PT Unknown
AU Christophersen, T
TI Usability im Online-Shopping: Entwicklung eines Fragebogeninstrumentes (ufosV2) unter Berücksichtigung formativer und reflektiver Messmodelle
SE Betriebswirtschaftliche Aspekte lose gekoppelter Systeme und Electronic Business
PY 2007
VL n/a
PU Christian-Albrechts-Universität zu Kiel
WP https://macau.uni-kiel.de/receive/diss_mods_00002117
LA de
DE Usability; E-Commerce; On-line Store; Questionnaire; Formative and Reflective Measurement Models; Evaluation; Success Factors; P; lose gekoppelter Systeme; virtuelle Unternehmen; elektronische Geschäftsprozesse; Betriebswirtschaftliche Aspekte; Loosely coupled systems; Virtual companies; Electronic business processes; Business management aspects
AB In usability research, difference between formative and reflective measurement models for the assessment of latent variables has been ignored largely. As a consequence, many usability scales are misspecified. This might result in reduced scale validity because of the elimination of important usability facets within the procedure of scale development. The aim of the current study was to develop a questionnaire for the evaluation of On-line store usability (UFOS-V2) that includes both a formative and a reflective scale. 378 subjects participated in a laboratory experimental study. Each participant visited two out of 35 On-line stores. The usability and intention to buy was assessed for both stores. In addition, actual purchase behaviour was observed by combining the subjects' reward with the decision to buy. In a two-construct PLS structural equation model the formative usability scale was used as a predictor for the reflective usability measure. Results indicate that the formative usability scale UFOS-V2f forms a valid set of items for the user-based assessment of online store usability. The reflective usability scale shows high internal consistency. In additional PLS-models, both UFOS scales showed positive effects on the intention to buy, suggesting high concurrent criterion validity. Relationships of UFOS scales with both trust and aesthetics were supported. Furthermore, a sequential multinomial regression analysis was performed on the decision to buy as outcome. Again, both UFOS-V2 measures provided good prediction of the criterion, confirming high predictive validity. In summary, the formative and the reflective UFOS measures fulfil all required scale criteria. Implications for human-computer interaction research are discussed.
PI Kiel
ER