The Mobilization Hypothesis and Voter Turnout in Congressional Elections, 1974-1982

• Published date: 01 June 1988
• DOI: 10.1177/106591298804100204
• Date: 01 June 1988
• Subject Matter: Articles

THE MOBILIZATION HYPOTHESIS AND VOTER
TURNOUT IN CONGRESSIONAL ELECTIONS, 1974-1982
PRISCILLA L. SOUTHWELL
University of Oregon
FTER
nearly a quarter-century of declining voter turnout, the 1982
congressional elections reversed this trend. According to surveys
jL
conducted by the University of Michigan, voter turnout in 1982
was 4.4 percentage points higher than in 1978.’ One interpretation of
this upswing is that the &dquo;have-nots&dquo; of American society surged to the
polls in 1982 in response to unfavorable changes in federal economic
policy under the Reagan administration (Cook 1983; Clark 1984). Spe-
cifically, blacks, working class citizens, and the unemployed all turned
out in greater numbers than in previous elections. Table 1 shows the per-
centage voting for each of these groups in both years. Although partici-
pation by such individuals still lags behind that of their white, employed
or middle-class counterparts, the sharpest increases in turnout occurred
among these less privileged individuals.
This &dquo;have-not&dquo; interpretation of the higher turnout in 1982 makes
the basic assumption that those individuals who experience some type
of economic difficulty will have an added incentive to get out and vote.
This point of view is often referred to as the &dquo;mobilization&dquo; hypothesis
(Rosenstone 1982). This paper attempts to explain the increased voting
among blacks, working class and unemployed individuals in 1982 by em-
phasizing the importance of context. We suggest that the &dquo;mobilization&dquo;
hypothesis is supported in certain election years, such as 1982, where
the combination of pressing economic problems and the nature of party
control of the White House created a situation in which a clearly iden-
tifiable &dquo;target&dquo; existed for the disgruntlement of those experiencing eco-
nomic adversity. Secondly, we emphasize that the contextual nature of
this motivation hypothesis has important methodological, as well as the-
oretical, implications. Previous models of the voting decision have meas-
ured only the effect of changes in distribution of a variable from year
to year rather than changes in the relationship of the variable to turnout
over time. The latter approach is appropriate for evaluating the mobili-
zation hypothesis, because this hypothesis assumes that a special set of
Received: June 12, 1987
First Revision Received: September 30, 1987
Accepted for Publication: October 5, 1987
1
Census data for these years also show increased voting among blacks, working class in-
dividuals, and the unemployed in 1982. Although Census data would provide a larger
sample of such individuals, we chose to rely on the SRC/CPS data because of the more
evaluative content of certain questions. The data were made available by the Inter-
University Consortium for Political Research. Neither the Center for Political Studies
nor the Consortium bears any responsibility for the analysis or interpretation presented
here.

---

274
TABLE 1
VOTER TURNOUT IN THE 1978 AND 1980 CONGRESSIONAL ELECTIONS BY RACE,
EMPLOYMENT STATUS, AND CLASS
ns = x2 not significant
*p<.10> **p<.05> ***p<.o1> circumstances encouraged voting among those experiencing economic
adversity in 1982. As a result, we can expect that certain variables, such
as race, employment status, and class, will have a different relationship
to voter turnout from that of previous years.
PREVIOUS RESEARCH ON ECONOMIC VOTING
The mobilization hypothesis flies in the face of conventional wisdom
with regard to economic voting. Previous researchers who have analyzed
the relationship between an individual’s personal economic condition
and his voting decision have found that persons experiencing financial
difficulties are, instead, less likely to vote (Caldeira, Patterson, and Markko
1985; Rosenstone 1982; Wolfinger and Rosenstone 1980; Brody and
Sniderman 1977). Such individuals are less likely to vote because the
stressful nature of economic adversity forces a preoccupation with such
personal economic problems and as a result makes the individual with-
draw from political or community matters.
Other researchers have rejected any relationship between economic
well-being and political participation by concluding that personal eco-
nomic problems have no effect on an individual’s decision on whether

---

275
or not to vote (Arcelus and Meltzer 1975; Conway 1981; Fiorina 1978;
Kinder and Kiewiet 1979). According to this argument, the ethic of self-
reliance in American culture encourages people to regard personal eco-
nomic problems as their own responsibility, and their &dquo;focus of expec-
tations&dquo; is not upon the government to deal with these difficulties (Brody
and Sniderman 1977). As Kinder and Kiewiet conclude, &dquo;Economic dis-
contents and
political judgments inhabit separate mental domains&dquo; (1979:
523).
This paper attempts to deal with these differing viewpoints by estimat-
ing several models of the voting decision which include certain key eco-
nomic variables. The topic of this paper thus overlaps considerably with
theories of &dquo;economic voting,&dquo; which address the question of how voters’
evaluations of the prior and current state of the economy shape subse-
quent vote choices.2 The main purpose here, however, is to evaluate the
effect of race, class, and employment status on voter turnout rather than
on partisan preference. This approach is similar to that of Arcelus and
Meltzer (1975) and Rosenstone (1982) who suggested that short-term eco-
nomic fluctuations may affect whether a person votes as well as how he
votes.3 3
CONTEXTUAL CONSIDERATIONS AND ECONOMIC ADVERSITY
On September 5, 1982, the Department of Labor issued its monthly
unemployment report which showed that the number of jobless Ameri-
cans had risen to 10.1 percent of the labor force. Although this two-digit
figure was not unexpected, the announcement sent political shock waves
across the country. The state of the economy had become salient again,
and it appeared that economic issues would dominate the upcoming con-
gressional elections. As Robert Teeter, Republican pollster, lamented at
the time, &dquo;This is an election of money issues; unemployment is the fo-
cal point&dquo; (Mann and Ornstein 1983: 38).
The conditions seemed ripe for the mobilization of those experienc-
ing economic adversity in 1982. When personal problems become
&dquo;socially-located,&dquo; they may activate turnout. As noted by Brody and
Sniderman:
Such problems reflect a looking outward, a directing of attention to-
ward the connections between self and society. Concern about this sort
of problem (e.g. inflation) should underline the salience of politics to the
individual, and therefore promote involvement (1977: 338).
2

See Bloom and Price 1975; Fiorina 1978; Goodman and Kramer 1975; Kinder and Kie-
wiet 1979; Kramer 1971, 1983; Tufte 1975; Weatherford 1978.
3
Retrospective voting models typically contain a measure of the individual’s evaluation
of his personal financial situation. The question in the American National Election Sur-
veys is: "Does R feel better off financially than a year ago?" We
have chosen to omit
this variable from the final analysis presented here although it shows a similar pattern
to the race and class variables. Retrospective economic well-being is only weakly related
to the policy evaluative aspect of our theoretical model for it does not distinguish be-
tween the effects on one’s financial situation due to unemployment or inflation.

---

276
However, the process by which economic problems,become socially
located is a complex one, and it occurs only under a particular...