On March 15, Ramya Balakrishnan, will be speaking about her cutting-edge research.

The abstract of her talk. “Understanding Leaders’ Most Pressing Challenges: Using Natural Language Processing to Develop a Leadership Challenge Ladder” is below.

From effective time management to working across boundaries, leaders at all levels face numerous challenges in the workplace. However, it is unclear which challenges are most salient and how the most important challenges employees face may vary based on their organizational position, sector, and social identity. In this study, we used natural language processing to identify leaders’ key leadership challenges and examine them across leader levels. The Leadership Challenge Ladder (LCL) is an evidence-based and theoretically grounded framework to describe key leadership challenges. Drawing from 11 years (2010-2020) of data collected from over 50,000 multi-level leaders working across more than 6,000 organizations. Participants provided open-ended responses to the question, “What are the three most critical leadership challenges you are currently facing?” Using natural language processing (i.e., Latent Dirichlet Allocation Topic modeling), 5 different topic models were identified based on 5 leader levels (from first-time managers to leading organizations). 6 to 11 topics were identified per leader level, with topics named by four subject matter experts to align with research on leadership development. Findings from this study serve as a foundation for academic and practical applications, including providing organizations with scalable approaches to understanding leadership challenges within their organizations. Based on the findings, this presentation will include three innovative ideas to enhance the interpretability of results, driving the potential for new insights from one’s data.

The Charlotte Women in Data Science (WiDS) Conference is an event hosted and sponsored by UNC Charlotte’s School of Data Science and Lowe’s. Learn more about the Women in Data Science Conference.