FINC6007 Chap.2 Measuring ESG Data and Ratings
Measuring ESG Data and Ratings
ESG measurement begins before the score. Analysts define the issue, choose sources, test data quality, select proxies, normalise unlike units and apply weights. Rating providers can disagree because they cover different issues, measure the same issue differently or assign different importance to each pillar.
This chapter reconstructs that chain and uses the IFRS S1 governance, strategy, risk-management, and metrics-and-targets structure as a scope check. It then works weighted pillar scores and z-score normalisation while showing why relative position does not establish absolute sustainability or impact.
What this chapter covers
- 01
Sovereign and corporate ESG sources
- 02
IFRS S1 core disclosure architecture
- 03
Raw data, metrics, themes and pillars
- 04
Scope, measurement and weighting divergence
- 05
Weighted company and portfolio scores
- 06
Z-score normalisation and interpretation
Worked example · free
Calculate and interpret a normalised ESG score
- 1Substitute the pillar values: 0.40(1.20) + 0.40(-0.40) + 0.20(0.80).
- 2Compute the weighted score: 0.48 - 0.16 + 0.16 = 0.48.
- 2Normalise: z = (0.48 - 0.30) / 0.50 = 0.36.
- 1Interpret the result as 0.36 standard deviations above this peer mean, not proof of impact.
Key terms
- ESG Metric
- A measured variable used to represent an environmental, social or governance condition.
- Materiality
- Relevance of an issue to the company, decision and stated user.
- Normalisation
- Transformation that places unlike measures on a comparable scale.
- Rating Divergence
- Difference between provider results caused by scope, proxy, method or weighting choices.
- Z-score
- A normalised value showing how many standard deviations an observation lies from its comparison-group mean.
- Data Provenance
- The source, method and processing history that allow an input to be assessed and reconstructed.
Measuring ESG Data and Ratings FAQ
Why do ESG ratings for the same firm disagree?
Providers may cover different issues, use different proxies, estimate missing data differently, normalise against different peers and apply different weights. The disagreement often reveals model design rather than a simple error.
What does IFRS S1 contribute to ESG analysis?
Its governance, strategy, risk-management, and metrics-and-targets structure helps analysts test whether material sustainability risks and opportunities are overseen, managed and measured, but it does not prescribe one universal rating.
When should ESG scores be normalised?
Normalisation helps when raw indicators use incompatible units or scales. It should follow alignment of scope and meaning, because mathematical comparability cannot repair conceptually different measures.
Does a high ESG score prove a company is sustainable?
No. An aggregate can conceal pillar trade-offs, omitted issues, estimation and weak outcomes. Report material drivers, boundaries and uncertainty alongside the total before making a decision.
How is a portfolio ESG score calculated?
Multiply each issuer score by its portfolio weight and sum the contributions. A simple average is correct only when all holdings have equal weight in the portfolio.
Exam move
Reconstruct one provider from raw input to final pillar score. Start with the decision and industry, choose material issues, list data sources, define each metric, identify reported versus estimated values, normalise the inputs and apply weights only at the end. Practise weighted company and portfolio scores by hand, keeping negative inputs in brackets and portfolio weights visible.
Next calculate a z-score and explain in words what the mean, standard deviation and comparison universe do. Always separate the weighted aggregate from the normalised result: one reflects the model's priorities and the other locates the result inside a chosen peer distribution. Create a divergence table for two imaginary providers. Change only scope in one row, only measurement proxy in another and only weights in a third.
Predict which issuers move and why. Use IFRS S1's governance, strategy, risk-management, and metrics-and-targets structure as a completeness check rather than a scoring formula. When revising sovereign data, note that income and data capacity can influence ratings, so development resources may be confused with sustainability quality.
Finish every exercise with a sentence that separates relative ranking from absolute performance and real-world impact. A high score should trigger two follow-up questions: which material driver produced it, and what raw evidence would contradict the ranking?
The strongest exam drill is to receive two conflicting scores, diagnose the source of disagreement and choose the measure that fits a specified credit, stewardship or impact decision. Add a time-stamp to the exercise and ask whether one provider updated after a controversy while another still uses older data.
Finally, change the peer group without changing company data and observe how the z-score moves; this makes the model dependence tangible. Run a missing-data stress test as well: replace one reported indicator with an estimate, recompute the aggregate and explain whether the ranking changed because of issuer performance or provider methodology. Record the confidence consequence separately from the numerical change.
Working through Measuring ESG Data and Ratings in FINC6007? Sia is AskSia’s AI Finance tutor — ask any FINC6007 Measuring ESG Data and Ratings question and get a clear, step-by-step explanation grounded in how FINC6007 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.
