Both question papers consist of multiple choice questions (MCQs) and extended matching items (EMIs). The balance of the two types of questions will vary but will be approximately two-thirds MCQs and one third EMIs.
MCQs are one of five, single best answer type. This type of MCQ comprises a question stem, which is usually one or two sentences long but may be longer. The question stem is followed by a list of five options. Candidates should choose the single best option that fits the question stem. One mark is given for the correct answer.
In the EMIs, candidates will be orientated by being given a theme for each set of EMIs. This is followed by an option list, set out in a logical order, followed by a lead-in statement explaining what the candidate is being asked to do. Finally, the vignettes are given and the candidate asked to choose one or more best options from the option list. One mark is given for each correct answer.
Candidates are advised to attempt all questions. No marks are deducted for incorrect answers.
A standard is set for each paper using an absolute criteria. This means that there is no pre-determined number of candidates who pass or fail. Instead, candidates pass or fail according to their own performance and not on how well they perform in relation to the other candidates.
Currently the Modified Angoff method is used to set the pass marks for Papers A & B.
In this procedure, the Standard Setting Panel which is made up of Subject Matter Experts (SMEs), i.e. trainers who are familiar with psychiatric training, for example College tutors, educational supervisors, and recently appointed ST 4 trainees, are appointed by the Examinations Sub-Committee (ESC). They first have a group discussion about the characteristics of a borderline candidate, i.e. a candidate who is minimally competent in the required standards, and then individually evaluate each question and estimate the proportion of these minimally competent candidates that would answer the question correctly.
For each question, the SMEs' assessments are averaged. Another round of estimates is made, allowing each SME to revise their scores on the basis of actual candidate performance statistics. This safeguards against large deviations in pass rates whilst maintaining the defensibility of criterion referencing. Criterion referencing means that every candidate who reaches a required standard will pass, rather than being relative to other candidates' performance. The pass mark is calculated as the average of all question level scores.
Finally, Standard Error of Measurement (SEM) is applied to this score to account for error around the cut point, increasing confidence that the candidate has reached the required standard. SEM is calculated using the number of items within each paper and the standard deviation of candidates' scores for that paper. It is therefore different for each paper. The final pass mark is calculated as the Angoff derived pass mark plus 2 x SEM, and approved by ESC.