print

Correlational findings

Study Deaton (2008): study ZZ 2006

Public:
15+ aged, general public, 132 countries, 2006
Sample:
Respondents:
N = 132000
Non Response:
n.a.
Assessment:
Interview: face-to-face

Correlate

Authors's label
Income
Our Classification
Operationalization
Logarithm of per capita Gross Domestic Product adjusted for purchasing power parities

Observed Relation with Happiness

Happiness Measure Statistics Elaboration / Remarks C-BW-c-sq-l-11-c r = + Life satisfaction is higher in countries with higher GDP per capita C-BW-c-sq-l-11-c r = + A doubling of GDP is associated with a constant increase in mean life satisfaction (middle broken line)

The upper line in figure 2 refers to young people (15 to 25) whereas the lower line refers to old people (60+):
- Old individuals are happier than young individuals in very rich countries
- Old individuals are less happy than young individuals in middle and low income countries
- Difference is stronger in middle income countries compared to low income countries
C-BW-c-sq-l-11-c b = +.84 p < .00 All countries, N=123

B does not change when controled for:
- economic growth 2000-2003
C-BW-c-sq-l-11-c b = +.87 p < .00 Without the countries of the former Sovjet-Union, N=111

B controled for:
- economic growth 1990-2000
- economic growth 2000-2003
C-BW-c-sq-l-11-c b = +.92 p < .00 All countries, N=120

B controled for:
- economic growth 2000-2003
- life expectance 2005
- change in life expectancy 1990-2005
C-BW-c-sq-l-11-c b = +.69 p < .00 All countries, N=123 C-BW-c-sq-l-11-c b = +1.6 p < .00 Among NOT POOR countries, income > $12.000, N=38 C-BW-c-sq-l-11-c b = +1.3 p < .00 Among NOT POOR countries, income > $12.000, N=37

B controled for:
- economic growth 2000-2003
- life expectancy 2005
- change in life expectancy 1990-2005
C-BW-c-sq-l-11-c b = +.38 ns Among RICH countries, income > $20.000, N=25

Insignificant due to low number of observation

B's similar across age groups

B (+.84) means that if income changes from $10.000 to $20.000 the happiness score increases by ((ln)20.000-(ln)10.000)*0,84 = 0,58 points due to the log scale