Abstract

The ability to express oneself verbally is essential to success in employment interviews and in a wide array of occupations. Yet, the exploration and measurement of audio features in video interviews remains largely unexplored. The current work reviews previous research of auditory features and relates them to pre-employment interviews, creating an operationalization of verbal expression, with both a definition and behavioral anchors. Across two pilot studies, we demonstrate the reliability of a pairwise comparison approach to the measurement of the construct and provide preliminary validity evidence. The construct and associated measurement techniques represent a measurement-driven approach that can create the foundation for the implementation of large-scale deep learning models that will automate the measurement of acoustic features as they relate to employment screening.