We analyze the attitudes and behavior on which the modern welfare state rests. Such support is studied broadly, including how people define their economic interests, how they perceive the fairness of procedures and outcomes, and whether they trust and int eract with fellow citizens in civil society. The overriding question is how orientations are affected by local and national context. We analyze effects of local variation in Norway, but also compare a large number of European countries with each other. Fo ur groups of contextual factors are considered: (1) public and private affluence, (2) the level and structure of economic inequality, (3) variations in ethnic heterogeneity, and (4) variations in the mass mediated public sphere. We make important contribu tions to research on citizens in context. We simultaneously consider a broader range of attitudes and behavior than much research. We jointly consider several groups of contextual factors, analyzing how they work in combination. We unite studies of nation al and local context under one framework. Finally, we systematically include the public sphere as a source of contextual variation, focusing on how agenda-setting and framing at the elite level affect citizens in combination with external conditions. We c onduct a citizen survey in Norwegian municipalities with a panel design (waves in 2013-15) and a stratified sample. This survey measures all key concepts and can be matched with unique contextual data. The design allows separation of self-selection proces ses from genuine contextual causal impact. It also allows analysis of dynamic contextual changes. Moreover, we study election campaigns in eight countries asking how important issues of the welfare state have been framed in the public sphere by political actors. The impact of such variation is studied in randomized experiments in Norway, Sweden and Germany. This allows studies of how "agenda-setting" and "framing" interacts with contextual conditions.