January 2025 Online Short Course

Machine Learning/Predictive Modeling

Machine Learning/Predictive Modeling

Dr. Louis Hickman

January 6-9, 2025 | 10:00 AM EST – 3:00 PM EST

Course Description

Organizational research often employs traditional statistical methods, such as linear regression, ANOVA, EFA, CFA, and SEM. However, these methods have limitations. They may not fully capture data complexity, overlook crucial relationships, or fail to optimally predict dependent variables. Additionally, traditional models can be overly complex, fitting well in the sample but failing to generalize to new, independent data due to potential chance capitalization. Machine learning algorithms address these limitations by avoiding underfitting (capturing complexity) and overfitting (cross-validating on new data). Predictive modeling forms, like random forests, LASSO regression, and neural networks, serve these purposes. These models complement traditional approaches in organizational research and may even replace them, especially when the number of variables exceeds the number of cases.This CARMA short course provides a hands-on experience using R and RStudio to analyze predictive models. If you’re unfamiliar with R basics, we recommend taking CARMA’s introductory R course. We’ll use available datasets and pre-developed R code to discuss, run, and interpret various machine learning models. Time permitting, we’ll explore methods to compare the performance of these models.

Meet the Instructor

My research focuses on the intersection of technology and work, with an emphasis on applications of machine learning and artificial intelligence to organizational science and practice (e.g., automatically scored interviews). More broadly, I use computers to measure verbal, paraverbal, and nonverbal behaviors in order to advance our understanding of how interpersonal perceptions form and how cultural, racial, and gender biases function. My current research projects include: understanding how first impressions form in professional settings, mitigating algorithmic bias, understanding how biases influence hiring decisions, and using machine learning and artificial intelligence to help individuals with personal and professional development. In my research, I collaborate with scholars in psychology, management, information sciences, and computer science.