Peace is notoriously difficult to define. The simplest way of approaching it is in terms of the harmony achieved by the absence of violence or the fear of violence, which has been described as Negative Peace. This complements Positive Peace which is defined as the attitudes, institutions and structures that create and sustain peaceful societies.
The GPI was founded by Steve Killelea, an Australian technology entrepreneur and philanthropist. It is produced by the Institute for Economics & Peace, a global think tank dedicated to developing metrics to analyse peace and to quantify its economic benefits.
The GPI measures a country’s level of Negative Peace using three domains of peacefulness.
Explore the 23 indicators of peace
The GPI comprises 23 indicators of the absence of violence or fear of violence, reviewed annually. All scores for each indicator are normalised on a scale of 1-5, whereby qualitative indicators are banded into five groupings and quantitative ones are scored from 1 to 5, to the third decimal point.
Two subcomponent weighted indices are calculated:
The GPI is a composite index of 23 indicators weighted and combined into one overall score. The weighting scheme within any composite index represents the relative importance of each indicator to the overall aim of the measure, in the GPI’s case, global peace.
To fully understand the representative nature or accuracy of any measure it is necessary to understand how sensitive the results of the index are to the specific weighting scheme used. If the analysis holds true for a large subset of all possible weighting schemes then the results can be called robust.
While it is expected that ranks will be sensitive to changes in the weights of any composite index, what is more important in a practical sense is the robustness of country comparisons. One of the core aims of the GPI is to allow for Country A to be compared to Country B. This raises the question that for any two countries, how often is the first ranked more peaceful than the second across the spectrum of weights. The more times that the first
country is ranked more peaceful than the second, the more confidence can be invested in the statement “Country A is more peaceful than Country B”.
To avoid the computational issue of evaluating every possible combination of 23 indicators, the robustness of pairwise country comparisons has been estimated using the three GPI domains militarisation, societal safety and security and ongoing conflict. Implementing an accepted methodology for robustness, the GPI is calculated for every weighting combination of three weights from 0 to 1 at 0.01 intervals. For computational expedience only weighting schemes that sum to one are selected, resulting in over 5100 recalculated GPI’s. Applying this, it is found that around 70 per cent of all pairwise country comparisons in the GPI are independent of the weighting scheme, i.e. 100 per cent robust. This is a similar level of absolute robustness as the Human Development Index.
The EIU’s Country Analysis team scores five qualitative indicators and country analysts are asked to suggest an alternative data source or provide an estimate to fill any gap for missing quantitative indicators when official data is missing. Supported by over 100 full-time analysts and 650 in-country contributors, they develop deep knowledge of each nation’s political, economic, and social conditions. Scoring is carefully reviewed through a multi-level process involving regional directors, the Custom Research team, and the external advisory panel to ensure consistency, reliability, and global comparability.
For the latest data and any updates to the methodology, refer to the most recent Global Peace Index report which can be downloaded from the Peace Resources page.
