The GGEI methodology draws from guidelines published through the OECD Handbook on Constructing Composite Indicators, which can be accessed here.  We have also consulted extensively with the publishers of other leading indices in this field to learn from their methodological approaches to similar measurement challenges.  Publishing an index like the GGEI is ultimately a series of decisions, often balancing the depth and breadth of issues covered against the available data. Furthermore, the concept of "green economy" is still a nascent one, gradually becoming more defined as the theoretical and practical parts of it are tested and developed. The sections below offer more background on how we have addressed methodological steps and challenges in calculating the 2016 GGEI.

Theoretical Framework

In the year before the publication of the first edition of the GGEI in late 2010, the Dual Citizen LLC team assembled a group of experts to define the theoretical framework of what became known as the Global Green Economy Index.  These experts - with backgrounds in climate change negotiations, renewable energy, policy advocacy on green economy and branding and communications - defined four main dimensions around which percpetions would be assessed: political leadership, policy, investment and tourism.  This first edition of the GGEI only calculated perceptions in a generalized manner, asking how respondents perceived national green reputations in these four areas.

The subsequent editions of the GGEI in 2011 and 2012 expanded upon this foundation in two important ways.  The first was to measure performance of the same values being polled in the survey, using datasets from third party sources and where appropriate, qualitative measures generated internally.  The second development was to expand the sub-categories of each of the four dimensions such that instead of just one measure for each dimension, an aggregate result could be generated from a series of related sub-categories (e.g. Political Leadership became defined by heads of state, media coverage, international forums etc.)

In early 2014, Dual Citizen LLC commissioned a strategic review of the GGEI with the purpose of revising its methodology and framework to more accurately reflect the different aspects of a green economy.  This process yielded two important changes.  The first was to expand the sectors covered beyond tourism to include other efficiency sectors like buildings, transport and energy.  The second was to integrate environmental performance to the GGEI such that both the economic and environmental pillars of green economy could be explored through the perception and performance results.  This review also lead to a more explicit linkage between leadership and climate change, such that linkages could be explored around whether national political rhetoric and policy was actually having a positive impact on the country's climate change performance.

Data Selection

The GGEI leverages data that best satisfy two central criteria: quality and coverage. Creating an index like the GGEI, one quickly absorbs the reality that datasets are often less complete than it may appear at first glance, and rarely cover a wide enough range of countries in a uniform manner.  Part of this is due to the manner in which countries are organized (i.e. the EU, OECD, the G20) whereby these organizing bodies lead data gathering efforts, and their associated parameters.  For example, robust datasets may exist for OECD countries but not for the entire G20. Furthermore, not all countries are compliant in reporting data in a timely manner, meaning that even if there is complete data coverage for a grouping of countries, the time series may be inconsistent with some countries having more recent data than others.

Given these realities, our approach to data selection has assumed a "top down" method, as opposed to a "bottom up" one. This means that we first defined the most important dimensions and associated sub-categories to measure given the purpose and goals of the Global Green Economy Index.  Then, based on this framework, we identified third party datasets that provided the best value measure given the required GGEI country coverage or, where appropriate, generated a system for calculating a qualitative scoring.  

We advocate for this "top down" approach to data selection for two main reasons.  The first is that existing data are not necessarily the most important values to measure related to a particularly topic.  Sometimes, data exist due to approaches or processes that may be outdated, or institutional priorities that no longer reflect the more important issues of the day.  Determining first what the overall framework is ensures that an index be defined by the topics that matter most, not simply the ones that are easiest to measure.  The second reason validating this "top down" approach is that it applies focus on areas where data are incomplete, providing incentive to statistical agencies, country ministries and international institutions to make collecting them a priority.

Admittedly, this "top down" approach has its limitations, and it is important to highlight them in the context of the green economy topic.  There remain vital components to understanding a green economy that simply can't be measured in a sound manner today.  One example is green jobs, where a working definition remains elusive and data are inconsistent across different country profiles.  Another example is manufacturing - a vital efficiency sector - where due to the complexity of supply chains and the diverse inputs required to create a reliable dataset, no sound approach exists yet to estimate the extent to which domestic manufacturing is "green."  In these cases, it would be imprudent to force a performance measure through an index like the GGEI, although we hope these limitations will change for future editions.

Imputation of Missing Data

While best efforts are made to identify data sources that provide adequate country coverage, it is inevitable that some data will be missing. In the 2016 GGEI, this issue came up mostly on the Markets & Investment dimension where due to the diversity of the 80 country profiles between more advanced economies and emerging ones, it was sometimes impossible to find complete data sources for the four sub-categories.  

On this dimension, our approach to imputing missing data was to make approximate scores for countries with missing data based on averaging scores from the five closest countries in terms of factors we could deduce.  For example, if a composite indicator estimating country attractiveness for renewable energy investment referenced in part a country's result on the World Bank's Doing Business report, a missing value would be estimated by looking at the scores of the five nations closest to it on the Doing Business report, and then averaging their scores on the composite indicator to compute the missing data point for the country in question. Like any approach to estimation, this method is imperfect and makes certain assumptions about national performance in one aspect of the economy based on results from another related one. But in terms of best practices for index creation and data aggregation, this is a more responsible approach than leaving the value blank or arbitrarily giving a mean score to countries with missing data.

To a lesser extent, we also grappled with missing data issues on the Environment & Natural Capital dimension, but for different reasons. In these cases, missing data resulted from the natural characteristics of each country.  For example, it is impossible to generate a value for Forests if a country doesn't have them or for Fisheries if a country is landlocked.  In these limited cases, countries in question received the highest score from the category.  Again, this approach is imperfect in that it basically credits a country for top performance on an environmental category that simply does not exist, potentially skewing overall results in its favor.  But the alternatives to our approach are less appealing, and expose the GGEI results to greater risk of imbalance. Excluding these categories for the countries in question would weight the other sub-categories more in the aggregate result, and create a situation where the internal weightings differed on a country by country basis. Alternatively, leaving these values blank would in effect punish these countries for natural characteristics of their territory beyond their control.

Normalization, Weighting and Aggregation

We applied a consistent normalization approach using GDP (PPP) to expressed values with inherent imbalances based on the size of the country economy.  Based on the "top down" approach to data selection, we generally applied equal weightings to both the four dimensions and their sub-categories.  One exception to this is on the Leadership & Climate Change dimension, where we lessened the weighting for the head of state and media coverage sub-categories, which in turn more heavily weighted international forums and climate change performance.

Obviously, the GGEI draws from a wide range of underlying datasets, and it is important to assume a consistent method for aggregating them.  Our approach was to calculate the mean and standard deviation for each indicator or dataset, which in turn enables calculating a z-score and associated percentile.  Then, these percentile values can be aggregated in a uniform manner, generating a country score that is expressed on the spectrum of 0-100.

Data Visualization and Results

The GGEI is a robust tool that enables a wide range of meaningful viewpoints into green economic performance and expert perceptions of it. In the public report released in September 2016 sharing the latest results, we will share a variety of graphs, charts and other visualizations to help readers digest these results and their linkages to other topics.  We encourage individuals and institutions to integrate these visuals to their own reporting and will create an image library accessible through the Usage section of this website to facilitate this.