Life-satisfaction is assessed by means of surveys in general population samples. See Rank Report Average Happiness .
In this ranking the focus is not on the level of happiness in the country. but on inequality in happiness among citizens. Inequality in happiness can be measured by the dispersion of responses to survey-questions. The degree of dispersion can be expressed statistically in the standard deviation and surveys items rated on a 10 step numerical scale are particularly usefull for that purpose. This method is explained in more detail in W. Kalmijn & R. Veenhoven
Measuring inequality in happiness in nations; In search for proper statistics, Journal of Happiness Studies 2005, 6:357-396.
An application of this method is reported in Veenhoven. R. Return of inequality in modern society? Test by dispersion of life satisfaction across time and nations, Journal of Happiness Studies 2005, 6:457-487 .
The scores are based on responses to a question about satisfaction with life, the answers to which were rated on a numerical scale ranging from 'dissatisfied' to 'satisfied'. The questions differ slightly in wording and answer format. Most questions are type O-SLW/c/sq/n/10/a (used in World Value Surveys) and O-SLW/c/sq/n/11/a (used in Gallup World Poll).
This classification is explained in section 4/3 of the introductory text. Data were taken from the two tables with measure types regarding the 10-step numeral LifeSatisfaction and the 11-step numeral LifeSatisfaction from this collection of distributional findings on Happiness in Nations.
Rating scales ranged from 1 to 10 or from 0 to 10. Scores on this 1-10 scale were transformed linearly to range 0-10. This transformation in explained in the introductory text, chapter 7.3.
Standarddeviations of 8 nations are based on responses to a somewhat different question:
Suppose the top of the ladder represents the best possible life for you and the bottom of the ladder the worst possible life. Where on this ladder do you feel you personally stand at the present time?The response was rated on a ladder scale ranging from 0 to 10 (measure code C-BW/c/sq/l/11/c). We transformed the scores using the information of nations in which both this item and the above question on life-satisfaction had been used in about the same years. There are 128 such cases. The standard-deviations of the responses on the two items appears to be correlated: r = +.51 after deletion of outliers (Angola, Dominican Republic, Mali, Lao People's Democratic Republic, Netherlands and Tanzania). Examination of the scatterplot shows a linear pattern. We computed the regression equation and used these to estimate the standard-deviation on 0-10 life-satisfaction. The formula is: Estimated SD 0-10 life satisfaction = 1.112 + 0.5552 x observed SD on the Best-Worst item. These estimates are reported in this table. The 95% confidence interval around these estimated values is about 1 point. which means that these estimates are quite rough.
This estimation technique is described in more detail in the Introductory Text. chapter 7 'Comparability of the data' in section 7/3.1 'Converting scores on measures of different happiness variants'.
Standard deviations are not yet available for the following countries of which average happiness is known: Central African Republic; Congo (Kinshasa),; Guatemala; Guinea; Haiti; Jamaica; Kuwait; Liberia; Malawi; Namibia; Sudan; Turkmenistan; United Arab Emirates; Yemen;
This list is included in the datafile 'States of nations' as variable SD_HappinessLSBW10.11_2000.09 .