Feeling thermometers versus 7-point scales: Which are better?

Duane Francis Alwin

Research output: Contribution to journalArticle

135 Citations (Scopus)

Abstract

This study addresses the issue of the relation between the number of response categories used in survey questions and the quality of measurement. Several hypotheses, derived from relevant theory and research, are tested through a comparison between 7-and II-category rating scales used in the 1978 Quality of Life Survey. One hypothesis derived from information theory, that rating scales with more response categories transmit a greater amount of information and are therefore inherently more precise in their measurement, is strongly supported. A second hypothesis, that questions with greater numbers of response categories are more vulnerable to systematic measurement errors or shared method variance, is rejected. This study supports the conclusion that questions with more categories are both more reliable and more valid.

Original languageEnglish (US)
Pages (from-to)318-340
Number of pages23
JournalSociological Methods and Research
Volume25
Issue number3
DOIs
StatePublished - Jan 1 1997

Fingerprint

rating scale
information theory
quality of life

All Science Journal Classification (ASJC) codes

  • Social Sciences (miscellaneous)
  • Sociology and Political Science

Cite this

Alwin, Duane Francis. / Feeling thermometers versus 7-point scales : Which are better?. In: Sociological Methods and Research. 1997 ; Vol. 25, No. 3. pp. 318-340.
@article{bcbbf417bdd54adaafaae61b5723f950,
title = "Feeling thermometers versus 7-point scales: Which are better?",
abstract = "This study addresses the issue of the relation between the number of response categories used in survey questions and the quality of measurement. Several hypotheses, derived from relevant theory and research, are tested through a comparison between 7-and II-category rating scales used in the 1978 Quality of Life Survey. One hypothesis derived from information theory, that rating scales with more response categories transmit a greater amount of information and are therefore inherently more precise in their measurement, is strongly supported. A second hypothesis, that questions with greater numbers of response categories are more vulnerable to systematic measurement errors or shared method variance, is rejected. This study supports the conclusion that questions with more categories are both more reliable and more valid.",
author = "Alwin, {Duane Francis}",
year = "1997",
month = "1",
day = "1",
doi = "10.1177/0049124197025003003",
language = "English (US)",
volume = "25",
pages = "318--340",
journal = "Sociological Methods and Research",
issn = "0049-1241",
publisher = "SAGE Publications Inc.",
number = "3",

}

Feeling thermometers versus 7-point scales : Which are better? / Alwin, Duane Francis.

In: Sociological Methods and Research, Vol. 25, No. 3, 01.01.1997, p. 318-340.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Feeling thermometers versus 7-point scales

T2 - Which are better?

AU - Alwin, Duane Francis

PY - 1997/1/1

Y1 - 1997/1/1

N2 - This study addresses the issue of the relation between the number of response categories used in survey questions and the quality of measurement. Several hypotheses, derived from relevant theory and research, are tested through a comparison between 7-and II-category rating scales used in the 1978 Quality of Life Survey. One hypothesis derived from information theory, that rating scales with more response categories transmit a greater amount of information and are therefore inherently more precise in their measurement, is strongly supported. A second hypothesis, that questions with greater numbers of response categories are more vulnerable to systematic measurement errors or shared method variance, is rejected. This study supports the conclusion that questions with more categories are both more reliable and more valid.

AB - This study addresses the issue of the relation between the number of response categories used in survey questions and the quality of measurement. Several hypotheses, derived from relevant theory and research, are tested through a comparison between 7-and II-category rating scales used in the 1978 Quality of Life Survey. One hypothesis derived from information theory, that rating scales with more response categories transmit a greater amount of information and are therefore inherently more precise in their measurement, is strongly supported. A second hypothesis, that questions with greater numbers of response categories are more vulnerable to systematic measurement errors or shared method variance, is rejected. This study supports the conclusion that questions with more categories are both more reliable and more valid.

UR - http://www.scopus.com/inward/record.url?scp=0031537636&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031537636&partnerID=8YFLogxK

U2 - 10.1177/0049124197025003003

DO - 10.1177/0049124197025003003

M3 - Article

AN - SCOPUS:0031537636

VL - 25

SP - 318

EP - 340

JO - Sociological Methods and Research

JF - Sociological Methods and Research

SN - 0049-1241

IS - 3

ER -