What is prosperity? It's often thought of simply in financial or economic terms. This is part of it, but there's much more to it than that. The root of the word 'prosperity' comes from pro and spere which comes from 'spes' which is Latin for hope. To be prosperous is to have hope, and to enjoy success in obtaining what we hope for.

We hope for many things in life, and therefore the Index seeks to measure a very wide range of factors contributing to prosperity. Nevertheless, it is impossible to measure everything. There are practical reasons for this: good data isn't available for everything, and it isn't available in every country. Even when it is, it's not available for each year we would like to measure. More fundamentally though, our job as analysts and statisticians is not to measure everything but to identify and process data to provide a robust estimate of the broad concept of prosperity.

An analogy here can be made with health: whilst tens of thousands of relevant datapoints could be obtained by extensive medical testing, physicians and actuaries both know very well that only a handful of well-chosen measures are necessary to accurately estimate a person's health. For example, age, blood pressure, and smoking status collectively contain a very large amount of information about a person's general health and expected lifespan. The human body is enormously complex, and yet its health can be estimated to a high degree of precision with only a handful of measurements. Moreover, given the current state of technology and science, it is not clear how one could reliably make use of tens of thousands of medical data points, and they would likely dilute the signals contained in essential variables such as smoking status. The key to reliably estimating a person's health, therefore, is figuring out which are the most important questions to ask. Clearly, the first is What is your date of birth?

We believe a similar phenomenon exists with measuring a country's prosperity. In this year's Index, we have broken down prosperity into just three core domains, which we have further subdivided into ten pillars, which are detailed in the next section. These three domains can be thought of as putting a number on the following three core questions:

1. How developed is the country?
2. How free is it?
3. How strong is its social fabric?

The pillars and the various data points that feed into them represent our attempt to break these three core questions down into their constituent parts. In spite of the numerous data points we use, we are still ultimately asking just three core questions.

Data cannot be used in its raw form with its varying units, sizes, and ranges: it must first be standardized by transforming it into a score. We call this transformation a scoring function, i.e. a mathematical map from the data to a number between 0 and 1. Generally, however, we give the worst-performing country a minimum score of 0.1 rather than a 0.

The scoring function that we use most often is known as "distance-to-frontier" (D2F), which simply maps the lowest data point to 0 (or in our case 0.1) and the highest to 1 and then interpolates every other intermediate value using a straight line to connect the best and the worst. Sometimes high data values are bad, e.g. murder rates, so we use what we call "reverse D2F", which is the same idea except the lowest data value gets a 1 and then highest a 0 (or a 0.1 in most instances).

In some cases, however, we deem it necessary to use a non-linear scoring function. A simple example is our transformation for inflation rates, where inflation that is too low (or worse: deflation) or too high are both harmful. Clearly a D2F or reverse D2F function would be inappropriate, as the optimum amount of inflation is low but not too low. It is clear, therefore, that the scoring function for inflation ought to resemble a "mound". In this particular case, we have elected to use a logistic distribution with a peak close to 2%.

After assigning scores to each indicator, they must be weighted before they can be averaged. We generally use uniform weights, e.g. if there are three indicators that are to be averaged together to produce a score for a higher-order concept, then a uniform weighting would entail assigning 1/3 to each indicator. In many cases, however, certain indicators are more important than others, e.g. murder is a far worse crime than petty theft and we deemed heroin use far more baneful to a nation's prosperity than cannabis. In both instances, we have used non-uniform weights to reflect the imbalances in the indicators' importance.

Once the process of averaging is complete, we have resulting scores for each country for the overall Prosperity Index and for each of the three domains and ten pillars as well. This enables us to rank countries based on how high their scores are relative to one another. The latest scores and ranks are available to browse on our rankings page for the overall Index as well as the domains and pillars. On each country profile page, the graph feature enables the comparison of different countries' prosperity across the ten years covered by our index data. You can also download the data here.

Our index uses data as described above, aggregating, transforming, weighting, and averaging it in line with our understanding of prosperity. We recognise that at each juncture, decisions regarding the way data is processed necessarily reflect our particular view of prosperity and the nature and relative importance of variables measured. Nonetheless, in every instance, we have aimed to be reasonable, emphasizing practicality over dogmatism.

Methodological decisions were taken by an in-house panel of Prosperity Institute staff. We consulted with a range of external experts who were provided with preliminary results and offered a presentation of our overall methodological approach. These experts come from a wide range of fields including economics, philosophy, law, political science, international development, and public policy. We collated feedback from internal and external reviews and implemented changes after making overall judgements. The bespoke software we have crafted to produce the Index was subject to a thorough audit by external software developers at a late stage in its development.