The Predictive Strength of the End of Curriculum Exam

Massey, Scott; Chadha, Rajat; Campbell, Dana; Rodgers, Christine

doi:10.1097/jpa.0000000000000407

Cited by 4 publications

(3 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Before incorporating predictive modeling, we identified at-risk students by using PACKRAT scores, a widely used approach. [32][33][34][35][36] Naturally, many of these students were not enthusiastic to learn that they were considered to be at risk. Because the terms "at risk" or "remedial" might sound harsh, we might instead choose to use terms such as "needing extra support" or "supplemental curriculum.…”

Section: Student Considerationsmentioning

confidence: 99%

The Application of Adaptive Minimum Match k-Nearest Neighbors to Identify At-Risk Students in Health Professions Education

Kumar,

DiJohnson,

Edwards

et al. 2023

J Physician Assist Educ

View full text Add to dashboard Cite

Introduction When learners fail to reach milestones, educators often wonder if any warning signs could have allowed them to intervene sooner. Machine learning can predict which students are at risk for failing a high-stakes certification examination. If predictions can be made well before the examination, educators can meaningfully intervene before students take the examination to reduce their chances of failing. Methods The authors used already-collected, first-year student assessment data from 5 cohorts in a single Master of Physician Assistant Studies program to implement an “adaptive minimum match” version of the k-nearest neighbors algorithm using changing numbers of neighbors to predict each student's future examination scores on the Physician Assistant National Certifying Exam (PANCE). Validation occurred in 2 ways by using leave-one-out cross-validation (LOOCV) and by evaluating predictions in a new cohort. Results “Adaptive minimum match” version of the k-nearest neighbors algorithm achieved an accuracy of 93% in LOOCV. “Adaptive minimum match” version of the k-nearest neighbors algorithm generates a predicted PANCE score for each student one year before they take the examination. Students are classified into extra support, optional extra support, or no extra support categories. Then, one year remains to provide appropriate support to each category of student. Discussion Predictive analytics can identify at-risk students who might need additional support or remediation before high-stakes certification examinations. Educators can use the included methods and code to generate predicted test outcomes for students. The authors recommend that educators use predictive modeling responsibly and transparently, as one of many tools used to support students. More research is needed to test alternative machine learning methods across a variety of educational programs.

show abstract

Section: Student Considerationsmentioning

confidence: 99%

The Application of Adaptive Minimum Match k-Nearest Neighbors to Identify At-Risk Students in Health Professions Education

Kumar,

DiJohnson,

Edwards

et al. 2023

J Physician Assist Educ

View full text Add to dashboard Cite

show abstract

“…3 However, the PA Education Association's (PAEA) End of Curriculum Exam and both Physician Assistant Clinical Knowledge Rating and Assessment Tools are statistically significant predictors of PANCE scores. 4,5 Effective from September 1, 2020, the Accreditation Standards, 5th edition, replaced the previous iteration. 6 These guidelines were established to comprehensively evaluate PA programs based on administration, curriculum/instruction, selfassessment, and accreditation maintenance.…”

Section: Introductionmentioning

confidence: 99%

“…3 However, the PA Education Association’s (PAEA) End of Curriculum Exam and both Physician Assistant Clinical Knowledge Rating and Assessment Tools are statistically significant predictors of PANCE scores. 4,5…”

Section: Introductionmentioning

confidence: 99%

Physician Assistant Program Accreditation Analysis and Survey of Program Directors

Lewis,

Peacock,

Miller

2023

J Physician Assist Educ

View full text Add to dashboard Cite

Purpose To inform physician assistant program directors through citation analysis after implementation of the Accreditation Review Commission on Education for the Physician Assistant (ARC-PA) Accreditation Standards, 5th edition. Methods This research used descriptive statistics, Pearson correlation, and the coefficient of determination to analyze the citations reported by ARC-PA during January 2021 to February 2023. Concurrent first-time taker Physician Assistant National Certifying Exam (PANCE) results were used to determine whether a correlation exists between pass rates and citations. In addition, a survey was sent to each institution’s current program director to investigate leadership concerns and differences between programs placed on a provisional, continued, or probation status by ARC-PA. Results Of the 98 program submissions for accreditation, 13 submissions resulted in a probation status outcome. For these 13 programs placed on probation, 46.2% and 30.8% were cited for being noncompliant with leadership Standards A2.09 and A1.02, respectively. Pearson correlation analysis indicates a significant negative correlation between ARC-PA citations and first-time taker PANCE pass rates (P = .023, 95% confidence interval [CI] = [−0.49 to −0.04]). This is particularly true for programs with continued accreditation (P = .007, 95% CI = [−0.67 to −0.13]) and programs that performed below the 85% benchmark (P = .013, 95% CI = [−0.94 to −0.22]) for first-time taker PANCE pass rates. Although a negative correlation is observed between the number of levied citations and PANCE pass rates, the coefficient of determination does not indicate that the number of citations can predict PANCE pass rates (R 2 = 0.0368). Regarding survey data, 42.86% of programs placed on probation cited institutional support as their biggest program weakness or threat. Conversely, 22.22% of programs with provisional status and 12.5% of programs with continued status reported institutional support as their biggest weakness or threat. Conclusion This research identifies leadership as a deficit of concern associated with programs placed on probation. In addition, a significant negative correlation exists between the number of citations and first-time taker PANCE pass rates—especially for programs with first-time taker PANCE pass rates below 85% or for programs with continued accreditation status.

show abstract