Can Causality Cure Confusion Caused By Correlation (in Software Analytics)?

TL;DR

This paper explores whether integrating causality-aware split criteria into symbolic models enhances their stability and robustness in software engineering tasks, without sacrificing predictive accuracy, and compares automated models with human expert judgments.

Contribution

It introduces causality-aware decision trees for software analytics and demonstrates their improved stability over traditional correlation-based models, validated through extensive empirical evaluation.

Findings

Causality-aware trees show higher stability than correlation-based trees.

Human expert judgments are more stable than automated models.

Causality-aware models maintain comparable predictive performance.

Abstract

Background: Symbolic models, particularly decision trees, are widely used in software engineering for explainable analytics in defect prediction, configuration tuning, and software quality assessment. Most of these models rely on correlational split criteria, such as variance reduction or information gain, which identify statistical associations but cannot imply causation between X and Y. Recent empirical studies in software engineering show that both correlational models and causal discovery algorithms suffer from pronounced instability. This instability arises from two complementary issues: 1-Correlation-based methods conflate association with causation. 2-Causal discovery algorithms rely on heuristic approximations to cope with the NP-hard nature of structure learning, causing their inferred graphs to vary widely under minor input perturbations. Together, these issues undermine trust, reproducibility, and the reliability of explanations in real-world SE tasks. Objective: This study investigates whether incorporating causality-aware split criteria into symbolic models can improve their stability and robustness, and whether such gains come at the cost of predictive or optimization performance. We additionally examine how the stability of human expert judgments compares to that of automated models. Method: Using 120+ multi-objective optimization tasks from the MOOT repository of multi-objective optimization tasks, we evaluate stability through a preregistered bootstrap-ensemble protocol that measures variance with win-score assignments. We compare the stability of human causal assessments with correlation-based decision trees (EZR). We would also compare the causality-aware trees, which leverage conditional-entropy split criteria and confounder filtering. Stability and performance differences are analyzed using statistical methods (variance, Gini Impurity, KS test, Cliff's delta)