# How are CPA Exam Simulations Graded?

Have you ever wondered how the CPA Exam is graded? If you have, your logical next questions were, “How are CPA Exam multiple choice questions graded? How are CPA Exam simulations graded?” After doing some digging, you may come across the weights for the multiple choice portion versus the non-multiple portion. You also may come across the notion of the Item Response Theory (IRT) that the AICPA uses to evaluate your responses.

## CPA Exam Simulations Grading: Item Response Theory

According to a Journal of Accountancy article, How the CPA Exam is Scored, “IRT is a class of mathematical models used for exam development and analysis, making it easier and more efficient to compare candidate scores when they are based on exams that have different questions.” The articles goes on to say, “Three statistics are used to describe the questions: difficulty—whether the question is easier or more difficult for candidates; discrimination—how well the question differentiates between more able and less able candidates; and guessing—the chances of candidates answering the question correctly just by guessing.” This method of evaluation seems pretty straightforward, especially for multiple choice questions.

However, how does the IRT work for task based simulations? After digging further, the Candidate Bulletin mentions, “Responses to the questions asked in the task-based simulations will receive credit when you provide a correct answer or complete a task correctly.”  Additionally, the AICPA’s How is the CPA Exam Scored? indicates, “The task-based simulation estimate is mapped to a score on a task-based simulation scale.” In other words, the IRT estimate is mapped to a scaled score, which would be the 40% weight. Likewise, this also seems to make sense.

## CPA Exam Simulations Grading: Questions Asked

Nevertheless, you as an astute, aspiring CPA may have some additional questions. First of all, you may be interested in knowing whether there is partial credit within a given simulation. By extension, how are carry through errors evaluated? Second, you may want to know if each simulation has a similar weight. For example, if there are seven simulations, does each simulation take on approximately one-seventh of the total weight? Third, how does the IRT evaluate individual responses to simulations in terms of difficulty, discrimination, and guessing?

In anticipation of your questions, I brought out my inner Nancy Drew, my inner Sherlock Holmes, my inner Scooby Doo and asked the AICPA more or less these questions.

## CPA Exam Simulations Grading: Response Received

I received the following response from the AICPA:

Please note the following response to your questions about the scoring process for the CPA Exam.

The CPA Examination uses a scoring algorithm called Item Response Theory (IRT); scores are not derived by simply counting the number of correct responses. For each section, answers to the multiple-choice questions across the 3 testlets are used as a single collection of responses to determine the single score for the MCQ portion of the exam section. Similarly, for the AUD, FAR, or REG section, answers to the questions across the 6 or 7 TBSs are used as a single collection of responses to determine the single score for the TBS portion of the section. The individual tasks are not weighted.

For BEC, each of the written communication tasks is scored individually and the single written communication score is then obtained as the average of the individual essay scores.

Thank you for contacting the AICPA. We appreciate your time and look forward to serving you in the future.