Applying the Deep Learning Method for Simulating Outcomes of Educational Interventions

Abstract

Predicting outcomes of educational interventions before investing in large-scale implementation efforts in school settings is essential for educational policy-making. However, due to time and resource limitations, conducting longitudinal, large-scale experiments testing outcomes of interventions in authentic settings is difficult. Here, we introduce the deep learning method as a way to address this issue and illustrate the use of the deep learning method for the prediction of intervention outcomes through a MATLAB implementation. The presented deep learning method extracts predictable patterns from an empirical dataset to simulate large-scale intervention outcomes. Findings from our simulations suggest that the deep learning applied simulation model can predict intervention outcomes significantly more accurately compared to the traditional regression analysis methods.

Improved model exploration for the relationship between moral foundations and moral judgment development using Bayesian Model Averaging

“…Third, we examined whether the BMA-identified model was more robust against overfitting compared with the full models with K-fold cross-validation. K-fold cross-validation examines whether a regression model can predict phenomena out of the boundary of the data used for regression (Han, Lee, & Soylu, 2020). When this method is employed, 1/K of the whole dataset is randomly sampled and used for validation.…”

Applying Deep Learning to Predict Civic Purpose Development: Within the Smart City Context

“…Categorical variables, i.e., Wave 1 political purpose status, gender, ethnicity, born in the US, mother and father born in the US, college boundedness, were converted into dummy variables. Deep learning is one of the widely used methodologies in machine learning for pattern recognition and prediction in the field of computer science [6]. Because deep learning utilizes a neural network consisting of multiple layers for training, it is assumed to have a deep structure, so it is called "deep" learning.…”

“…Although the previous studies examined the factors that facilitate the development of political purpose during emerging adulthood, (e.g., [3][4][5]) several points related to model prediction shall be reconsidered and improved. Recent methodological advances in data science, such as the wide employment of the deep learning method, have demonstrated that novel methods can more accurately predict outcomes compared with traditional regression-based methods in educational and psychological research [6]. This advantage becomes more significant when the prediction is involved in diverse and complex predictors [7].…”

“…Related to the improvement of smart cities, which is the main focus of the current edited volume, examining the developmental trajectories of political purpose with an advanced computational method, the deep learning method, will provide scholars, educators, and policy makers with useful insights [6,13]. We are primarily interested in how to improve citizens' quality of life by implementing smart city solutions based on computational methods, including but not limited to, smart transportation, smart surveillance, smart health solutions, smart infrastructure, and smart energy.…”

“…Such understanding would provide us with more concrete evidencebased ideas about how to design our education as well as policies addressing civic engagement among citizens. In this process, the application of advanced computational methods, the deep learning method in particular, is deemed to increase the accuracy of modeling and prediction (e.g., [6]). So, it would be worthwhile to carefully consider how to apply the methods in the context of research on political purpose development.…”