Prior Beliefs and Voter Turnout in the 1986 and 1988 Congressional Elections

• Author: Stephen P Nicholson,Ross A. Miller
• Published date: 01 March 1997
• Date: 01 March 1997
• DOI: http://doi.org/10.1177/106591299705000110
• Subject Matter: Articles

199

Prior

Beliefs

and

Voter

Turnout

in

the

1986

and

1988

Congressional

Elections

STEPHEN

P

NICHOLSON,

UNIVERSITY

OF

CALIFORNIA,

DAVIS

ROSS

A.

MILLER,

SANTA

CLARA

UNIVERSITY

The

costs

of

political

information

differ

between

and

among

elections.

In

those

elections

where

information

costs

are

high,

voters

should

rely

on

information

from

previous

elections.

Although

research

on

voter

choice

has

long

recognized

that

voters

use

past

information

in

their

assessments

of

candidates,

studies

of

voter

turnout

are

solely

concerned

with

informa-

tion

available

in

the

current

election.

Specifically,

the

closeness

of

elections

is

a

central

concern

in

rational

actor

models

of

voter

turnout.

As

such,

these

studies

neglect

the

effects

of

prior

electoral

competitiveness

on

citi-

zens’

decisions

to

vote.

In

this

study

we

propose

that

actors

rely

on

prior

beliefs

when

deciding

to

vote

in

information-poor

elections.

Controlling

for

information

available

in

the

current

election,

we

explore

this

possibility

in

the

1986

and

1988

House

elections

and

find

that

prior

beliefs

have

a

significant

effect

on

turnout.

The

closeness

of

elections

is

central

to

rational

actor

approaches

to

voter

turnout.

In

the

dominant

rational

choice

approach,

the

calculus

of

voting

(Riker

and

Ordeshook

1968),

citizens

base

their

decisions

to

vote,

in

part,

on

the

probability

that

their

vote

will

affect

the

electoral

outcome.

However,

schol-

ars

have

implicitly

assumed

that

the

amount

of

information

available

to

voters

in

assessing

the

closeness

of

elections

is

equivalent

across

and

among

elec-

tions.

This

assumption

is

problematic

since

the

amount

of

information

avail-

able

in

an

election

varies

with

office

and

the

profile

of

the

contest.

Aldrich

NOTE:

An

earlier

draft

of

this

article

was

presented

at

the

Western

Political

Science

Asso-

ciation

Annual

Meeting,

Albuquerque,

New

Mexico,

March

10-12, 1994.

We

would

like

to

acknowledge

the

helpful

comments

of

Nathaniel

Beck,

Scott

Gartner,

Gary

Jacobson,

Robert Jackman,

Gary

Segura,

Randy

Siverson,

Roger

Rose,

and

Michael

Wilkening.

In

addition,

we

would

like

to

thank

Walter

Stone

and

the

anonymous

reviewers

of

the

Political

Research

Quarterly.