Quality & Quantity (2007) 41:601–626 © Springer 2007
Ordinal Methodology in the Analysis of Likert
, CHRISTOPHER McCOLLIN
Institute for Applied Mathematics and Statistics, University of W
urzburg, Sanderring 2,
urzburg, Germany. E-mail: firstname.lastname@example.org;
Trent University, University Burton Street, Nottingham, NG1 4BU, United Kingdom.
Instituto Superior T
ecnico, Maths Dept., Av. Rovisco
Pais, 1049-001 Lisbon, Portugal
Abstract. Likert scales are widely used in survey studies for attitude measuring. In particu-
lar, the questionnaires propagated by the SERVQUAL approach are based on Likert scales.
Though the problem of attitude suggests an ordinal interpretation of Likert scales, attitude
survey data are often evaluated with techniques designed for cardinal measurements. The
present paper discusses the interpretation of scales for attitude measuring and gives a survey
of data analysis techniques under the proper ordinal understanding.
Key words: attitude measuring, likert scales, ordinal scales, cardinal scales, SERVQUAL, sta-
Likert scales are widely used for measuring attitudes, e.g., opinions, psy-
chic and mental dispositions, preferences. Questionnaires and surveys based
on Likert scales are used in various areas, e.g., in psychometrics for the
analysis of subjective well-being, see Diener et al. (1985) or Watson et al.
(1988), in social studies and panels, or for purposes of business administra-
tion. The use of Likert scales has increased especially in the service sector
with consumer surveys now being commonplace within the hotel, leisure
and public utility sectors. In particular, the SERVQUAL approach intro-
duced by Parasuraman et al. (1985, 1988) has received enormous interest.
The ways of collecting survey data vary widely from the use of telephone
questionnaires to on-line designed web pages for automatic input.
The statistical analysis of survey data can range from simple dot plots
to logistic regression and cluster analysis to determine any hidden struc-
ture. However, many studies conﬁne themselves to a descriptive analysis.
Author for correspondence: E-mail: email@example.com