Considerable research focuses on explaining trends in party identification
in the American public. Somewhat less attention has been devoted to trends
in ideological identification, although such research has been on the incre
ase. However, the potential relationship between macrapartisanship and macr
oideology remains largely unexamined We use alternative methodologies to te
st whether shifts in macropartisanship have occurred independently of, alon
g with, before, or after shifts in macroideology. We examine the time serie
s properties of the two series, which provides insight into their persisten
ce and memory. Our framework provides a flexible approach to studying short
- and long-run behavior, and the evidence is consistently at variance with
the idea that there is a relationship between macroideology and macropartis
anship. Thus, shifts in party identification and ideology are not driven in
response to a set of common factors, nor can shifts in one partisan direct
ion or the other be interpreted as ideological mandates at the aggregate le
vel.