Participant_Dropout_Classification

Summary

C5.0 classifcation decision trees are useful because the algorithm can handle numerical and multi-categorical data. For this use case, a C5.0 decision tree was used to classifiy participant dropout from an online research study. Features of participants who drop out can be used to guide future remote protocol development by building in support for demographic-specific retention. Teleresearch protocols require extra planning and thoughtful implementation to keep participants engaged in an environment that is less controlled than in-person research. Using data-driven methods to evaluate the performance and efficiency of teleresearch protocols is critical, including factors that classify dropout.

Programming Language: R

Corresponding Publication: https://doi.org/10.3389/fnhum.2023.1251174