportunities within the simulation.
Then they conclude that the answer contained in the exam
answer key is correct or whether another response could or
should be considered an appropriate and valid answer.”
So simulations are curved based on the answers of moderate and high performers?
Paper Exam May 2002; AUD 62, LPR, 63, FAR 49, ARE 70
Paper Exam Nov 2003; AUD 72, LPR, 72, FAR 67, ARE 71
CBT 2010; AUD 83 FAR 76 BEC 78 REG 73,(2/28/11 NTS # 446) 75 and DONE!
.