Causal Inference Methods in Social Return on Investment Analysis: A Scoping Revie
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Project Description The Problem Social Return on Investment (SROI) is a vital framework for valuing the social, environmental, and economic impact of interventions. However, its core methodological weakness lies in the construction of the counterfactual. Key discount factors (specifically deadweight and attribution) are routinely estimated through subjective stakeholder consultation rather than robust empirical analysis. This reliance on assumption over causal identification limits the evaluative integrity of reported SROI ratios and undermines evidence-based commissioning. Project Purpose This overarching research project seeks to resolve this tension by formally integrating modern causal inference methods (such as Difference-in-Differences, Propensity Score Matching, and Randomised Controlled Trials) into the SROI framework. The goal is to transition SROI from an assumption-heavy accounting exercise into a rigorous, scientifically defensible evaluation methodology. Phase 1: Methodological Mapping (Current) The foundational phase of this project is a comprehensive scoping review mapping the extent to which causal inference methods have been applied within, or proposed as improvements to, SROI evaluations. The primary output of this phase will be the publication of a novel, tiered "Typology of Counterfactual Quality," designed to help evaluators and commissioners appropriately match causal rigor to the stakes of the evaluation. (Note: The protocol for this review is pre-registered under this OSF project). Phase 2: Empirical Prototyping (Upcoming) Following the scoping review, the next phase of this research programme will involve the empirical application of the proposed typology. Upcoming work includes a comparative prototype study applying quasi-experimental designs (e.g., Propensity Score Matching) to an existing social intervention dataset. This phase will empirically demonstrate the variance between a "Standard SROI" ratio (relying on generic assumptions) and a "Causal SROI" ratio (usinga statistically matched control group) to quantify the risk of overclaiming in conventional practice. Expected Outcomes Ultimately, this research project aims to equip the impact evaluation sector with the methodological tools and practical frameworks necessary to improve the evidential basis of social value claims. By bridging the gap between participatory stakeholder valuation and rigorous econometric causal inference, this work will provide a roadmap for more credible, transparent, and accurate policy and commissioning decisions.
Fields of Study & Subjects
Evaluation and Performance Assessment
Advanced Causal Inference Techniques
Community Development and Social Impact
Publication details
- Publication date
- 1 January 2026
- Language
- EN
- Document type
- Preprint