The moderator role of Gender in the Unified Theory of Acceptance and Use of Technology (UTAUT): A study on users of Electronic Document Management Systems

Abstract

Venkatesh et al. [1] tried to integrate predictability capabilities from the different existing models of technology acceptance. This produced the Unified Theory of Acceptance and Use of Technology (UTAUT). This comprehensive model resulted in the identification of common aspects. It proposed several constructs with a greater explanatory power and analyzed moderating drivers, such as age, Gender, experience and voluntariness of use. By doing so, UTAUT identifies three major drivers of behavioral intention: performance expectancy, effort expectancy and social influence. On the other hand, facilitating conditions and behavioral intention were identified as determinant factors of actual use [1]. In addition to previous considerations about UTAUT, empirical research has scarcely analyzed the moderating role of Gender [2]. This is why this paper particularly aims to fill this gap. Hofstede [3] describes strength, competitiveness and guidance for material success as social roles linked to male values, whilst modesty, tenderness, sensitivity and concern for the quality of life are values associated with women. With respect to UTAUT, existing studies have shown that performance expectancy positively influences behavioral intention more strongly for men (cf. [4], [5], [6] and [7]). Moreover, it has been observed that effort expectancy positively influences behavioral intention more strongly for women (cf. [4], [5] and [6]), while social influence positively affects behavioral intention more strongly for women (cf. [5], [7] and [8]). In our research, with the aim of testing the moderating effects of Gender, a sample of 2,175 users of Electronic Document Management Systems (EDMS) in Portuguese municipalities was used. Taking into account that Gender is a categorical variable, we have adopted a multi-group or multi-sample analysis [9] -dividing the sample into two groups (male = 748; female = 1,427) and estimating each group of observations separately. Before comparing the groups, an analysis of the measurement invariance was carried out to make sure that the construct measures were invariant between both groups [10]. Once the metric invariance had been assessed, we carried out a set of multi-group analyses –interpreting statistically-significant differences in path coefficients as moderating effects. On the one hand, the parametric approach considering both equal variances and different variances has been used [11, 12]. On the other hand, we have applied non-parametric approaches exemplified by the permutation test [13], and Henseler’s PLS multi-group analysis [10, 12, 14]. This study notes slight differences in the results of the aforementioned methods. As a result, the moderating effect of Gender on the relation between performance expectancy and behavioral intention showed that this relationship is stronger among men than women. Finally, a discussion on the implications of Gender as a moderator for the UTAUT model is included.