Detecting Micromanagement During Pair Programming

Solomon Ubani, Rodney Nielsen

Abstract: One of the biggest challenges hindering effective collaboration and therefore, pair programming, is micromanagement. In this paper, we investigated the capabilities of a pretrained transformer-based machine learning model to detect such micromanagement, following a text classification approach. We collected and labeled a dataset of 42,820 utterances from three course assignments involving 32 pair programmers in remote environments. This is the first successful automated detection of such micromanagement.