New Scoring Model FAQs
If you have questions not answered here, please send an email to email@example.com. Please allow 48 hours for responses.
ABOG is continuously working to improve the fairness, reliability, and validity of the certification process. The new model accounts for two major variables:
- item difficulty (how hard a question is) and
- examiner severity (how hard an individual examiner grades).
The new model also aligns with the Standards for Initial Certification and Organizational Standards set by the American Board of Medical Specialties (ABMS).
In April 2020, ABOG will be implementing a new scoring model for grading the Subspecialty Certifying Examination. The new process requires statistical analysis of the grades, resulting in a four-week period post-exam to accurately analyze and then release the results.
The new model is called the Multi-Facet Rasch Model. It is a statistical measurement model that can account for some factors that influence the outcome of a candidate's performance on an exam.
No. For many years, ABOG has used a Compensatory Model for the Certifying Examination. This model generates one grade per section of the examination and allows a candidate who does poorly in one area to pass the examination if they do well in other areas. This model has proven to be a highly valuable tool for evaluating Certifying Examination candidates. However, ABOG is continuously seeking to improve processes that exemplify high standards.
Key points of the Compensatory Model (prior grading model used):
- It is based on a holistic system, where a candidate can fail one domain and still pass the full exam.
- It employs a single grade per team of examiners.
- It does not account for item difficulty.
- It does not account for examiner severity.
Key points about the Multi-Facet Rasch Model (new measurement mode):
- It is a statistical model requiring analysis of data.
- Each examinee's score is determined by all grades from each examiner.
- It does account for examiner severity.
- It does account for item difficulty.
- Because more time is needed to accurately conduct the mathematical analysis used to determine the cut score, final scores will be released to candidates no more than four weeks after the Subspecialty Certifying Examination.