Based on a survey of 707 residents of a deprived area in Cologne, Germany, the paper assesses whether the area is internally heterogeneous. The theoretical approach is based in a multi-level model combining census data, such as percentage of social assistance and migration rate, further social collective efficacy and intergenerational closure at the meso level, and several variables at the micro level, such as perceived deviant behaviour, disorder and willingness-to-move out. Decomposing the total area into six neighbourhoods, considerable variation is found in the extent of social mix between these areas. Further, heterogeneity varies by indicator of social mix. Due to this heterogeneity, effects on dependent variables differ significantly. Further, neighbourhood exhibits different means on all variables but comparatively similar values of heterogeneity. Since multi-level analyses could not be applied due to the small number of neighbourhoods, correspondence analysis is suggested as a method to specify and visualize the relations between the variables of different levels. Correspondence analysis reveals the six neighbourhoods to have different levels of norms and different correlations with the explanatory variables, such as transfer income. Two major conclusions emerge from the analyses. First, seemingly homogeneous deprived areas are internally heterogeneous, which may account for the inconsistent findings on social mix on behavioural outcomes. Second, studies and policies of social mix should give more attention to the problem of dimensions (and their combination) of social mix.
In an experimental split ballot design, I test four different ranking techniques (drag and drop, numbering, arrows, and most–least) to explore potential effects on substantive answers, dropouts, and item nonresponse and response time between the groups. As an example, I use six items from Inglehart’s materialism–postmaterialism index. Data come from 1,225 members of an access panel who entered the set of items to be rank ordered. With respect to sex, education, and age, there are no significant differences between the four experimental groups. However, the groups differ extensively in response time, item nonresponse, and estimation of the percentage of materialists and postmaterialists. Drag and drop is shown to be the best-suited method for collecting rank data in web surveys.