Methodological Notes

Basic structure of the household income variables

Exceptions: in EE00, IT04/08/10, and PL95, the total gross income and income taxes and contributions are available, while the sub-components are net. In the French series the income variables report amounts that are gross of taxes and net of social security contributions.

Socio-demographic and employment characteristics

Individual-level socio-demographic and employment characteristics variables are available for a pre-defined group of the population (see Universe column in List of Variables). When creating these variables out of the underlying microdata, care was taken in suppressing the information each time the variable was missing for more than 10% of the cases included in the universe. This action was taken to avoid that the aggregate-level figure would be based on a non-representative portion of the population.

However, those variables should still be used with care when directly comparing their distributions across datasets (whether over time within countries, or across countries at one point in time). These variables are best used as additional measures to be included in the comparison of income statistics (e.g. mean wage income by education level, poverty rates by gender, or by employment status).

Income Conversion using 2011 Purchasing Power Parities (PPPs)

All income variables have been expressed from nominal local currency units to 2011 international dollars. Converting income amounts in PPP terms is common in comparing incomes across countries and results in incomes that hold roughly equal purchasing power measured in international dollars. The conversion was done by applying first a national consumer price deflator to the nominal amounts to express them in terms of year 2011 prices. Those amounts were then converted to international dollars using PPPs. The national deflators and PPPs were taken from the World Development Indicators from the World Bank.
PPPs and deflators.


The Web Tabulator gives you the choice to produce results that are either weighted or unweighted. Frequency tables should be unweighted for the simple sample frequency counts, and weighted for population frequency counts and frequency percentages. Income statistics tables should always be weighted in order to obtain results which are representative of the total population.

Level of the analysis

Please note that all variables are reported at the individual level, so that even if only household variables are selected (e.g. poverty by tenure of dwelling), the results will be based on individual-level numbers (e.g. frequencies of poverty by tenure of dwelling will be expressed as numbers of persons living in poor households by tenure of the dwelling).

In addition, whenever the income statistics are analysed with one (or more) individual-level variables, the results will be displayed only for the defined universe of the individual-level variable(s) (e.g. mean wage by employment status will refer to the working-age population only).