دانلود رایگان مقاله مدل برای پیش بینی اولیه از دانش آموزان در معرض خطر در یک دوره

Abstract

Using predictive modeling methods, it is possible to identify at-risk students early and inform both the instructors and the students. While some universities have started to use standards-based grading, which has educational advantages over common score-based grading, at–risk prediction models have not been adapted to reap the benefits of standards-based grading in courses that utilize this grading. In this paper, we compare predictive methods to identify at-risk students in a course that used standards-based grading. Only in-semester performance data that were available to the course instructors were used in the prediction methods. When identifying at-risk students, it is important to minimize false negative (i.e., type II) error while not increasing false positive (i.e., type I) error significantly. To increase the generalizability of the models and accuracy of the predictions, we used a feature selection method to reduce the number of variables used in each model. The Naive Bayes Classifier model and an Ensemble model using a sequence of models (i.e., Support Vector Machine, K-Nearest Neighbors, and Naive Bayes Classifier) had the best results among the seven tested modeling methods.

6. Conclusion

This study took a first and successful step in examining different prediction methods for identifying at-risk students early in the semester using performance data during the semester in a course with standards-based grading. For courses with at least 120 students and less than 10% failing rate, it may be possible to use NBC to identify at-risk students during the semester with high accuracy. The more accurate prediction results compared to generic prediction models showcase the possibility of building specific prediction models for each course. In addition to the course-specific models, another aspect of the successful predictions was using in-semester performance data and possibly standards-based grading. Adding this to the already known educational benefits of standards-based grading (although more research is needed) may encourage more instructors to use this kind of grading instead of the traditional score-based grading. All the data that was used in the models, which was insemester performance data, is available to the course instructors during the semester. Thus, by providing specific guidelines on how to create an accurate prediction model (i.e., which prediction model and what types of data to use, how to train, verify, and test the model), course instructors can create and use course-specific models to identify at-risk students and help these students improve their performance.