公衛學會五十周年紀念專文二：台灣「公共衛生核心課程基本能力驗」的發展歷史與回顧分析

Taiwan Public Health Association 50th Anniversary Commemorative Article II: development history and retrospective analysis of“The Credential Examination on Public Health Core Course Competencies”in Taiwan

目標：公共衛生核心課程基本能力測驗（簡稱：核心能力測驗）從民國98年至109年舉辦了十二年。本研究目的欲藉由核心能力測驗發展的回顧（包括題庫建置過程），分析歷年報名與通過率，及試題難易度與鑑別度，以為未來發展強化核心能力課程之參考。方法：本研究使用文獻回顧台灣公共衛生學會舉辦之核心能力測驗的歷史發展，檢閱由台灣公共衛生學會提供之試務發展過程的文本與資料，並以次級資料的分析方法，（生物統計、流行病學、社會行為科學、環境與職業衛生、衛生行政與管理）的測驗結果。本文利用正確率、難易度、鑑別度分析歷年的考試結果。使用統計軟體為R（版本4.0.4），統計方法包括次數分配與百分比、皮爾森相關係數、變異數分析、無母數變異數分析與卡方檢定。結果：98-109年之報名人數為4,225；2,183人通過，佔未缺考者的60.4%。自99年起五科全採選擇題，因而以99-109年資料進行試題難易度與鑑別度分析。結果顯示每年考試通過率約在六成上下（55.6%~71.5%）。其中，生物統計與環境與職業衛生的平均正確率（標準差），分別為0.54（0.19）及0.51（0.26），略低於其他科目如流行病學的0.56（0.22）、社會行為科學0.67（0.25）、衛生行政與管理0.58（0.25）；但生物統計的鑑別度0.39（標準差0.17）則較其他科目（鑑別度落在0.14~0.31之間）顯著較好（p<0.001）。若比較試題的試前認定難易程度（容易、適中、困難）與試後考生填答的正確率（高、中、低），這兩者的分布在生物統計沒有太大關係，但這兩種分布在其他科目則比較相符（p<0.001）。結論：公共衛生核心能力測驗之通過率穩定，但不同科目之間的難易度及鑑別度有明顯不同，未來或許應該檢驗加強某些特別科目，或是發展參考教材。

Objectives: The Credential Examination on Public Health Core Course Competencies (or the Core Competency Exam) was conducted annually in Taiwan for 12 years, from 2009 to 2020. This study aims to analyze the initiation and development of the exam, encompassing the creation of the questions database and evaluation of pass rates, item difficulty, and the discrimination index, and provide directions for enhancing core course competencies. Methods: This study sought to understand the history of the exam, its development, and its annual conduction by reviewing documents provided by the Taiwan Public Health Association. Additionally, the exam questions on five subjects (Biostatistics, Epidemiology, Social and Behavioral Sciences, Environmental and Occupational Health, and Public Health Policy and Administration) were analyzed to determine their accuracy, difficulty, and discrimination ability. The statistical analyses included calculation of frequencies, percentages, and Pearson’s correlation coefficients, analysis of variance, nonparametric analysis of variance, and chi-square tests. Results: A total of 4,225 individuals registered for the exam between 2009 and 2020, and 2,183 of them passed, which was approximately 60.4% of the non-absentees. From 2010, only multiple-choice questions were used for all five subjects. Therefore, item difficulty and the discrimination index were analyzed for the data collected from 2010 to 2020. The average pass rate was approximately 60% (range: 55.6%–71.5%). The participants scored lower in Biostatistics (average accuracy and standard error across problems: 0.54±0.01) and Environmental and Occupational Health (0.51±0.01) than in Epidemiology (0.56±0.01), Social and Behavioral Sciences (0.67±0.01), and Public Health Policy and Administration (0.58±0.01). However, the discrimination index for Biostatistics (0.39±0.01) was significantly higher (p < 0.001) than that for the other four subjects (range: 0.14–0.31). The before-test problem difficulty (easy, fair, or difficult) and the after-test problem accuracy (high, medium, or low) were significantly similar (p < 0.001) for all subjects except Biostatistics. Conclusions: The pass rate of the Core Competency Exam was found to have been stable. However, the difficulty level and item discrimination index varied among the five subjects. Focusing on specific topics and developing reference textbooks for these subjects may be necessary.