This article analyzes the growing impact of partisanship on citizens' evaluations of the president's job performance at the individual level. The literature seeking to explain variation in presidential approval has analyzed aggregate trends over time. Although studies of aggregate trends have contributed to our understanding of the conditions that influence presidential approval, they are unable to model the individual‐level process underlying the aggregate results. The authors estimate the effects of individual citizens' party identification and retrospective and prospective economic evaluations on presidential approval in thirteen national samples over the two and one‐half decades from 1972 to 2000. It is found that party identification has a stronger influence on evaluations of the president's job performance in the period since 1982 than in the 1970s. Evidence is also found of an increasing tendency for partisanship to influence assessments of the economy in the period since 1982.