Sears & Funk (1999) Evidence of the Long-Term Persistence of Adults' Political Predispositions

Sears and Funk (1999) Evidence of the Long-Term Persistence of Adults' Political Predispositions

long term
large sample
longitudinal

Abstract: The persistence hypothesis holds that core political pre-dispositions tend to be highly stable through the life span. It has rarely been tested directly, given the scarcity of long-term large-sample longitudinal studies. We address it through the Terman longitudinal study, in which the party identification and ideology of 1,272 respondents were measured on four occasions between 1940-1977, from roughly age 30 to retirement age. These partisan attitudes were lightly stable over this long period, yielding continuity coefficients of about .80 between each measurement (separated by at least 10 years) and .65 for the full 37-year span. Examination of the trajectories of individual attitudes reveals that the most common pattern was constancy across time. A substantial minority changed in small but consistent ways, but changes form one partisan side to the other were not very common. Surprisingly, early-life racial attitudes ha a resurgent effect on partisan attitudes in the 1970s. There was evidence of increasing attitude crystallization through the life span, infusing core pre-dispositions with increasing psychological strength over time. Limitations of the study include the high intelligence of the respondents and the "steady" state of the party system through most of this period.

Our primary goal in this paper is to test empirically the stability of core political predispositions across the full adult lifespan.
*incorporates a persistence hypothesis: the attitudes persist from early adulthood through adulthood.

The Data:
mail questionnaires distributed 1940, 1950, 1960, 1977
N= 1,272 (had to respond to at least 2 questionnaires)
measures:
pid, ideology, etc.


General Pre-Dispositional Model-
attitudes on certain objects become crystallized; as these attitudes become more crystallized, these attitudes serves as a predisposition for evaluating new attitude objects (issues or political candidates).
-even "poorly crystallized attitudes may be quite stable if they confront little challenges" (2).
-crystallization is not necessarily incremental. there can be periodic leaps forward triggered by information-rich environments or other external prodding.
-IMPORTANT!! "Not all attitudes fit the predisposition description. If persistence depends on the intensity and direction of information flow, it is likely to vary across attitude domains as well, because they vary a great deal in the amount of public attention they receive. This is why party id is more stable--its more salient. other issue domains exert less attitude stability, arguably because they suffer from low political salience, or they are complex (Feldman and Zaller 1992).
-priming plays a role; attitudes need to be primed to be activated, kind of? "The priming variant of the persistence hypothesis suggests that predispositions can change as old issues drop off the political agenda and new issues become more salient"


Findings:
For both PID and ideology, the most common pattern was to hold the same position at all three time points (3?). 59% were perfectly stable in PID and another 6% had returned to their original decision by the end. Thus 65% ended up where they started. 42% for ideology were at the same point at al three points; 12% defected for a short time, but the longitudinal study allows us to see that 54% wound up where they started from. "So the most common individual trajectory for either party identification or ideology was to hold the same basic preference at both the beginning and the end of adulthood, not withstanding the fact that a few had some sojourn elsewhere" (13).
The most movement came from those in the middle, moving from moderate conserve (lib) to moderate lib (conservative).

However, this describes change. it does not explain it.

AGENDA SETTING/PRIMING EFFECTS: What explains change can somewhat be attributed to the political agenda. For example, racial attitudes were measured in 1940; but they don’t have any PID or ideological predictive power until the 1977 survey. the women's movement questions as well, though smaller increase. "The effect of attitudes toward blacks also dropped steadily from 1940-1960, in a period of relative neglect of racial issues. but their effect surged considerably in 1977, and actually became somewhat more predictive of L/R orientation that in 1940, when originally measured.

AGE EFFECTS: Stability does indeed rise with age (18). as does attitude consistency (PID and ideology).

KNOWLEDGE LEVELS: those with more education (some college) are more likely to have stable attitudes (.71) than high school grads (.64). But these authors don’t make any confirming conclusions on the effects of education---the jury is still out.

Where does this leave us?
"It is noteworthy that their early-acquired political pre-dispositions show particularly high levels of persistence through their full adult life spans. The findings therefore provide a look at an important segment of the public over an unusually long and politically interesting period in American history" (21).

Conclusions:
As adults their party OD and political ideologies were remarkably stable over four widely spaced intervals between 1940 and 1977.

Changes did occur, but they were usually in a consistent direction. i.e. in PID 32% changed in consistent direction, as opposed to the 8% that did not change in a consistent direction. for ideology, 41% changed in a consistent direction; 17% were inconsistent.

the change is real and systematic, from 1940, 50, 60 to 77.

Shifts might be occurring on account of issue salience, as the priming hypothesis states. "Apparent attitude changes might therefore reflect changes in issue salience rather than change in the predispositions themselves, reflecting, in Asch's (1948) terms, "changes in the attitude object," rather than "changes in attitude" (23).

The long-term impact of early socialization is therefore not a psychological given, but depends on several facilitating conditions that are themselves much influenced by the ongoing political environment. One is that early-life experiences must yield crystallized attitudes, which should be most likely when exogenous events stimulate high levels of relevant communication. This occurs routinely for partisanship as part of the election cycle. Extraordinarily events such as the Kennedy assassination, the Watergate scandal, the protests of the 1960s, or the GD have also left lasting traces on both the memory and political attitudes of entire generations (23).

social environment matters. frequent reinforcement is important (as indicated by election cycle reinforcement of PID).
information flows matter. "Elites also need to frame political issues such that ordinary citizens can link them cognitively to relevant pre-dispositions (Kinder and Sanders 1996; Sears and Huddy 1992). Of course, alternate frames may evoke different pre-dispositions, so elites must compete to secure the most favorable framing. Such competition over the public agenda may advantage the partisan camp that succeeds in focusing attention on issues it owns" (Petrocik 1996).