Premium
A Congruence Construct of User Involvement
Author(s) -
Doll William J.,
Torkzadeh Gholamreza
Publication year - 1991
Publication title -
decision sciences
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.238
H-Index - 108
eISSN - 1540-5915
pISSN - 0011-7315
DOI - 10.1111/j.1540-5915.1991.tb00358.x
Subject(s) - congruence (geometry) , construct (python library) , psychology , contingency , construct validity , social psychology , computer science , cognitive psychology , knowledge management , psychometrics , developmental psychology , programming language , linguistics , philosophy
The MIS literature has devoted considerable attention to the relationship between user involvement and MIS success; unfortunately, this research has produced conflicting results. Recent research on a discrepancy model of user involvement has provided a framework for reconciling studies showing positive, negative, or no impact of involvement on user satisfaction. In this discrepancy model, individual differences between perceived and desired levels of involvement define three conditional states or frames of reference that govern the relationship between involvement and end‐user satisfaction. The discrepancy model indicates that studies of user involvement need to control for individual differences by dividing respondents into three groups corresponding to these conditional states. Building upon the discrepancy concept, this paper presents a congruence construct of user involvement (i.e., a measure of involvement “relative” to an individual's desire to get involved) as an alternative way of modeling this contingency relationship. The reliability and validity of the involvement congruence construct are assessed. Perceived and congruence constructs of involvement are compared as predictors of end‐user computing satisfaction. The results suggest that involvement congruence is a better predictor than perceived involvement and may offer theoretical as well as empirical advantages over the use of component measures. Furthermore, the congruence method of modeling the discrepancy effect has research design advantages (i.e., it does not require dividing respondents by frame of reference).