MEASURING ANTICIPATED SATISFACTION: A FULL-SAMPLE APPROACH TO TECHNOLOGY SATISFACTION SCALES IN PRE-ADOPTION CONTEXTS

Abstract

Satisfaction is often only measured among current users of a technology since standard survey questions (e.g., 'I am satisfied with...') assume prior experience. This convention necessarily excludes any participant who has not yet adopted the technology — usually the majority of a sampled population in early technology-diffusion scenarios — and prevents direct, one-sample comparison of projected satisfaction across firms with different adoption statuses. This study presents and substantiates a comprehensive alternative: utilizing a singular four-item projected-satisfaction scale across an entire cross-sectional sample irrespective of adoption status, by administering identical, forward-looking satisfaction items to every respondent.. The methodology relies on Oliver's (1980) expectation-confirmation theory (ECT), which considers pre-consumption expectations as a separate, quantifiable cognitive state against which satisfaction assessments are made, and on the later adaptation of the expectation-confirmation model to technological environments. Utilizing survey data from 450 micro, small, and medium enterprises (195 users and 255 non-users of a governmental e-procurement platform), the SAT scores obtained from adopters and non-adopters demonstrate evidence of cross-group measurement comparability: reliability is notably high and nearly uniform across subgroups (α = .942 for users, α = .936 for non-users, α = .946 combined), a single-factor structure accounts for a similar proportion of variance in both cohorts (85.3% vs. 84.0%), and the scale's correlations with four theoretically relevant constructs (attitude, perceived relative advantage, performance expectancy, and awareness) show no significant differences between users and non-users (all |z| < 1.3, p > .20). Simultaneously, the average projected satisfaction level is higher among adopters (M = 3.53) than among non-adopters (M = 2.65, p < .001), demonstrating that the scale continues to effectively differentiate between groups rather than devolving into a mere artifact of shared method variance. Collectively, these findings endorse a comprehensive, anticipated-framing methodology for assessing satisfaction, serving as a versatile research instrument for scholars examining groups with varied adoption statuses, restoring statistical efficacy and facilitating comparisons of projected satisfaction across firms with different adoption statuses within a unified cross-sectional framework.