Search for: All records

Predictive modeling often ignores interaction effects among predictors in high-dimensional data because of analytical and computational challenges. Research in interaction selection has been galvanized along with methodological and computational advances. In this study, we aim to investigate the performance of two types of predictive algorithms that can perform interaction selection. Specifically, we compare the predictive performance and interaction selection accuracy of both penalty-based and tree-based predictive algorithms. Penalty-based algorithms included in our comparative study are the regularization path algorithm under the marginality principle (RAMP), the least absolute shrinkage selector operator (LASSO), the smoothed clipped absolute deviance (SCAD), and the minimax concave penalty (MCP). The tree-based algorithms considered are random forest (RF) and iterative random forest (iRF). We evaluate the effectiveness of these algorithms under various regression and classification models with varying structures and dimensions. We assess predictive performance using the mean squared error for regression and accuracy, sensitivity, specificity, balanced accuracy, and F1 score for classification. We use interaction coverage to judge the algorithm’s efficacy for interaction selection. Our findings reveal that the effectiveness of the selected algorithms varies depending on the number of predictors (data dimension) and the structure of the data-generating model, i.e., linear or nonlinear, hierarchical or non-hierarchical. There were at least one or more scenarios that favored each of the algorithms included in this study. However, from the general pattern, we are able to recommend one or more specific algorithm(s) for some specific scenarios. Our analysis helps clarify each algorithm’s strengths and limitations, offering guidance to researchers and data analysts in choosing an appropriate algorithm for their predictive modeling task based on their data structure.

 

The outbreak of the COVID-19 pandemic early in 2020 forced universities to shut down their campuses and transition to emergency remote instruction (ERI). Students had to quickly adapt to this new mode of instruction while dealing with all other distractions caused by the pandemic. This study integrates extensive data from students’ institutional records at a large Historically Black College and University (HBCU) institution with data from a students’ survey about the impact of COVID-19 on learning during the Spring 2020 semester to examine the impact of the transition to ERI on students’ performance and identify the main factors explaining variations in students’ performance. The main findings of our analysis are: (a) students’ university experience was positively correlated with performance (continuing students who spent at least one academic year at the university prior to the outbreak had better performance than freshman and new transfer students), (b) students’ perceived change in performance after the transition was positively associated with actual performance (students who perceived a decline in their performance after transition to ERI had significantly worse performance than other students), and (c) students’ prior online learning experiences and students’ emotional experiences with the COVID-19 disease were not significantly associated with performance. These results suggest that the approaches adopted by higher education institutions to support students during times of crisis should pay special attention to certain groups of students.