Towards Causal Analysis of Empirical Software Engineering Data: The Impact of Programming Languages on Coding Competitions

TL;DR

This paper explores causal analysis techniques for observational software engineering data, demonstrating how causal models can improve understanding of factors like programming languages' impact on coding competition performance.

Contribution

It introduces novel structural causal model techniques to analyze observational data and applies them to assess programming languages' effects in a large coding contest.

Findings

Weak overall effect of programming languages on performance

Significant differences between causal and correlational analyses

Causal analysis provides more robust insights

Abstract

There is abundant observational data in the software engineering domain, whereas running large-scale controlled experiments is often practically impossible. Thus, most empirical studies can only report statistical correlations -- instead of potentially more insightful and robust causal relations. To support analyzing purely observational data for causal relations, and to assess any differences between purely predictive and causal models of the same data, this paper discusses some novel techniques based on structural causal models (such as directed acyclic graphs of causal Bayesian networks). Using these techniques, one can rigorously express, and partially validate, causal hypotheses; and then use the causal information to guide the construction of a statistical model that captures genuine causal relations -- such that correlation does imply causation. We apply these ideas to analyzing public data about programmer performance in Code Jam, a large world-wide coding contest organized by Google every year. Specifically, we look at the impact of different programming languages on a participant's performance in the contest. While the overall effect associated with programming languages is weak compared to other variables -- regardless of whether we consider correlational or causal links -- we found considerable differences between a purely associational and a causal analysis of the very same data. The takeaway message is that even an imperfect causal analysis of observational data can help answer the salient research questions more precisely and more robustly than with just purely predictive techniques -- where genuine causal effects may be confounded.