Reliable and Unreliable Respondents : Party Registration and Prestige Pressure

• Published date: 01 March 1966
• Date: 01 March 1966
• DOI: http://doi.org/10.1177/106591296601900104
• Subject Matter: Articles

RELIABLE AND UNRELIABLE RESPONDENTS :
PARTY REGISTRATION AND PRESTIGE PRESSURE
CHARLES G. BELL, California State College at Fullerton,
AND
WILLIAM BUCHANAN, University of Tennessee
ALIFORNIA
LAW
provides for permanent registration of voters. Further-
~
more, registration is made easy. Deputy registrars set up cardtables on street
corners
and at supermarkets; political parties send patrols into those neighbor-
hoods where they are strongest to sign up partisans on their doorsteps. And, as a
result, a vigorous drive in Los Angeles County before the 1960 general elections in-
creased registration from 67.7 to 78.5 per cent of the adult population.’
In the summer of 1961, a sample of Los Angeles County residents was inter-
viewed for a study of residential mobility. A multi-stage probability sample of housing
units was used in view of the major objective of obtaining mobility data. In each
household one person, controlled by age rank, and sex, was interviewed.2
2
On the hour-long schedule dealing with housing, neighborhood preference, his-
tory of past moves, and expectation of future moves was a series of questions on vot-
ing behavior. These include: (a) standard questions about party identification; (b)
the question, &dquo;As of today, are you registered to vote in California?&dquo; If &dquo;yes,&dquo; then
( 1 ) &dquo;are you registered at this address?&dquo; (2) if registered elsewhere, give registration
address; (c) respondent’s party designation; (d) if not registered, data on any pre-
vious registration; (e) respondent’s report of his 1960 presidential vote; and, finally
(f) respondent’s name and address. Although the survey did not deal with voting
in the depth and detail a political study would have used, the questions were thor-
ough enough to check for internal inconsistencies and haphazard responses.
Registration data in Los Angeles County are kept up by career ofhcials using
advanced filing techniques and electronic equipment. Therefore, it was possible to
search the registrar’s files to discover which of the respondents who claimed to be
registered voters were actually registered and the accuracy of their reported party
registration.3
NOTE: The investigation was supported by grants from the John Randolph and Dora Haynes
Foundation of Los Angeles. The political portion of the survey was under the direction
of Professors T. J. Anderson and William Buchanan; the mobility study, under the direc-
tion of Professors Georges Sabagh and Maurice D. Van Arsdol, Jr. -
all of the University
of Southern California. Professor Buchanan subsequently accepted a position at the Uni-
,
versity of Tennessee.
1
Los Angeles County population over 21 years of age, 1960 census, was 3,830,936. The total
number of registered voters has varied considerably: in June 1960, it was 2,548,971; in
November 1960, it was 3,011,279; following the legally required post-election "purge" of
the lists for failure to vote, it was 2,666,711; in January 1962, it was 2,714,267. Thus,
the percentage of registered adult population varied from 67.7 per cent in June 1960 to
78.5 per cent in November 1960, and then dropped back to 69.9 per cent following the
"purge." Registration in mid-1961 was estimated at 70 per cent on the basis of these
figures.
2
Sampling method is described in detail in a publication available from the Population Research
Laboratory, University of Southern California.
3
Grateful acknowledgment is made to County Registrar of Voters Benjamin S. Hite and to his
deputy, James Allison, for their generous assistance.
37

---

38
There were 769 respondents in the Los Angeles County sample, of whom 23
were eliminated because there was doubt of identity: e.g., was the...