Summary of Major Voter Behavior Studies

1. Quantitative Methods of Politics (1928) - Stuart Rice -
based on aggregate voting statistics; first effort to connect
quantitative research on voting behavior with such social
science phenomenon as social change and attitude

2. The People's Choice (1944) - P.F. Lazarsfeld, B.R. Berelson,
and Hazel Gaudet - sought to study impact of political
campaigns on voters' choices among competing candidates;
carried out in Erie County, Ohio; used a panel of 600;
interviewed panel 7 times before and after 1940 presidential
election; findings were: (1) partisan loyalties stymied the
effect of campaign efforts to convert voters; (2) religion,
social status, and place of residence were important
predictors of voting choice; (3) media had little direct impact
on voter choice; and (4) media did stimulate partisan
activists, who in turn made a stronger commitment

3. Voting (1954) - P.F. Lazarsfeld, Bernard Berelson, and
William N. McPhee - located in Elmira, N.Y.; used a l000 voter
panel; dealt with 1948 election; found cross-pressured
persons to be late deciders; discovered perceptual screening
by which voters held a distorted perception of a candidate's
or party's position to try to reduce the degree of
disagreement with our views; led authors to conclude that
political disinterest and inactivity make a democratic
political system more stable and flexible than would be the
case with high levels of citizen involvement

4. University of Michigan Survey Research Center (Center for
Political Studies) "American Voter Model" studies: The
Voter Decides
(1954) - Angus Campbell, Gerald Gurin, and
Warren Miller: The American Voter (1960) - Campbell et al.;
Elections and the Political Order(1966) - Campbell et al. -
dealt with 1952 and 1956 presidential elections, developed
concept of "funnel of causality" (be able to label on final
exam); findings: (l) party loyalty most important factor,
(2) independents were low on political involvement, (3)
most citizens limit political activity to voting; (4) voters are
not rational citizens but rather uninvolved and uninformed,
blind partisans - know four points of American voter model
(p. 22 - Hill & Luttbeg)

5. Philip Converse - "normal vote" concept - argues that partisan
attachment is the prevailing long-term force in voting;
short-term forces can prevail over partisanship and cause a
deviation from the normal vote; two major short term forces
are candidate images and issue orientations of voters

6. Classification of presidential elections:
(l) maintaining - one in which the normal vote is maintained
and therefore the majority party captures the presidency
(2) deviating - occurs when short-term forces cause many
majority party voters to support the minority party's
(3) realigning - characterized by a "more or less durable"
shift in the underlying pattern of partisanship in the
electorate with the minority party becoming the majority

7. Converse study of non-issue oriented voter (l950s) - (l) most
Americans do not hold truly meaningful political attitudes -
no coherent "belief system"

8. The Responsible Electorate (l966) - V.O. Key - took issue with
"American voter model"; argued voters were conscious
decision makers rewarding and punishing candidates based
on expected benefits

9. Partisanship decline theory - (l) growth in split ticket voting,
(2) growth in switch voting; (3) growth in percentage
declaring themselves "independents", (4) increase in issues
and candidate images in voting behavior; (5) the decline in
influence of parties in elections

10. Growth in political alienation - (l) reduced trust in
government, (2) reduced sense of political efficacy, (3)
growth in cynicism of political process and political leaders
was evident in the l970s. It has declined somewhat in the

11. The Gallup Polls note that in all presidential elections
between l968-l980 over 50 percent of the electorate were
split-ticket voters.

12. In The Responsible Electorate, Key developed a topology of
voters, dividing them into three categories: standpatters,
switchers, and new voters. Key felt that switchers, because
of their willingness to cross party lines, represented an
important strain of independency. In contrast to The
American Voter, he did not see the "independent" as
necessarily uninformed politically, with low political
motivation. Other studies since have reinforced this view.
Gerald Pomper in Voter's Choice (l975) found independent
identifiers to be as knowledgeable about government as weak
party identifiers, though not as knowledgeable as strong
party identifiers. Too, differences between level of political
activity was not great. Other research has divided
independents into two categories: those who admit a party
leaning and those who do not. The former fit Key's switcher,
while the latter tend to fit the model of the independent
described by The American Voter.

13. The American Voter identified three main motivating forces
for a voter: - issues, images, and partisanship. The latter
was found to be the strongest force. In order for a person to
engage in "issue voting," three conditions had to be present.
First, the voter had to know and understand the issue.
Second, the issue had to be important to the voter. Third,
the voter had to be able to distinguish the candidates'/
parties' positions on the issue. In the l950s issue voting was
not very high and surveys found few voters to be "issue
ideologues." Subsequent researchers have claimed that the
l950s had few issues thus the conclusion about low issue
awareness was time bound. Events of the l960s and l970s
did polarize voters and raise the level of issue voting.
Voters have come to be more aware of policy differences
between the parties and to more correctly perceive these
party differences.

14. In l974 Norman Nie and Kristi Anderson, in a follow up to
Converse's earlier study of belief systems, found a striking
improvement in ideological constraint in the l960s. Too,
they found that levels of ideological consistency had risen
more for domestic policy issues than for foreign policy
issues. Too, studies by the SRC/CPS in the l970s also found
a similar trend. They argued that political events of the
'60s and '70s had led to this shift. However, other studies
have taken issue with these findings. Two studies in the
late '70s charged that Nie's findings were a result of changes
in the format of the SRC/CPS survey questions. Too, Converse
and Gregory Markus, after comparing a panel study of
1952-56 with a panel study of l972-76 on similar issues,
found little increase in issue consistency.

15. While there has been an evident increase in issue voting, it
has been accompanied by an increase in candidate-image
voting. This, according to Eugene De Clercq, Thomas Hurley,
and Norman Luttbeg, had outstripped growth in the role of
issues between l956-72. In another study, the authors also
introduced a fourth voter motivation variable - party image
(as separate from party identification). The impact of these
four variables on voter behavior through l980 was studied.
Based on trend lines, the authors conclude, "It is difficult
to predict the future importance of party identification or
any other of the four factors analyzed, other than to say
that issues are slowly and consistently rising in importance
over time. It seems likely that the erratic performance of
these factors over the last four elections will continue as we
enter an era of dynamic electoral behavior in which the
absolute and relative effects of partisanship, issues, and
images will fluctuate."

16. A number of factors have contributed to a decline in the
influence of political parties. These include: (l) loss of
patronage power, (2) non-partisan elections, (3) direct
primaries, (4) new campaign technology, (5) candidate-
centered campaigns, and (6) party convention reform. Jack
Dennis, using longitudinal analysis, found that in recent years
Americans have become alienated from the parties and the
party system. Also, the pollster Louis Harris, in l973, found
that Americans were not likely to see parties as active or as
agents of change.

17. Party realignments have been described as cyclical in nature,
reflecting the dissolution of old coalitions and building of
new ones, and also as reflective of critical elections in
American history when new forces or issues created a crisis
in the political order. From a time perspective, America is
ripe for a new party alignment from the old New Deal
coalition that made the Democrats the majority party.
Reasons noted have been (l) the increase in independents,
(2) the success of third parties, (3) the increase in issue
voting, (4) the declining confidence in the parties, (5)
reduced impact of transmission of party labels from parents,
and (6) the crumbling of the social group foundations of the
New Deal coalition. However, some scholars doubt if a
realignment is imminent because a necessary component of
the realignment process is missing. These scholars believe
that realignment can occur only after a party perceived as
unsuccessful in governing is replaced by a party which is
ultimately perceived as successful in governing where the
former party failed. Because "successful" governance today
would require control of all branches of government
simultaneously, and because government's problems might
be unsolvable, at least to the public's satisfaction, the
requirement for realignment is not likely to be met.

18. Factors that might be considered in assessing whether or not
there has been a realigning election include the following:
(l) percent of popular vote won
(2) number of states won
(3) location of states won
(4) electoral votes won
(5) number of House/Senate seats won
(6) number of governorships won
(7) percent of voter turnout
(8) socio-economic description of supporters
(9) former political leaning of supporters
(l0) reasons stated for supporting

19. One major area considered likely for realignment is the South.
James Sunquist sees the South dividing in the future on class
lines similar to the rest of the country. Carol Cassel refutes
this thesis. Her study of native white Southerners over the
period l952-l972 found that group moving away from its
traditional Democratic partisanship toward the independent,
not Republican label. Philip Converse argues that most change
in Southern partisanship is not due to political conversions,
but instead the relocation of northern Republicans as industry
and commerce move south of the Mason-Dixon Line. Of
course, in l976, Carter ran very well in the South; however,
in l980 and l984, the South went strongly Republican. Some
have argued that it is not how a person identifies himself/
herself but how that person votes. Based on presidential
electoral votes, the South has seemed to move into the
Republican column; however, at the state-level the returns
are mixed, not only between states but within as well.

20. Some scholars feel what we are in is a period of dealignment
when parties are ceasing to be very relevant as voter cues.
And, some feel that the emergence of a new party system in
the near future is unlikely.

21. Looking at voter trends, one might reach the following
conclusions: (l) While both major parties have lost weak and
strong supporters to the ranks of independents, most of the
independents continue to at least admit to some partisan
leaning and to vote in that direction. (2) Only the "pure"
independents reflect a change from the l950s. (3) Party
defectors show no unique increase in the quality of their
candidate evaluations. Too, the rates of defection between
l952-l980 have been erratic and reflect no overall trend.
(4) Voter sophistication, as judged by the use of issues or
ideology in evaluating candidates, has improved, but it is
neither the cause nor the result of disenchantment of the
public with the major political parties.

22. Polsby & Wildovsky offer the following observations about
voter behavior: (l) The number of issue-oriented "pure"
independents represents a very small percent of the voting
populace. (2) Switchers are likely to be concentrated among
those who receive some but not too much information about
parties, issues, and candidates. (3) Most changes in party
identifications involve switching in and out of the
independent category rather than between the parties.

23. In recent years interest has developed in what has come to
be termed as "retrospective voting". By this term, it is meant
that when a voter goes in to cast a vote she/he will cast that
vote based upon the citizens assessment of the candidate on
party's past performance.

24. Gerald Pomper and Susan Lederman have put forth in more
recent time, in Elections in America (2nd ed., l980), the
concept of the voter as "the protective meddler." In this
structure party is important but not overpowering. Voter's
involvement is more in the nature of meddling rather than
continuing participation. This meddling comes when the
voter becomes concerned. Political efforts at persuasion
have limited success due to the voter's level of self-interest
and often indifference. Political messages must be consonant
with perceived interests to be heard. Party loyalty can be
shed when appropriate. Voters make their individual
assessment on a few issues, those relevant to them. By
looking back at prior performance retrospectively and by
projecting future expectations, voters will express their will.
This meddling allows the voter to exercise indirect popular
influence over public policy by determining who will be in
power to make that policy.

Dr. C's Course Links