All Academic, Inc. Research Logo

Info/CitationFAQResearchAll Academic Inc.
Document

Explaining Information Effects in Collective Preferences
Unformatted Document Text:  assertions, 6 questions with ordinal scales containing a middle response option, and questions that are part of a series sharing the same response scale, 7 since these format characteristics make it easy for respondents to give substantive responses with little cognitive effort. Similarly, item nonresponse rates are slightly higher for questions appearing later in a survey, 8 since respondent fatigue should be higher at the time such questions are asked. This much is in line with expectations. INSERT TABLE 1 ABOUT HERE However, while satisficing is expected to be more common among respondents with lower ability and motivation levels, it turns out that these question characteristics influence the lopsidedness of group opinions to similar degrees in both the highest and lowest knowledge quartiles. In each case, question formats known to produce satisficing behavior significantly increase the lopsidedness of group opinions. While these patterns are consistent with the predictions of the satisficing literature, the novel finding here is that the effect of question ordering is so similar among the most and least able respondents. And in the case of questions worded as assertions, respondents in the highest knowledge quartile appear to satisfice more than those in the lowest. Because both groups are affected in similar ways, none of these question attributes has a consistent or significant impact on the size of information effects, despite the fact that they are notorious in the literature for reducing the quality of survey responses. Information effects should become smaller to the extent that respondents across levels of political knowledge reliably satisfice when presented with such questions, because this form of satisficing should reduce differences in 6 The assertion variable was coded 1 for all questions using agree/disagree, willing/not willing, favor/oppose, for/against, or approve/disapprove formats, 0 otherwise. 7 A series is defined as three or more adjacent questions that share the same response scale. For the 1988 and 1992 data, this attribute was coded from copies of the actual survey instruments presented to respondents (available on the ANES CD-ROM), while for the 1996 data this attribute was coded from the NES codebook. 8 This variable was constructed by arranging sequentially every question stem (labeled with a letter and a number in the NES codebook) within each wave of the NES data and numbering them from 1 to n, with nrepresenting the last question in the survey instrument. The identifying letter and number combination for each question was obtained from the questionnaires archived on the NES CD-ROM for the 1988 and 1992 data, and from the codebooks for the 1996 data. For each wave, the number of each question was divided by the total number of questions in the wave, to produce a variable ranging from approximately 0 to 1, with lower scores representing items that appeared relatively early in the questionnaire and higher scores representing later questions.

Authors: Althaus, Scott.
first   previous   Page 9 of 53   next   last



background image
assertions,
6
questions with ordinal scales containing a middle response option, and questions that are
part of a series sharing the same response scale,
7
since these format characteristics make it easy for
respondents to give substantive responses with little cognitive effort. Similarly, item nonresponse
rates are slightly higher for questions appearing later in a survey,
8
since respondent fatigue should be
higher at the time such questions are asked. This much is in line with expectations.
INSERT TABLE 1 ABOUT HERE
However, while satisficing is expected to be more common among respondents with lower
ability and motivation levels, it turns out that these question characteristics influence the
lopsidedness of group opinions to similar degrees in both the highest and lowest knowledge
quartiles. In each case, question formats known to produce satisficing behavior significantly increase
the lopsidedness of group opinions. While these patterns are consistent with the predictions of the
satisficing literature, the novel finding here is that the effect of question ordering is so similar among
the most and least able respondents. And in the case of questions worded as assertions, respondents
in the highest knowledge quartile appear to satisfice more than those in the lowest.
Because both groups are affected in similar ways, none of these question attributes has a
consistent or significant impact on the size of information effects, despite the fact that they are
notorious in the literature for reducing the quality of survey responses. Information effects should
become smaller to the extent that respondents across levels of political knowledge reliably satisfice
when presented with such questions, because this form of satisficing should reduce differences in
6 The assertion variable was coded 1 for all questions using agree/disagree, willing/not willing, favor/oppose,
for/against, or approve/disapprove formats, 0 otherwise.
7 A series is defined as three or more adjacent questions that share the same response scale. For the 1988 and
1992 data, this attribute was coded from copies of the actual survey instruments presented to respondents (available
on the ANES CD-ROM), while for the 1996 data this attribute was coded from the NES codebook.
8 This variable was constructed by arranging sequentially every question stem (labeled with a letter and a
number in the NES codebook) within each wave of the NES data and numbering them from 1 to n, with n
representing the last question in the survey instrument. The identifying letter and number combination for each
question was obtained from the questionnaires archived on the NES CD-ROM for the 1988 and 1992 data, and from
the codebooks for the 1996 data. For each wave, the number of each question was divided by the total number of
questions in the wave, to produce a variable ranging from approximately 0 to 1, with lower scores representing items
that appeared relatively early in the questionnaire and higher scores representing later questions.


Convention
All Academic Convention is the premier solution for your association's abstract management solutions needs.
Submission - Custom fields, multiple submission types, tracks, audio visual, multiple upload formats, automatic conversion to pdf.
Review - Peer Review, Bulk reviewer assignment, bulk emails, ranking, z-score statistics, and multiple worksheets!
Reports - Many standard and custom reports generated while you wait. Print programs with participant indexes, event grids, and more!
Scheduling - Flexible and convenient grid scheduling within rooms and buildings. Conflict checking and advanced filtering.
Communication - Bulk email tools to help your administrators send reminders and responses. Use form letters, a message center, and much more!
Management - Search tools, duplicate people management, editing tools, submission transfers, many tools to manage a variety of conference management headaches!
Click here for more information.

first   previous   Page 9 of 53   next   last

©2012 All Academic, Inc.