Happiness Index

Team: HappinessMatters

Governments around the world have started to investigate alternative methods for measuring “success” besides just GNP. One of these methods is Gross National Happiness, which measures quality of life or social progress in more holistic terms than GDP (http://en.wikipedia.org/wiki/Gross_national_happiness)

Team HappinessMatters embarked with the idea of applying this global concept at the Local Government Association (LGA) level. In other words, is the area where you live happier than the other areas in Australia?

Our first task was to establish “what is happiness”. Instead of just making assumptions what quantitative elements make up happiness, or generic surveys asking “are you happy”, we used the research from the paper “Factors Predicting the Subjective Well-Being of Nations” (Diener et al, 1995). From this paper we isolated the variables with the highest significance and located data to support a possible model.

The Happiness Index is rooted in statistics, which means it doesn’t just display data, but displays the index as a result of modelling multiple predictors. The usable output is a map depicting the happiness index per LGA.

Notable interests in the data analysis include:

– Finding the income growth mean per LGA (using the formula to calculate GDP)
– Comparing income levels spatially between neighbouring LGAs. In other words, compare the income level of every LGA to average income levels of the surrounding LGAs.
– Analysis of income inequality based on the Gini coefficient of income levels within the LGA

The predictors we used and their associated data sets:
Complaint Statistics (Rights measure): http://humanrights.gov.au/about/publications/annual_reports/2009_2010/complaint-statistics.html

National Regional Profile (Money, Growth, Neighbouring LGA’s Income)

Income Distribution from Basic Community Profiles, from the 2006 Census Data (Income Gini)

Local Government Areas ASGS Non ABS Structures Ed 2011

Consumer Price Index

Tools used: QGIS, R, Microsoft Excel, Google Fusion Tables, PostGIS, and Python

Our final data can be seen in Google Fusion Tables, and our analysis can be visualized in a Google Fusion Table map.

Intermediate data and most of the scripts used to prepare the data are available on Bitbucket. During the data-gathering phase, we considered many other sets of data that we didn’t end up using in our analysis.



  1. Harold W Schranz

    June 3, 2012

    Cute idea ala Bhutan. However happiness is not directly proportional to income. In fact, less is more, and real happiness ( fulfillment) is more likely bimodally related to income; as long as basic needs are taken care of the cohort is likely to be happy in a family & community sense. More material goods ( mortgages) don’t equal happiness – more likely related to stress.

    • James Polley

      June 3, 2012

      Actually, the research that Bhutan uses (http://www2.eur.nl/fsw/research/veenhoven/Pub1990s/91a-full.pdf) suggests that the relationship is not bimodal – it just stops having any significant effect above a certain level. Our model could certainly be improved by taking this into account, but we’d need to have some data about where the threshold is.

      Our model (like Bhutan’s model) also gives a fairly high weighting to income inequality, as the same research also suggests that being poorer than people around you is at least as significant as the absolute value of income.

      We would have liked to take more factors into account (for instance: average commute time, and average dwelling size, as there’s research to show that both of these are correlated with happiness – http://www.planetizen.com/node/43787) but we weren’t able to find this data in time.

  2. Kelvin Nicholson

    June 3, 2012


    Thanks for the comment! Mark and James may contribute more shortly, but if this is a topic you’re interested in, I would encourage you to read the following paper that details the science behind the data analysis we performed:

    “Factors Predicting the Subjective Well-Being of Nations” (Diener et al, 1995).

    You are completely correct about Bhutan, and that is what we were discussing on the train home. While the research didn’t state happiness was directly proportional to income, the variable in the model was strongly significant.


Add a comment

You must be logged in to comment.

GovHack Sponsors

IP Australia
IAG Google Ancestry

Platinum Sponsors 2016

Announcement coming soon

Gold Sponsors 2016


Silver Sponsors 2016

DLA Piper
Department of Finance Bureau of Meteorology
Department of Industry, Innovation and Science AIHW
Who’s behind GovHack? This is a non-profit event proudly run by volunteers who form the GovHack Coordination Team. Our ongoing thanks to everyone who gets involved and makes GovHack awesome! That is, the hackers, data providers, sponsors, mentors and a special thanks to the volunteers who run Local GovHack events.
Sponsoring GovHack 2016 If you are interested in sponsoring this event, please contact GovHack Australia at sponsors@govhack.org or GovHack New Zealand at sponsors@hacknz.org.
The GovHack Global Operations Team Click through to find us on Twitter and Google+ Richard TubbJan Bryson,  Alex Sadleir, Alysha Thomas, Keith Moss,  Gavin Tapp

Privacy Statement/Disclaimer
GovHack is run under the auspices of Linux Australia.