Using d3.parcoords to explore the influence of various factors on the feeling of happiness between 2010 and 2014
Restrict the happiness brush filter to a quarter of its height
Get a feeling of the correlation by staring at each dimension (vertical axis) while moving the brush filter up and down the happiness axis
The results are somewhat unclear and the number of countries lacking, but unexpectingly the importance of politics seems to have the clearest positive influence on happiness, as well as the importance of leisure time.
This d3.parcoords display allowed me to go through the painful process of dealing with massive data, cleaning data, transforming data to keep only relevant fields (i.e. plotted dimensions), mapping ordinal data to numerical data, and trying to play the Multidimensional Detective.
The derived values for each country are simple averages (sum+normalization), some more complex stats functions could have been used for more relevant result. The scaling could be improved, and the extremal values removed.
My browser couldn't process it, but we could probably sharpen the insight with more countries loaded.