Reliability of measurement scales: Tarkkonen's general method supersedes Cronbach's alpha

Show simple item record

dc.contributor University of Helsinki, Faculty of Social Sciences, Department of Mathematics and Statistics en
dc.contributor Helsingin yliopisto, Valtiotieteellinen tiedekunta, Matematiikan ja tilastotieteen laitos fi
dc.contributor Helsingfors universitet, Statsvetenskapliga fakulteten, Matematiska och statistiska institutionen sv
dc.contributor.author Vehkalahti, Kimmo
dc.date.accessioned 2009-09-08T09:29:30Z
dc.date.available 2009-09-08T09:29:30Z
dc.date.issued 2000-11-25
dc.identifier.uri http://hdl.handle.net/10138/10570
dc.description Endast sammandrag. Inbundna avhandlingar kan sökas i Helka-databasen (http://www.helsinki.fi/helka). Elektroniska kopior av avhandlingar finns antingen öppet på nätet eller endast tillgängliga i bibliotekets avhandlingsterminaler. sv
dc.description Only abstract. Paper copies of master’s theses are listed in the Helka database (http://www.helsinki.fi/helka). Electronic copies of master’s theses are either available as open access or only on thesis terminals in the Helsinki University Library. en
dc.description Vain tiivistelmä. Sidottujen gradujen saatavuuden voit tarkistaa Helka-tietokannasta (http://www.helsinki.fi/helka). Digitaaliset gradut voivat olla luettavissa avoimesti verkossa tai rajoitetusti kirjaston opinnäytekioskeilla. fi
dc.description.abstract Methods for assessing the reliability of measurement scales were investigated. Reliability, which is defined as the ratio of the true variance to the total variance, is an important property of measurement. In order to estimate the reliability, the concepts of measurement model and measurement scale are required. The model specifies the structure of the measurement, and the scale, which is a combination of the measured items, represents a realization of the theoretical notions. The focus was on two measures of reliability: Cronbach's alpha, which is widely applied, and Tarkkonen's more general measure. Both measures are founded on the same definition of reliability, but they imply different assumptions about the model and the scale. Cronbach's alpha is based on the classical true score model of psychometrics, while Tarkkonen's measure belongs to a general framework of modelling the measurement. The measures were examined theoretically and by extensive Monte Carlo simulation experiments implemented in Survo environment. Cronbach's alpha is shown to be a restricted special case of Tarkkonen's measure. According to the simulation experiments, the statistical properties of Tarkkonen's measure proved to be acceptable. Conversely, the study revealed additional evidence concerning the unsuitability of Cronbach's alpha as a measure of reliability. The results suggest that Tarkkonen's measure of reliability should supersede Cronbach's alpha in all applications. en
dc.language.iso en
dc.title Reliability of measurement scales: Tarkkonen's general method supersedes Cronbach's alpha en
dc.identifier.laitoskoodi 710
dc.description.note ISBN 952-91-2818-5. Statistical Research Reports, 17, The Finnish Statistical Society (ISSN 0356-3499). PDF version (ISBN 952-91-2819-3) available in http://ethesis.helsinki.fi/ en
dc.type.ontasot Doctoral thesis en
dc.type.ontasot Väitöskirja fi
dc.type.ontasot Doktorsavhandling sv
dc.type.dcmitype Text
dc.format.content abstractOnly

Files in this item

Total number of downloads: Loading...

Files Size Format View
abstract.pdf 48.43Kb PDF View/Open

This item appears in the following Collection(s)

Show simple item record