Machine Learning and Data Science for ESG Compliance Measurement in Financial Software Engineering Projects: A Single-Case Socio-Technical Systems Analysis

Summary

What question does this paper answer?

Can an open-source data science and machine learning workflow measure Environmental, Social, and Governance (ESG) compliance in financial software engineering projects — and what does a firm’s own evaluation practice allow such models to learn?

What did the study find?

In 207 anonymized project records from one financial firm, the stakeholder-rated social and governance factors were almost perfectly correlated with the overall project score (r = +0.995 and +0.966), while the environmental factor was unrelated to it. Machine learning classifiers performed weakly: kNN was at chance (AUC = 0.497) and SVM reached 61.8% accuracy, only modestly above baseline.

Why does it matter?

The study shows that ESG measurement is shaped as much by the human rating process and organizational templates as by the algorithm — a socio-technical systems view. The near-unity correlations describe how the firm scores projects rather than distinct ESG constructs, and the weak classifiers show the limits of learning from such records. It offers a proof of concept and a research agenda for AI-enabled, project-level ESG measurement.

Study at a glance

Abstract

The integration of artificial intelligence (AI) and data science into organizational evaluation practices creates socio-technical systems in which human rating behavior, organizational templates, regulatory pressure, and analytical algorithms jointly determine what can be measured and learned. This single-case study examines the measurement of Environmental, Social, and Governance (ESG) compliance in financial software engineering projects, treating one firm’s project evaluation practice as a socio-technical system and applying an open-source data science and machine learning workflow to 207 anonymized archival project records. Correlation analysis revealed near-unity associations between the stakeholder-rated social and governance factors and the overall project score (r = +0.995 and +0.966, p < 0.001), while the environmental factor was unrelated to the score; post-hoc diagnostics (a seven-component principal-component structure, Harman’s screen, selective near-zero same-source correlations, and marker-variable partial correlations) bind, but cannot eliminate, method-based explanations, so the coefficients are interpreted as a descriptive property of the firm’s evaluation system rather than as estimates of relationships between validated, distinct constructs. Exploratory machine learning classifiers performed weakly—kNN at chance (AUC = 0.497) and SVM only modestly above the no-information baseline (accuracy 61.8%)—a result consistent with the constraints that the social subsystem imposes on the learnability of the records it generates, although technical factors, including the dichotomization of the target variable, the modest sample size, and model configuration, cannot be ruled out as contributing explanations; descriptive statistics are reported for all variables, and a diagnostic analysis reconciles the apparent divergence between the near-unity correlations and the weak classification performance by showing that the two rest on different feature sets, the near-redundant social and governance ratings having been withheld from the classifiers. The findings offer a proof of concept and a structured agenda for AI-enabled, project-level ESG measurement in socio-technical systems.

Abstract as published in Systems under a Creative Commons licence.

Key terms

ESG compliance

The degree to which an organization or project meets Environmental, Social, and Governance criteria set by regulators, investors, or internal policy.

Socio-technical system

A system in which people, organizational rules, and technology jointly determine outcomes, so that none can be understood in isolation.

Common method bias

Inflated correlations that arise because several variables were measured by the same source or instrument rather than because the constructs are truly related.

Limitations

• Single-case design: one firm’s evaluation practice, so results are descriptive of that system.
• Classifier performance may also reflect dichotomization of the target variable, modest sample size, and model configuration.

How to cite

Strang, K. D., & Vajjhala, N. R. (2026). Machine Learning and Data Science for ESG Compliance Measurement in Financial Software Engineering Projects: A Single-Case Socio-Technical Systems Analysis. Systems, 14(8), 990. https://doi.org/10.3390/systems14080990

@article{strang2026machine,
  title = {Machine Learning and Data Science for ESG Compliance Measurement in Financial Software Engineering Projects: A Single-Case Socio-Technical Systems Analysis},
  author = {Strang, Kenneth David and Vajjhala, Narasimha Rao},
  journal = {Systems},
  volume = {14},
  number = {8},
  pages = {990},
  year = {2026},
  publisher = {MDPI},
  doi = {10.3390/systems14080990},
  url = {https://doi.org/10.3390/systems14080990}
}