ProgressVote: Description
From Open Knowledge Foundation
(Click here for a shorter version)
The motivation for ProgressVote
How this all started
After the topic received close to no attention for decades, there are now government agencies working on the measurement of social progress across Europe. These initiatives were kicked off by a report of several Nobel laureates in 2009 who mainstreamed the critique of using only economic production (GDP) as progress-measure. Their report reinforced that GDP as a single metric cannot express the complexities of developments. But this was only the first of two key contributions! Importantly, they also argued against the standard methodology used in non-GDP progress indicators: composite indices. The assumptions used to construct these indices “are either arbitrary or reflect value judgements on which there is not broad consensus”.
2009-recommendations to governments
Instead of composite indices, Stiglitz et al held that governments should use “dashboards of indicators”: presenting tables with multiple key indicators.
Problem from Open Governance viewpoint
The problem with the report's conclusion, however, is that such dashboards are very complex: we already have some of them for a long time and they feature just so much data that the standard political debate just ignores them altogether and sticks with GDP: that one at least gives a simple, single metric.
Joe Public just cannot absorb as much data as Nobel Laureates can. But Joe matters!, at least for open politics; he is the souvereign, the principal. A social progress indicator must be understandable by the public, it must involve the public. A huge data-dashboard is exclusive.
Why is this politically of the utmost (!) importance
If mainstream politics just does what can be measured and where progress boils down to numbers that can be presented, then anyone hoping for more political consideration of issues that are not immediately captured by economic output has an immediate interest in more holistic measures.
The debate for such measures exists since the 1930s, now is the first time it truly hits politics. And it does so in several European countries and institutions at the same time. What a chance for a coordinated shift!
How policymakers reacted
While some initiatives now do dashboards (e.g. the EU decided to compose one large body of data on sustainability indices) most initiatives have returned/continue to use indices - the critiques raised in the 2009 report notwithstanding.
Much progress has been made to improve these indices, but the main problems remain: the assumptions are arbitrary.
This arbitrariness, unfortunately, takes credibility from the the whole "development beyond GDP"-endeavor. While social progress measurement could be a breakthrough in aligning policy with modern sustainable economics, current government projects risk loosing that chance.
What would ProgressVote do?
Progress Vote suggests a way how dashboards could be used in a way that is open to the non-expert public, incentivising public participation by engaging the user in easy data-mining. The website circumvents the weighting problem inherent to indices by allowing individualised weights for each user.
Step 1: Users are presented with visualisations of past development of key aspects of social progress. Of these variables, both the past development of the stock as well as the current flow would be shown.
Step 2: After seeing the data visualisations the user would be presented with a few short statements by experts. These experts would each present their interpretation of how all the different trends sum up towards the "big picture": how did social progress evolve over the last quarter? Below the statements the expert would also quantify his statement in an index number.
The statements and the quantification are essentially the same as composite indices, the weighting/proxies/shadow prices are just implicit, but each expert is extracting a combined central trend from the various developments.
Step 3: The big difference is that after seeing the visualised data and the experts suggested evaluations the user is asked with which expert they most agree.
They choose this without seeing the different experts' identity, to avoid framing bias. Additionally, the ordering of the experts statements (as well as the visualisations) is randomised, which again eliminates framework effects (in large samples).
Step 4: The key step: the public's and the experts voting combined form the social progress index. To do this, all expert index figures are weighted according to the number of users that agreed with them.
So imagine there are four experts and they rate the development of a number of key time series on a certain index scale, next to text-based interpretations explaining their rating. Then users select which experts they agree with. We take each of the experts rating, multiply them by how many users agreed with each rating, divide by the number of total votes and get the final social progress index.
This index now has those weights that the users collectively want, not an arbitrary ranking.
Step 5: At the end of each quarter the index would be communicated coinciding with the publication of GNI, but in format of an “Open Letter” from the user-society to the government.
