We assess the tendency for the public to use group-centric policy evaluations with evidence from a survey experiment concerning two issues within the social policy domain, health care and aid to cities. By randomly varying target group identity within each issue and using both negatively and positively regarded groups our evidence shows that differences exist in the tendency for members of the public to use group-centric heuristics. Group-centric evaluations are related to party identification and political ideology. Across both issues conservatives and Republicans are more likely than liberals or Democrats to adopt a group-centric heuristic. Partisan and ideological differences suggest that established theories miss the mark by emphasizing how universal policy designs are preferred to designs that target unpopular groups.