Exploring the Choice for Default Systems

Introduction

Few technology acceptance studies have been undertaken where respondents have had to evaluate two substitutive systems. The original Technology Acceptance Model (TAM) paper (Davis, 1989; see literature review) compared acceptance of the Chart-Master system and the Pendraw system. Other seminal papers in this domain assess the acceptance of a single system (Venkatesh et al., 2003; Venkatesh et al., 2012). A wider journal literature review shows researchers systematically focus on acceptance of a single system in isolation. With the exception of one paper (Strader et al., 2007), all TAM papers published in the European Journal of Information Systems since 2007 only studied the acceptance of one system (Van Slyke et al., 2007; Klein, 2007; Neufeld et al., 2007; Bhattacherjee & Hikmet, 2007; Po-An Hsieh & Wang, 2007; Khalifa & Liu, 2007; Dickinger et al., 2008; Sipior et al., 2011; Deng et al., 2010; Stafford & Hamit Turan, 2011; Yang et al., 2011; Schwarz et al., 2012). This study’s primary goal is to assess how acceptance of two substitutive systems are interrelated.

Technology acceptance model studies typically report a relatively high coefficient of determination (R²). Technology acceptance papers published since 2006 in Information & Management – which is the journal that publishes most technology acceptance papers (Bradley, 2012) – give R²s of 0.75 (Hasan, 2006), 0.46 (Cyr et al., 2006), 0.74 (Castaneda et al., 2007), 0.78 (Hsu et al., 2008), 0.60 (Chi et al., 2008), 0.58 (Jin, 2013) and 0.78 (Ho et al., 2015). The lowest R² reported since 2006 is thus 0.46 (Cyr et al., 2006). The R²s mentioned above for Information & Management are completely in line with R²s found in other journals. This suggests independent variables in technology acceptance literature explain a large part of the variance in the intention to use every information technology. However, it seems reasonable to believe that little reason exists for researchers who find a low R² to try to publish papers with little explanatory power, and journals are unlikely to accept papers with a low R² if no theory is present to explain the low R². Hence, readers may simply not get the opportunity to see reports of studies with a low R².

This article argues a classic application of technology acceptance theory is unlikely to explain much of the variance related to intent to use default systems. The researchers anticipate TAM to provide low R² in such cases. This study assesses the applicability of the large body of existing technology acceptance theory to explain the use of default systems and alternative systems.

Technology acceptance literature has not yet investigated defaults’ role in consumers’ IT choices. Partly this is because researchers have wanted to extend TAM and UTAUT (Unified Theory of Acceptance and Use of Technology, Venkatesh et al., 2003) models with context-specific constructs that provide insight into the acceptance of a single specific system. Researchers did not aim to define boundaries round the area where TAM and UTAUT models are highly explanatory. Researchers have tended to add constructs in order to improve those models’ validity in specific circumstances (Murphy et al., 2011).