Effects of Rating Scale Direction under the Condition of Different Reading Direction

Because response scales serve as orientation for respondents when mapping their answers to response categories, it can be expected that the decremental (from positive to negative) or incremental (from negative to positive) order of a response scale provides information that influences response behav...

Full description

Bibliographic Details
Main Authors: Daagmar Krebs, Yaacov G. Bachner
Format: Article
Language:English
Published: GESIS - Leibniz-Institute for the Social Sciences, Mannheim 2018-01-01
Series:Methoden, Daten, Analysen
Subjects:
Online Access:https://mda.gesis.org/index.php/mda/article/view/2017.08/223
Description
Summary:Because response scales serve as orientation for respondents when mapping their answers to response categories, it can be expected that the decremental (from positive to negative) or incremental (from negative to positive) order of a response scale provides information that influences response behavior. If respondents interpret the first category on a scale as signifying “most accepted,” then starting an agree/disagree scale with “agree completely” or “disagree completely” may result in their forming different subjective hypotheses about the “most acceptable” response. If this principle applies in general, respondents’ reactions to horizontal response scales with different orders of response categories should be similar in the two directions of reading – right to left or left to right. This paper tests two hypotheses: first, that decremental scales elicit more positive responses than incremental scales; second, that this pattern holds under the condition of different reading direction. These hypotheses were tested using a German and an Israeli student sample. Seven-point decremental and incremental scales were applied in each sample; only the scale endpoints were verbally labeled. The questions asked related to extrinsic and intrinsic job motivation and achievement motivation. For data collection, a split-ballot design with random assignment of respondents to decremental and incremental scales was applied in both samples. Results revealed that response-order effects occur similarly in the right-to-left and the left-to-right reading direction.
ISSN:1864-6956
2190-4936