Would a science-based approach to ESG data resolve credibility problems?

Dr Katarzyna Wilk argues that sustainability disclosure and analysis needs to look to scientific practice for guidance

The ongoing ESG debate focuses on two main issues: quality of data and the most relevant approach to ESG analysis. The first pertains to problems such as the lack of validity, reliability and comparability of available ESG data. The second asks for a holistic and more informative approach, allowing all stakeholders – including finance, business and policymakers – to exchange views. I believe both challenges can be effectively resolved with a science-based approach and methods.

Holistic Approach to ESG Data Analysis 

Financial analysis of ESG data and a science-based approach to ESG data analysis dominate the current debate. Practitioners and academics, however, seem to contradict each other – tend to use different narratives and, given a limited access to raw company-level data, they use different ESG data sets. Comprehensive and insightful analysis of ESG data should simultaneously meet the objectives of investors, policymakers and academics – keeping in mind that an ultimate goal is an improvement in overall sustainability.

To ensure comparability of ESG data across different data providers, company sustainability assessment tools should include a core standardised questionnaire be used by all data providers

ESG data presents enormous potential for more informative analysis, capturing both the differences and the similarities between – and within – companies, sectors and regions. ESG analysis, thus, should go beyond simply providing scores, and serve all stakeholders.

Advanced statistical analysis offers a comparative approach across time and entities and sectors, putting ESG risks in broader, more meaningful contexts. Financial analysis, therefore, can be significantly improved by using concepts and methods from science, which could become an integral part of investment decision-making. Examples of such analysis include: regressions and logistic regression, and hierarchical linear and non-linear modelling.

ESG Data Quality

ESG data quality problems are rooted in the lack of standardised ESG/sustainability assessment tools. Measurement asymmetries, the lack of comparability and validity – and consequently, the lack of credibility of ESG analysis – lead not only to confusing investment recommendations, but also inhibit informative debates between academia, practitioners and policymakers.

In order to ensure high-quality ESG data analysis, company-level data should be collected in a way that meets the methodological standards of science – in particular, social sciences and survey methods – including common standards for ensuring data validity, reliability and comparability across various contexts (e.g. sectoral, regional or country) through standardised questionnaire design (i.e. ESG assessment tools), relevant measurement tools, and standardised processes for coding data into variables.

Moreover, in order to ensure comparability of ESG data across different data providers, company sustainability assessment tools should include a core standardised questionnaire be used by all data providers, in analogy with the reputable world or regional surveys in social sciences. This would not eliminate a possibility for data providers to include additional questionnaire items that serve to distinguish them from competitors. Fulfilling these requirements will lead to credible analysis, formulation of credible investment recommendations and, importantly, to overall sustainability improvements.

In conclusion, financial and science-based analysis of ESG data, when taken into account simultaneously, provide a holistic picture of ESG risks and opportunities, measured at multiple levels. In order to achieve this holistic view, ESG data providers need to work with academia, research institutes and policymakers to exchange expertise, foster collaborative projects and increase transparency. This would result in improved and measurable sustainability performances of companies over time, better informed policy, and more accurate tools and standards for measuring both impact and sustainability progress. All of the above is essential to generating well-informed changes and a significant progress towards sustainable development, measurable at the corporate, regional and global level.


Dr Katarzyna Wilk is CEO & founder of impact investment advisory Swiss Impact Lead based in Zurich. She has served as a policy advisor at the European Commission, and was a collaborator of the Polish Academy of Science, specialising in advanced comparative analysis, cross-national survey design and inequalities.