校務研究：教育場域數據分析實作

臺北科技大學 114-2 學期課程，課號 360358。

課程大綱

校務研究是教育數據分析在教育場域的應用，在真實的教育場域中應用所學的教育數據分析技術，並針對所服務的教育機構，提供有科學意義且淺顯易懂的數據分析報告，以支持教育機構進行校務運作與教學活動的改善。本門課程首先讓學生認識各職能部門的核心項目及衍伸的校務研究議題；其次瞭解教育場域中因應各項校務與教學活動所出現的數據結構類型；接著教授學生使用教育場域中的真實數據，執行包括數據清理、數據調處、數據水平與垂直整併等步驟，以獨立生成有意義的數據結構；最後則聚焦在教育方案的效果分析，以及將數據分析資料視覺化，輔以淺顯易懂的數據解讀報告，增加學員對於數據分析結果的瞭解。 Institutional Research (IR) serves as the vital intersection of data science and higher education. In this field, researchers leverage advanced analytics to transform complex data into accessible, evidence-based reports that guide institutional decision-making and enhance pedagogical effectiveness. This course provides a comprehensive foundation in IR, beginning with an overview of university administration and the specific research priorities of various campus offices. Students will then explore educational data structures and develop essential skills in data manipulation and cleaning. The final phase of the course focuses on practical application, training students to conduct program evaluations, create compelling data visualizations, and perform rigorous statistical analyses with meaningful result interpretation.

課程進度

week 1: 課程介紹與R環境設定Introduction to R & IR What is institutional research and how can R help us answer educational questions? 校務研究是什麼？R如何幫助我們回答教育問題？ week 2: 校務研究資料來源與資料讀取Data Sources & Reading How do graduation rates vary across different types of higher education institutions? 不同類型的高等教育機構畢業率如何差異？ week 3: 獨立研究：數據清理與整理Data Wrangling（無實體課） Which states have the largest achievement gaps between different student subgroups? 哪些州在不同學生群體之間有最大的成就差距？ week 4: 數據視覺化 I - ggplot2基礎Data Visualization I What is the relationship between school poverty rates and student performance? 學校貧困率與學生表現之間有什麼關係？ week 5: 數據視覺化 II - 進階視覺化Data Visualization II How has state-level education spending changed over the past decade, and where are the geographic disparities? 各州教育支出在過去十年如何變化？地理差異在哪裡？ week 6: 描述統計與探索性數據分析Descriptive Statistics & EDA How does student digital literacy vary by country and student socioeconomic background? 學生的數位素養如何因國家和社經背景而異？ week 7: 假設檢定與推論統計Hypothesis Testing Do charter schools perform significantly better than traditional public schools after controlling for student demographics? 在控制學生背景後，特許學校是否顯著優於傳統公立學校？ week 8: 迴歸分析 I - 線性迴歸Linear Regression What factors predict student mathematics achievement? How much variance can we explain? 哪些因素預測數學成就？我們能解釋多少變異？ week 9: 期中報告 - 撰寫資料視覺化分析報告 week 10: 迴歸分析 II - 多層次模型Multilevel Models How much of student achievement variation is due to differences between schools vs. within schools? 學生成就變異有多少來自學校間vs.學校內差異？ week 11: 因果推論 I - 倍差法與綜合控制法DID & Synthetic Control Did the state's class size reduction policy causally improve student achievement? 該州的小班級政策是否因果性地提升了學生成就？ week 12: 因果推論 II - 斷點迴歸與傾向分數配對RDD & PSM What is the causal effect of receiving remedial tutoring on student outcomes? 接受補救教學對學生成果的因果效應是什麼？ week 13: 地理空間分析 (GIS)Geospatial Analysis Where are the "education deserts" - areas with limited access to high-quality schools?「教育沙漠」在哪裡？哪些地區缺乏高品質學校？ week 14: 社會網絡分析 (SNA)Social Network Analysis How do friendship networks influence student academic performance? Who are the most influential students? 友誼網絡如何影響學生學業表現？誰是最有影響力的學生？ week 15: 自然語言處理 (NLP)Text Mining & NLP What are students' main concerns and sentiments about online learning based on their course evaluations? 學生對線上學習的主要關注點和情感態度是什麼？ week 16: 代理人基礎模型 (ABM)Agent-Based Modeling How does school choice policy lead to racial/socioeconomic segregation? Can we simulate the long-term effects? 學校選擇政策如何導致種族/社經隔離？我們能模擬長期效果嗎？ week 17: 互動式儀表板與資料產品Shiny Dashboards How can we create an interactive dashboard to monitor student enrollment trends and predict future demand? 我們如何創建互動式儀表板來監控學生入學趨勢並預測未來需求？ week 18: 期末專題報告

評量方式

〇 課堂討論參與15% 以學生在課堂中參與討論情形做為考核指標。 〇 出席15% 以學生出席率作為考核指標。 〇 報告(期中報告：資料視覺化報告) 30% 以學生為單位向校務部門主管呈現資料視覺化的結果，並說明重要發現。 〇 專題(期末專題報告) 40% 以小組為單位向校務部門主管呈交校務研究數據分析報告，並以口頭方式報告重要發現。

教材

【遵守智慧財產權觀念，請使用正版教科書，不得使用非法影印教科書】 使用外文原文書：否 必讀教材 Wickham, H., & Grolemund, G. (2017). R for Data Science. O'Reilly Media. 線上〇費版：https://r4ds.had.co.nz/ Ismay, C., & Kim, A. Y. (2019). Statistical Inference via Data Science: A ModernDive into R and the Tidyverse. CRC Press. 線上〇費版：https://moderndive.com/ Peng, R. D. (2016). Exploratory Data Analysis with R. Leanpub. 校務研究專書 Howard, R. D., McLaughlin, G. W., & Knight, W. E. (2012). The Handbook of Institutional Research. John Wiley & Sons.