Caroline Schaerer and Andrea Baranzini
To summarize, we observe that segregation indices are generally relatively low in Geneva and Zurich. Moreover, segregation appears to be more related to nationality than to the education level (and thus probably to income), although foreigners with low education level are those who are relatively more segregated. It should be emphasized however, that a low level of geographic concentration does not exclude that some specific population group might be particularly disadvantaged in terms of living conditions. Therefore, in the next section we analyse whether there are differences in the residential conditions depending on the origin and/or the education levels.
Analysing residential conditions: how are the people living?
In order to describe the living conditions and to test for quality differentials between dwellings occupied by different categories of individuals, we need informations on dwellings and buildings characteristics, as well as data measuring the neighbourhood quality, in addition to the individual socio-economic characteristics. Concerning building and dwelling characteristics, they are reported in the 2000 Swiss Popula- tion Census. However, since the same dwelling will appear more than once in the dataset for each household composed of more than one individual, we keep the household’s head only.3 In order to limit the scope of our analysis, we limit it to rented apartments. From the dataset, we thus drop the owners, the members of a housing cooperative, the single family houses, and the holders of special rent con- tracts (i. e. holders of a free-rent dwelling, holder of service dwelling, or of a farming lease). To this dataset we add several variables in order to measure environmental and accessibility characteristics of the building and the neighbourhood.
Firstly, we add location characteristics calculated using the Information System of the Geneva Territory (SITG) and the GIS-centre of the Zurich office of land use regulation and measurement (ARV), two very rich and well-developed GIS databases. Using these datasets, we calculate accessibility variables, which measure precisely the proximity of the buildings to environmental amenities and main public infrastructures, such as the distance to the city centre, to the nearest urban park and to the nearest public transportation stop. In addition, we define several neighbourhood characteristics (at the district level) quantifying the percent- age of different land-use features, such as the percentage of tree-covered area and the percentage of urban parks.
The household head is defined according to the following criteria, by order of priority: 1) an older individual is preferred over a younger one; 2) a full-time working individual is selected over a part time working, an unemployed, a retired individual, an individual in an education process, or over a individual who is not in the labor force; 3) an individual occupying an executive job is chosen over an individual with an independent activity, a intermediate job, an employee, a factory worker, or an apprentice.
© Seismo Verlag Zürich