European Journal of Information Systems (2014), 1–17 © 2014 Operational Research Society Ltd. All rights reserved 0960-085X/14 www.palgrave-journals.com/ejis/

EMPIRICAL RESEARCH

Five-factor model personality traits as predictors of perceived and actual usage of technology Tim Barnett1, Allison W. Pearson1, Rodney Pearson1 and Franz W. Kellermanns2 1 Department of Management and Information Systems, Mississippi State University, Starkville, U.S.A; 2UNC-Charlotte, Charlotte, NC, U.S.A

Correspondence: Tim Barnett, Department of Management and Information Systems, Mississippi State University, P.O. Box 9581, Mississippi State, MS 39762, U.S.A. Tel: 662-325-2419; Fax: 662-325-8651

Abstract Understanding the adoption and use of technology is extremely important in the field of information systems. Not surprisingly, there are several conceptual models that attempt to explain how and why individuals use technology. Until recently, however, the role of personality in general, and the five-factor model (FFM) of personality in particular, had remained largely unexplored. Our study takes an interactional psychology perspective, linking components of the FFM to the use of technology within the conceptual framework of the Unified Theory of Acceptance and Use of Technology (UTAUT). After empirically confirming previous research findings linking performance expectancy, effort expectancy, and social influence to technology use, we test direct relationships between FFM personality traits and technology use in the context of a webbased classroom technological system, utilizing measures of perceived and actual use of technology. Consistent with expectations, conscientiousness and neuroticism are associated with perceived and actual use of technology, with conscientiousness demonstrating a positive association with both perceived and actual use and neuroticism, a negative association. Extraversion was also significantly associated with actual use, although not in the positive direction expected. Further, the significant relationships between the personality traits and the actual use of technology were direct and not mediated by expressed intentions to use the system. European Journal of Information Systems advance online publication, 3 June 2014; doi:10.1057/ejis.2014.10 Keywords: personality; five-factor model; UTAUT; TAM; system use; individual differences; technology acceptance; technology adoption

Introduction

Received: 15 December 2011 Revised: 15 September 2012 2nd Revision: 30 April 2013 3rd Revision: 07 October 2013 4th Revision: 18 February 2014 5th Revision: 18 March 2014 Accepted: 25 March 2014

Understanding how and why users accept and use technology is of critical interest to Information Systems (IS) research. Not surprisingly, theories of technology adoption, such as the Technology Acceptance Model (TAM) (Davis, 1989) and TAM2 (Venkatesh & Davis, 2000), have proliferated. After synthesizing eight such models, Venkatesh et al (2003) developed the Unified Theory of Acceptance and Use of Technology (UTAUT), concluding that it ‘provides a useful tool for managers needing to assess the likelihood of success for new technology introductions and helps them understand the drivers of acceptance’ (p. 425). The UTAUT ‘represents a shift from a fragmented view of technology acceptance to a unified view that integrated the major theories and technology acceptance models into a single theory’ (Abu-Shanab et al, 2010, p. 495) and ‘has been applied to the study of a variety of technologies in both organizational and non-organizational settings’ (Venkatesh et al, 2012). Thus, we position our study of technology acceptance and use within this widely accepted framework.

2

Five-factor model personality traits

Like other models of technology acceptance, the UTAUT focuses primarily on situational constructs, such as system usefulness and ease of use. Although individual differences such as gender and age are considered as potential moderators of the aforementioned variables’ impact on technology use, the UTAUT does not consider the direct impact of individual differences on technology use. Researchers have suggested that individual differences of various kinds should be incorporated into studies of technology use (e.g., Zmud, 1979; Nelson, 1990; Harrison & Rainer 1992; Neufeld et al, 2007; Theotokis et al, 2008; Pramatari & Theotokis 2009), but as Thatcher and Perrewe (2002, p. 382) state ‘Although mounting evidence suggests individual differences influence IT use, more integrative research is needed to better understand the nomological net among individual differences that relate to IT acceptance and use’. More specifically, personality differences, which were previously ignored (Agarwal & Prasad, 1999), have received increased scholarly attention within the broad domain of technology use (Stewart & Segars, 2002; Vishwanath, 2005; Ehrenberg et al, 2008; Hunsinger et al, 2008; Junglas et al, 2008; Korzaan & Boswell, 2008; Lin, 2008; Theotokis et al, 2008; Jacques et al, 2009; Pramatari & Theotokis, 2009; Lin & Ong, 2010; Wilson et al, 2010; Witt et al, 2010; Al-Natour et al, 2011; Bansal, 2011; Zhou & Lu, 2011; Buckner et al, 2012; Devolder et al, 2012; Koenigstorfer and GroeppelKlein, 2012; Lane & Manner, 2012; Terzis et al, 2012; Venkatesh & Windeler, 2012; Wang et al, 2012a, b; Jackson et al, 2013; Kober & Neuper, 2013; Liu et al, 2013). For example, recent studies have linked personality traits to attitudes toward Radio Frequency Identification (Pramatari & Theotokis, 2009), participation in online social networking, video game playing, and virtual reality (Wilson et al, 2010; Witt et al, 2010; Venkatesh & Windeler, 2012; Kober & Neuper, 2013), trust in, perceived usefulness and use of mobile technologies (Zhou & Lu, 2011; Koenigstorfer and Groeppel-Klein, 2012; Lane & Manner, 2012), various types of online shopping (Al-Natour et al, 2011; Bansal, 2011; Liu et al, 2013), blogging and instant messaging (Wang et al, 2012a, b), and even the excessive use of technology (Buckner et al, 2012). However, most studies of personality and technology have not incorporated traits directly into comprehensive technology acceptance models (TAM). However, beginning with McElroy et al, 2007, several studies, summarized in Table 1, illustrate the utility of including personality within the framework of the TAM (Devaraj et al, 2008; Aldas-Manzano et al, 2009; Svendsen et al, 2013), the UTAUT (Abu-Shanab et al, 2010), or both (McElroy et al, 2007). In a TAM-based study, Aldas-Manzano et al (2009) looked at the relationship between innovativeness and intentions to use mobile technology, finding that it was positively associated with intentions. Devaraj et al (2008) also drew upon the TAM, incorporating the five-factor model (FFM) traits of conscientiousness, extraversion, agreeableness, neuroticism, and openness to experience as predictors of TAM constructs and as moderators of TAM constructs’ relationships with technology use. Within

European Journal of Information Systems

Tim Barnett et al

their student sample, conscientiousness, extraversion, and agreeableness moderated relationships between perceived usefulness and/or subjective norms and the students’ intentions to use a course management system. In another TAM-based study, Svendsen et al (2013) focused on the impact of the FFM personality traits on behavioral intentions to use a hypothetical software tool, finding that perceived usefulness and ease of use fully mediated the effects of conscientiousness and extraversion on intentions, while emotional stability directly affected intentions. McElroy et al (2007) incorporated constructs from both the TAM and the UTAUT into a model with the FFM traits and other personality constructs, finding that openness to experience was associated with self-reported internet use and neuroticism was associated with willingness to buy and sell online. Abu-Shanab et al (2010) also drew upon the UTAUT and incorporated four personality traits into a model of internet banking usage, finding that self-efficacy and locus of control were positively associated with intentions to use the technology. These studies are consistent with an influential theoretical framework in applied psychology, interactional psychology, (Bowers, 1973; Terborg, 1981; Schneider, 1983), which suggests that ‘simultaneous consideration of both the person and the situation’ (Terborg, 1981, p. 569) is essential to understanding behavior. Dating back to Lewin’s (1951) classic conceptualization of behavior as a function of the environment and the person (i.e., B = f {E,P}), this perspective rejects purely environmental or trait-based explanations of behavior as inadequate (Bowers, 1973), concluding that ‘both situational factors and personality factors are important for understanding affect, cognition, attitudes, and behavior’ (George, 1992, p. 192) and that ‘it is generally accepted that both situation and personal variables in complex combinations affect attitudes and behavior’ (Nelson, 1990, p. 80). Although within this framework there are multiple ways in which person–situation interactions can occur (Terborg, 1981; Schneider, 1983), it does not advocate a particular interactional form, but instead makes the broader point that behavior is a joint function of the person and the situation (Bowers, 1973). Although IS researchers have taken an interactional perspective to demonstrate the relevance of personality within the framework of the TAM or UTAUT, the extant research is neither comprehensive nor definitive. Of the illustrative studies shown in Table 1, only one (Devaraj et al, 2008) utilized an objective measure of system use, with the others focusing only on behavioral intentions or perceived use (McElroy et al, 2007; Aldas-Manzano et al, 2009; Abu-Shanab et al, 2010; Svendsen et al, 2013). Further, the FFM, which despite criticism (Block, 1995; Epstein, 2010; Boag, 2011) is the dominant personality taxonomy in applied psychology ( John et al, 2008; John & Naumann, 2010; Chang et al, 2012), was included in only three studies (McElroy et al, 2007; Devaraj et al, 2008; Svendsen et al, 2013), none of which considered all of the contextual constructs in the UTAUT. Finally, no

Study

Conceptual model addressed

Abu-Shanab UTAUT et al, 2010

Illustrative recent studies on personality traits incorporated directly into TAM or UTAUT-based models Sample

Summary of findings related to personality traits

Behavioral intention

Self-efficacy; anxiety; innovativeness; locus of control Big Five (conscientiousness, extraversion, agreeableness, openness, and neuroticism)

523 banking customers

Self-efficacy and locus of control predicted behavioral intentions to engage in internet banking

180 students

Innovativeness

470 mobile phone users

Conscientiousness moderated (1) perceived usefulness→behavioral intentions and (2) subjective norms→behavioral intentions. Extraversion moderated subjective norm→behavioral intentions. Agreeableness moderated subjective norm→behavioral intentions. Neuroticism negatively associated with perceived usefulness. Openness and agreeableness positively associated with perceived usefulness Innovativeness positively associated with M-shopping intentions

Devaraj et al, 2008

TAM

Actual logged usage of a collaborative technology

AldasManzano et al, 2009 McElroy et al, 2007

TAM

Mobile shopping intentions and patronage

Cited both TAM and UTAUT – did not incorporate all variables from either TAM

Self-reported internet use, Big Five, self-efficacy, locus of 153 students willingness to buy online, control, Myers–Briggs and willingness to sell online cognitive style

Svendsen et al (2013)

Behavioral intention to use a Big Five hypothetical software tool

Big Five explained additional variance in one or more of the selfreported dependent variables. Specifically, openness was positively associated with self-reported internet use, and neuroticism was positively associated with willingness to buy and sell online The impact of conscientiousness and extraversion on behavioral intentions was fully mediated by perceived usefulness and ease of use. Emotional stability had direct effect on expressed behavioral intention. Openness was associated with perceived ease of use, but not behavioral intention

3

European Journal of Information Systems

1004 Norwegians drawn from a statistical panel

Tim Barnett et al

Primary dependent variable(s) Specific personality traits examined

Five-factor model personality traits

Table 1

4

Five-factor model personality traits

Situational Constructs Main Effect Antecedents from UTAUT Facilitating Conditions Main effect only

Tim Barnett et al

H4

Performance Expectancy H1 - H3

Usage Outcomes

Effort Expectancy Social Influence

Individual Constructs Big Five Personality Constructs

Perceived System Use Behavioral Intention

H9 Actual Usage

H5a – H8a

H5 – H8

Conscientiousness Openness to Experience Neuroticism Extraversion Agreeableness No relationship

Figure 1 Adapted UTAUT (Venkatesh et al, 2003), Interactional Psychology Perspective. H1–H3 suggest that behavioral intention mediates the effects of performance expectancy, effort expectancy, and social influence on usage. H4 suggests a direct effect of facilitating conditions on usage as suggested by the UTAUT. H5–H8 suggest main effects of four of the Big Five personality traits on usage. H5a–H8a suggest that behavioral intention will partially mediate the relationship between the personality traits and usage.

studies examined the direct impact of personality on actual system use, although one found direct effects on self-reported use (McElroy et al, 2007). Thus, there is a need for more research on the FFM and technology use within the framework of the TAM or UTAUT. Illustrating this, Svendsen et al (2013) state ‘surprisingly few studies have investigated the relation between general personality traits and TAM constructs’ (p. 324) and Devaraj et al (2008, p. 104) state ‘Personality has been largely ignored in the MIS literature over the past two decades’. McElroy et al (2007) concur, remarking that Future models of IS adoption and use may be improved by incorporating personality along with existing attitudinal and situational determinants … The first step is to reintroduce disposition factors into models of technology use and adoption. TAM2 (Venkatesh & Davis, 2000) and UTAUT (Venkatesh et al, 2003) are two viable candidates. (p. 817) Therefore, in this study, we draw upon the interactional perspective and extend the fledgling IS research on the FFM and technology use. Our study is the first, to our knowledge, that does all the following: (1) utilizes the UTAUT as the underlying technology acceptance model, (2) includes actual, in addition to perceived use of technology, and (3) examines direct effects of FFM traits on actual, as well as perceived, system use. As shown in Figure 1, we test contextual hypotheses emanating from the UTAUT, but also test hypotheses linking FFM traits to behavioral intentions, self-reported use, and the actual use of technology. We first provide a brief review of the development of the UTAUT, and present

European Journal of Information Systems

several hypotheses drawn directly from it. We then focus on how trait-based differences are relevant to the use of technology, deriving hypotheses that link FFM traits to technology use. Finally, we report the results of an empirical test of these hypotheses and discuss the implications of our findings.

Theoretical development of the UTAUT – emphasis on situational aspects of the technological system The UTAUT builds upon the theories of reasoned action and planned behavior (Ajzen & Fishbein, 1980; Ajzen, 1991; Armitage & Christian, 2003) as well as prior models of technology acceptance (Venkatesh et al, 2003). It focuses on four situational or contextual constructs: (1) performance expectancy, (2) effort expectancy, (3) social influence, and (4) facilitating conditions. Performance expectancy is the user’s belief about how the technological system will help that user perform. Effort expectancy is defined as the ease of use of the system. The social influence construct is the degree to which influential others believe that the user should use the system. Each of these constructs is theorized in the UTAUT to affect system use through its impact on behavioral intentions (Venkatesh et al, 2003). Facilitating conditions is defined as the perceived level of organizational and technical support for the system. This construct is conceptualized in the UTAUT as a direct predictor of technology use (Venkatesh et al, 2003). In addition, the UTAUT suggests that individual factors such as age and gender may moderate the relationships between the situational constructs and technology acceptance and use.

Five-factor model personality traits

In addition to the original validation of the UTAUT (Venkatesh et al, 2003), several other researchers have tested its main hypotheses and/or extended the model by (1) testing it in new contexts, (2) introducing new constructs predictive of behavioral intentions and technology usage, or (3) examining predictors of the UTAUT constructs themselves (Venkatesh et al, 2012). In general, these empirical tests support the UTAUT (e.g., Brown et al, 2010; Chan et al, 2010; Fillion et al, 2012). However, some of the UTAUT constructs have received more consistent support than others, and the integration of additional predictors into the original UTAUT appears to increase variance explained in behavioral intentions and actual use. Here, we first provide empirical tests of the main effects hypotheses derived from the UTAUT, in terms of both perceived and actual system use. These hypotheses are stated formally below: H1: Behavioral Intention will fully mediate the relationship between Performance Expectancy and IT usage, both actual and perceived. H2: Behavioral Intention will fully mediate the relationship between Effort Expectancy and IT system usage, both actual and perceived. H3: Behavioral Intention will fully mediate the relationship between Social Influence and IT system usage, both actual and perceived. H4: Facilitating Conditions will be positively related to IT system usage, both actual and perceived.

The FFM of personality and technology acceptance and use Within the conceptual framework of interactional psychology, multiple types of person–situation interaction are possible (Terborg, 1981). Statistical, or ‘algebraic’ interaction, is commonly explored in A×B moderator variable models (Terborg, 1981; Schneider, 1983). The incorporation of FFM constructs as moderators of the TAM model’s predictors of behavioral intentions and technology use (Devaraj et al, 2008) is consistent with algebraic interaction. Another form of person–situation interaction that is particularly relevant to the present study is additive interaction (Terborg, 1981; Schneider, 1983; George, 1992), which occurs when person and situation factors independently explain variance in a dependent variable or variables (Terborg, 1981). McElroy et al’s (2007) consideration of the FFM constructs as possible predictors of behavioral intentions and willingness to buy and sell online, AldasManzano et al’s (2009) TAM-based study of the effect of innovativeness on mobile shopping intentions, Abu-Shanab et al’s (2010) incorporation of self-efficacy, anxiety, innovativeness, and locus of control into the UTAUT as predictors of behavioral intentions, and Svendsen et al’s (2013) study of the FFM constructs’ effect on perceived usefulness, ease of use, and behavioral intention, are all consistent with additive interaction (Terborg, 1981). Additive interaction is

5

Tim Barnett et al

also consistent with a type of UTAUT extension identified by Venkatesh et al (2012), in which constructs hypothesized as predictive of IT system use are added to the UTAUT model to expand the model’s scope. Our study considers this form of interaction, in that we hypothesize that the FFM constructs have independent and direct effects on behavioral intentions, perceived usage, and the actual use of a technological system. We now turn to a brief review of the FFM and the development of hypotheses linking conscientiousness, openness to experience, neuroticism, and extraversion to technology use.

The FFM categorization of personality Early research on personality, lacking a parsimonious taxonomy (John et al, 2008), was dominated by exhaustive lists of hundreds of traits (e.g., Allport & Odbert, 1936). In the second half of the twentieth century, a number of researchers (e.g., Norman, 1963; Digman, 1990; Goldberg, 1990) contributed to the development of the FFM (conscientiousness, openness to experience, neuroticism, extraversion, and agreeableness), an ‘integrative personality taxonomy that offers a common nomenclature for scientists working the field’ ( John & Naumann, 2010, p. 48). Although some have criticized the FFM as atheoretical, incomplete, and inflexible (Block, 1995; Epstein, 2010; Boag, 2011), the field of personality research appears to have arrived at an ‘initial consensus on a general taxonomy of personality’ (John et al, 2008, p. 116) and has concluded that ‘on the whole, the wide base of research that has supported development of these models suggests that people differ on five general dimensions’ (Chang et al, 2012, p. 408). This conclusion is supported by the fact that, although it was relatively unknown as late as 1990, there have now been more than 3000 studies incorporating the FFM, compared with less than half this number for all other personality taxonomies ( John & Naumann, 2010). Thus, although the FFM may not represent the ‘last word’ on personality structure (John & Naumann, 2010, p. 48), it has ‘much to offer’ to those interested in the effect of personality traits on behavior (Epstein, 2010, p. 35). The FFM is not only parsimonious, but also has as strong, or even stronger, predictive validity than narrow personality traits (Driskell et al, 1994; Dudley et al, 1996; Ones & Viswesvaran, 1996). A significant body of research supports the stability of the FFM across a variety of samples (Costa & McCrae, 1988, 1992; McCrae & Costa, 1989). These and other findings led Digman (1990, p. 436) to conclude that the FFM creates a ‘set of very broad dimensions that characterize individual differences. These dimensions can be measured with high reliability and impressive validity’. The FFM has been widely adopted in personnel selection and job performance research, and a series of meta-analytic studies demonstrate that the FFM is associated with a wide variety of attitudes and behaviors (Barrick & Mount, 1991; Tett et al, 1991; Mount & Barrick, 1995; Salgado, 1997; Hurtz & Donovan, 2000; Judge & Ilies, 2002). Extending this perspective to the IS field, McElroy et al (2007) state

European Journal of Information Systems

6

Five-factor model personality traits

‘Taken together, the Big Five capture the essence of one’s personality’ (p. 811). From this rich research foundation, we set out to explore which, if any, of the Big Five dimensions directly predict the UTAUT outcome variables (1) intentions to use, and (2) use of technology, both perceived and actual. Following McElroy et al (2007), we are interested in exploring personality’s ‘direct effect on IS use, not in casting personality as a precursor to cognitive behaviors’ (p. 811), and within the context of the existing UTAUT, we consider the personality traits’ potential to explain variance in IT system adoption, over and above that explained by the UTAUT’s situational constructs.

Tim Barnett et al

intentions to use a hypothetical software technology. On the basis of theory and the empirical findings of related studies, we expect that the effect of conscientiousness on IT system use will be partially mediated by users’ intentions to use the system. Therefore we propose the following hypotheses: H5:

Conscientiousness will be positively related to IT system use, both actual and perceived.

H5a: Behavioral Intention to use the IT system will partially mediate the relationship between Conscientiousness and IT system use.

Openness to experience Conscientiousness Conscientiousness includes thoroughness, dependability, responsibility, and achievement orientation (Digman, 1990). Studies have demonstrated a relationship between this trait and task proficiency in a variety of contexts, with individuals who are highly conscientious generally performing better (Barrick & Mount, 1991; Hurtz & Donovan, 2000). Conscientiousness is linked to motivation to learn (Major et al, 2006) because conscientious individuals set clear goals and engage in behaviors that help them succeed. It is also positively correlated to a learning goal orientation, a characteristic of individuals who value acquiring new skills and/or knowledge (Payne et al, 2007). It has been linked to academic achievement (Laidra et al, 2007) and is thus relevant to students’ classroom performance (Hunsinger et al, 2008), as well as workplace performance (Barrick & Mount, 1991). Conscientiousness is likely to be associated with classroom behaviors, such as the use of a course management system, that are consistent with the motivation to learn, a strong goal orientation, and a desire to acquire knowledge. With regard to the specific relationship of conscientiousness to technology use, McElroy et al (2007, p. 811) point out that the FFM is of interest ‘because of its established link to behaviors and cognitions’ and Devaraj et al (2008, p. 93) state that the FFM is ‘associated with a number of organizational processes, behaviors, and outcomes’ and that ‘we expect conscientious people to be more likely to carefully consider whether technology provides an opportunity to further on-the-job achievement and then act based on that assessment; conscientiousness will be related to the enactment of intentions’. Landers & Lounsbury (2006) found that highly conscientious students were more likely to use the internet for academic purposes than leisure, which is consistent with earlier research suggesting that conscientiousness is linked to a learning goal orientation (Payne et al, 2007). Hurtz and Donovan (2000) note that conscientiousness has motivational implications, and recommend that it be considered not only as a direct predictor of behaviors, but also as an indirect predictor through intentions. Consistent with this perspective, in their TAM-based study Svendsen et al (2013) found that conscientiousness was associated with behavioral

European Journal of Information Systems

Openness to experience is characterized by curiosity, originality, inquisitiveness, and artistic sensitivity (McCrae & Costa, 1989). Like conscientiousness, openness is linked to a strong motivation to learn (Major et al, 2006), and with a learning goal orientation (Payne et al, 2007), perhaps because individuals characterized by openness desire to learn for ‘learning’s sake’. It appears to be a valid predictor of training proficiency (Barrick & Mount, 1991), with individuals who are curious and broad-minded being ‘more likely to have positive attitudes toward learning experiences in general’ (Barrick & Mount, 1991, p.19). Individuals scoring highly on openness to experience may also be more eager to engage in new and learning-oriented experiences. Openness has also been linked to classroom achievement (Laidra et al, 2007). Thus, openness appears relevant to both the workplace and classroom contexts. As technological systems often require active learning, openness may be useful in identifying users who are more willing to learn and have high intentions to use technology in the classroom. In their study of students’ internet use, McElroy et al (2007) found openness to experience to predict overall use. In a closely related study, Agarwal & Karahanna (2000) examined the relationship of cognitive absorption to perceived usefulness and ease of use of an IT system. The authors defined cognitive absorption as including ‘responsiveness to engaging stimuli; responsiveness to inductive stimuli’ (p. 667), where an individual becomes deeply interested and engrossed in activities. Cognitive absorption has some theoretical overlap with openness, which is described in terms of an ‘inquiring intellect’ (Digman, 1990, p. 423). Using a student sample, Agarwal & Karahanna (2000) showed that cognitive absorption predicted perceived usefulness, ease of use, and intentions to use an IT system. In addition, Barrick & Mount (1991) note that openness may relate to both ability to learn and motivation to learn, which are closely related to behavioral intentions. We therefore offer the following hypotheses: H6:

Openness to Experience will be positively related to IT system usage, both actual and perceived.

H6a: Behavioral Intention to use the IT system will partially mediate the relationship between Openness to Experience and IT system use.

Five-factor model personality traits

Neuroticism Neuroticism is the degree to which a person is ‘anxious, depressed, angry, embarrassed, emotional, worried, and insecure’ (Barrick & Mount, 1991, p. 4). Neuroticism is not associated with a motivation to learn and is negatively linked to a learning goal orientation (Payne et al, 2007). Thus, neurotic individuals are not expected to seek out opportunities to learn new things (Major et al, 2006), because of their generally negative affect and expectations. In the classroom, neuroticism appears to be negatively associated with academic performance (Laidra et al, 2007). Findings regarding neuroticism and other types of task performance tend to confirm a negative relationship across a wide variety of job tasks (Barrick & Mount, 1991; Hurtz & Donovan, 2000), leading researchers to conclude ‘It appears that being calm, secure, well-adjusted and low anxiety has a small but consistent impact on job performance’ (Hurtz & Donovan, 2000, p. 876). The relationship between neuroticism and technology use is less clear (McElroy et al, 2007). Although Svendsen et al (2013) did not find a significant relationship between neuroticism and intentions to use a hypothetical software technology, McElroy et al (2007) actually found a positive association between the trait and willingness to engage in internet buying or selling. Marakas et al (2000) explored the theoretical role neuroticism plays in influencing a user’s personification of information technologies, concluding that when users are anxious about novel situations, they may be more susceptible to feelings of helplessness and struggle with their social interpretations of computer usage. In explaining how users adjust to information technology, Nelson (1990) recommends research exploring the relationship of the user’s negative affectivity. She proposes that users who are prone to negative assessments and conclusions will experience more tenuous adjustments to technology. Devaraj et al (2008, p. 97) allude to the workplace literature and conclude ‘neurotic personalities are likely to view technological advances in their work as threatening and stressful, and to have generally negative thought processes when considering it’. Thus, given that neuroticism does not promote a motivation to learn or a learning goal orientation, and that learning and adopting computer systems often require patience and trial and error, we hypothesize that individuals who are high in neuroticism will be less likely to form intentions to use an IT system, or to actually adopt and use the system. H7: Neuroticism will be negatively related to IT system use, both actual and perceived. H7a: Behavioral Intention to use the IT system will partially mediate the relationship between neuroticism and IT system use.

Extraversion Extraversion is characterized by sociability, assertiveness, and gregariousness (Barrick & Mount, 1991). It is associated

Tim Barnett et al

7

with a strong motivation to learn (Major et al, 2006) and with a learning goal orientation (Payne et al, 2007), probably because of the assertive characteristics that are associated with extraverts. Further, it is predictive of a wide variety of job tasks, and thus appears relevant to both job and classroom tasks related to learning (Barrick & Mount, 1991). Zmud (1979) reports that extraverts generally possess more positive attitudes toward IS than others. Empirical findings are mixed, however, with Devaraj et al (2008) finding that extraversion moderated the subjective norm– behavioral intention relationship in the TAM model, and Svendsen et al (2013) finding that extraversion impacted behavioral intentions through perceived usefulness and ease of use in a TAM model, whereas McElroy et al, (2007) found no significant effect for extraversion in a combined TAM/UTAUT model. Perhaps extraverts may readily adopt communication systems, such as e-mail, to complement their interaction avenues. The purpose of many systems, however, is to provide meaningful information rather than to facilitate social interaction. Thus, the advantage that extraverts may have in classroom system use exists in the training and learning required for systems. By questioning and interacting, extraverts may be more likely to gain access to multiple information sources (instructor, other students) for learning. Following this logic, Barrick & Mount (1991) found that extraversion predicted training proficiency, concluding that since it is associated with general activity levels, such as being assertive, active, and talkative, it may facilitate more efficient learning on the job or task, as the learner is actively engaged with others. H8:

Extraversion will be positively related to IT system use, both actual and perceived.

H8a: Behavioral Intention to use the IT system will partially mediate the relationship between Extraversion and IT system use.

Agreeableness Agreeableness is characterized by kindness, good-naturedness, trust, and tolerance (Barrick & Mount, 1991). Although agreeableness is associated with a learning goal orientation (Payne et al, 2007), it does not appear to contribute to a stronger motivation to learn (Major et al, 2006). Although agreeableness may be a desirable personality characteristic in individuals, it generally has the weakest link to task performance. Barrick & Mount (1991) conclude that ‘the results for agreeableness suggest that it is not an important predictor of job performance, even in those jobs containing a large social component.’ (p. 21). Further, previous work on the FFM in TAM or UTAUT frameworks, with the exception of Devaraj et al (2008), have not found that agreeableness significantly impacts technology acceptance or use (McElroy et al, 2007; Svendsen et al, 2013). Therefore, because agreeableness does not appear to foster a strong motivation to learn and because its relationship to a wide variety of tasks, including technology use, appears relatively weak, we do

European Journal of Information Systems

8

Five-factor model personality traits

not hypothesize a direct relationship to behavioral intentions to use an IT system or actual IT system use. However, for the sake of completeness in terms of the FFM, we do include agreeableness in our overall model (e.g., Major et al, 2006)

Perceived and actual usage Prior IS research has recognized that actual behavior and perceived behavior are not necessarily interchangeable (Straub et al, 1995; Barnett et al, 2007). Indeed, the task of self-reporting usage in complex systems may be difficult (Barnett et al, 2007). It is, however, important to assess whether the UTAUT has predictive validity for both actual and perceived usage behavior. Accordingly, our abovestated hypotheses include both actual and perceived usage behavior. In addition, we need to know if perceived usage behavior serves as a viable and sufficient proxy for actual usage behavior in the realm of the unified theoretical model. Accordingly, we investigate whether actual usage predicts perceived usage. H9: Actual IT system usage will be positively related to perceived IT system usage.

Research method We conducted this study during an academic semester in four undergraduate business classes at a large state university in the United States as representative of our population of interest – computer users. The instructors invited students to use a custom-designed web-based course management system, which for the purposes of this paper we term m-web. Students received information at the beginning of the semester on accessing the system, which includes a course website (consisting of lecture outlines, assignments, and other course material), an online gradebook (which includes each student’s grades on all exams and assignments), and a class e-mail archive (which include copies of all the electronic communication to the class from the teacher during the semester). Students could use each feature as often as desired; use of the system was totally voluntary, in that students had alternative means of accessing the information, although the comprehensiveness and convenience of the system was beneficial to the student.

Sample The sample frame was 382 undergraduate students in the four classes. Almost all students were in their early to midtwenties, with an average age of 21.9 years. Fifty-eight percent of the students were male. All aspects of student participation, data collection, and analysis were reviewed and approved by the Institutional Review Board at the authors’ institution and all students provided informed consent before completing surveys. We collected data from and about the students at three different points in time. During the second week of the semester, we administered a survey assessing the UTAUT constructs and the Big

European Journal of Information Systems

Tim Barnett et al

Five personality constructs. During the last week of the 15-week semester, students self-reported their perceived system use. A total of 347 students participated in both surveys (a response rate of 91%). Four responses were excluded because the students were enrolled in more than one of the classes. We collected actual logged system usage data for each student at the end of the term from the m-web system.

Measures Following accepted best practices (Hinkin, 1995, 1998), we utilized existing measures to assess the UTAUT constructs, and used well-established measures from the applied psychology literature to assess the FFM constructs. Except where noted, all items were assessed with 7-point Likert scales, anchored by ‘strongly disagree’ and ‘strongly agree’. Following the general approach outlined by Anderson & Gerbing (1988), we assessed the measurement model before testing the hypothesized relationships. Confirmatory factor analyses (CFA) and reliability analyses on the final measures indicated that all measurement scales demonstrated acceptable construct validity and strong internal consistency. A list of items and the Cronbach’s α for each scale utilized in the study are presented in the Table A1 in Appendix. Effort expectancy (perceived ease of use) and performance expectancy (perceived usefulness) were each assessed by 3item scales developed by Davis (1989) and Davis et al (1989), and further refined in later studies (e.g., Venkatesh & Davis, 2000; Venkatesh et al, 2003). We slightly modified each item’s wording to fit the IT system context in the four classes. The 3-item measures yielded coefficient’s of 0.94 and 0.89, respectively. Social influence was assessed with an abbreviated 2-item scale that has been frequently used (e.g., Taylor & Todd, 1995a, b; Venkatesh et al, 2003), with slightly revised wording for the specific web-based system of interest. The α for this measure was 0.90. Facilitating conditions was measured with a 3-item scale adapted from Taylor and Todd (1995a) and Thompson et al, (1991), as suggested by Venkatesh et al (2003). This scale had an α of 0.77. To assess the FFM personality traits, we initially chose 50 items from the International Personality Item Pool (Goldberg, 1999). These scales have been used often in organizational behavior and human resource management research (e.g., Vasilopoulos et al, 2005; Cote & Miners, 2006), and they appear to have adequate convergent and discriminant validity (Lim & Ployhart, 2006). However, we conducted a CFA to obtain additional evidence related to the psychometric properties of the items. This process revealed the need to eliminate several items from each personality measure, which resulted in measures that exhibited acceptable αs and also demonstrated better fit with the data. The items used to assess conscientiousness, openness to experience, neuroticism, extraversion, and agreeableness are shown in the Table A1 in Appendix. The αs for the five personality measures ranged from 0.71 to 0.79.

Five-factor model personality traits

Table 2

Descriptive statistics and correlations

Variables

Mean Standard deviation

1. GPA 2. Effort expectancy 3. Performance expectancy 4. Social influence 5. Facilitating conditions 6. Extraversion 7. Neuroticism 8. Conscientiousness 9. Openness 10. Agreeableness 11. Behavioral intention 12. Perceived use 13. Actual use

2.97 5.85 5.15 4.49 5.92 3.46 3.52 3.60 3.72 3.84 5.57 3.03 1.22

0.59 0.98 1.19 1.30 0.85 0.73 0.66 0.63 0.73 0.64 1.09 1.17 0.75

9

Tim Barnett et al

1

2

3

4

5

6

7

8

9

10

11

12

0.11 0.03 0.38 0.05 0.21 0.33 0.08 0.54 0.41 0.32 0.01 −0.02 −0.02 −0.01 0.04 0.11 0.03 0.01 −0.10 0.14 0.29 0.14 0.25 0.13 0.12 0.21 0.19 0.24 0.08 0.12 0.05 −0.07 0.14 0.18 0.16 0.20 0.13 0.23 0.24 0.09 0.32 0.01 0.20 0.45 0.28 −0.06 0.53 0.59 0.36 0.51 −0.02 −0.07 0.15 0.06 0.29 −0.26 0.13 0.18 0.16 0.14 −0.03 −0.09 0.14 −0.10 0.10 0.32 0.09 0.20 0.18 0.07 0.10 −0.21 −0.18 0.17 −0.05 0.11 0.24 0.33

Correlations greater than 0.11 are significant at p<0.05.

Self-reported usage of the system was assessed with a 3item measure, adapted from Davis (1989), which asked for a self-report of the student’s weekly average number of accesses of the course website, the online grades page, and the e-mail archive. An 8-point scale ranging from ‘0 times per week’ to ‘7 or more times a week’ was utilized. Coefficient α for the measure was 0.79. In addition, we assessed the actual system usage for each student at the end of the semester. To access the system, each student entered a unique identification code, which allowed us to obtain data for each user’s frequency of accesses to the website, email archives, and grade records from m-web’s internal log. These items were used to form a construct, following a similar technique used by Devaraj et al (2008). Finally, behavioral intentions was measured with a 3-item scale (e.g., Davis et al, 1989; Taylor & Todd, 1995a; Venkatesh & Davis, 1996), which had an α of 0.91 in the present study. A student’s grade point average (GPA) relates to cognitive ability as well as intrinsic motivation, which are constructs suggested to have a potential impact on the study variables (e.g., Venkatesh et al, 2003). Thus, we included overall GPA as a control variable.

Results Table 2 presents descriptive statistics and correlations for the variables of interest. We used Amos 20.0TM for structural equation modeling (SEM) to test the study hypotheses. Although OLS regression is often used for tests of mediation, SEM is also capable of testing mediation and offers several advantages, namely, the simultaneous estimation of the measurement and structural models, inclusion of multiple dependent variables (as in our study), and error term estimation. First, we conducted CFA to verify that the measures used were sound. Following Anderson & Gerbing (1988), we examined separate measurement submodels for the endogenous and exogenous constructs so as not to confound

the measurement. The CFA for the dependent variables yielded a χ2 (24) = 54.585, p = 0.000, comparative fit index (CFI) of 0.980, an IFI of 0.980, a goodness of fit index (GFI) of 0.966, and root mean square of approximation (RMSEA) = 0.061. These 9 items yielded significant loadings on their respective constructs. The CFA for the independent variables, which included 41 items and 9 latent constructs – which by nature of the size of the model introduced the possibility of confounded measurement – resulted in a χ2 (743) = 1476.182, p = 0.000, CFI of 0.873, an IFI of 0.875, a GFI of 0.821, and RMSEA = 0.053. The items measuring the independent variables all yielded significant loadings on their respective constructs. In the last step, we estimated an overall CFA including endogenous and exogenous variables, allowing the error terms of the corresponding perceived and actual use items to co-vary. The overall model also showed acceptable fit: χ2 (1109) = 1957.379, p = 0.000, CFI of 0.892, an IFI of 0.893, a GFI of 0.813, and RMSEA = 0.047. On the basis of the CFAs, an assessment of the residuals, modification indexes, the squared multiple correlations, and Cronbach’s α (the Cronbach’s αs are reported for each measure, along with the specific items, in the Table A1 in Appendix), the measurement model was deemed satisfactory to proceed to structural assessments for hypothesis testing. Once we found our measurement model to be satisfactory via CFAs, we proceeded to test the proposed theoretical model. To assess the goodness of fit of the proposed model, we utilized a variety of fit indices, including the GFI, the normed comparative fit index (NFI), the CFI, and the RMSEA. For the model discussed below, we allowed the error terms of the individual usage components of actual and perceived usage to correlate. The hypothesized model demonstrated reasonable fit, as most values for the fit indices approached 0.90, and the value for the RMSEA remained below 0.08 (Hu & Bentler, 1995; Mulaik et al, 1989). The following values were observed: χ2 (1151) = 2093.12, a CFI of 0.89, an IFI of 0.89, a GFI of.81, and a RMSEA of 0.049, with a 90%

European Journal of Information Systems

10

Five-factor model personality traits

confidence interval = (0.045; 0.052). Accordingly, we will report the findings of our hypothesized model, which are summarized in Table 3. The first four hypotheses were drawn directly from the existing UTAUT model. As predicted by the UTAUT, the relationships between effort expectancy, performance expectancy, and social influence and usage behavior (both perceived and actual usage) were fully meditated by behavioral intentions, which supported H1, H2, and H3. However, contrary to the UTAUT, facilitating conditions did not have a direct relationship to either perceived or actual usage, and H4 was not supported. The second set of four hypotheses concerned the independent effects of the FFM personality traits on system use, over and above the effects of the UTAUT’s contextual predictors. Our additive interaction perspective received partial support. H5 was supported, as conscientiousness was directly and positively associated with both perceived and actual system use, with the standardized path loading slightly higher for actual use. However, H6 did not receive support from the data, as openness to experience did not demonstrate a statistically significant relationship with either perceived or actual system usage. Neuroticism was directly and negatively associated with both perceived and actual system use, with the standardized path loading somewhat stronger for actual use. These results provided support for H7. Extraversion also exhibited a statistically significant association with actual use, but not perceived use, of the m-web system. However, it was a negative relationship, with higher levels of extraversion associated with lower levels of actual system use. Thus, H8 was not supported. Hypotheses 5a, 6a, 7a, and 8a stated that behavioral intentions would partially mediate the relationship between each of the personality traits and system use. However, none of the partial mediation hypotheses received support, as there was not a statistically significant relationship between any of the personality traits and the ‘intent to use’ measure. As noted, the FFM constructs of conscientiousness and neuroticism had direct effects on perceived and actual use of the system that were positive and negative, respectively, consistent with our expectations. Conversely, extraversion had a direct (but unexpectedly negative) effect on actual use. Interestingly, however, we did find support for H9 – actual IT system use and perceived IT system use were positively associated, although the strength of the association between the two was modest. Moreover, we had proposed that agreeableness would have no relationship to system use, and our results supported this argument. Finally, the study results were found while accounting for a significant control variable, GPA. As previous research (Venkatesh et al, 2003; Venkatesh et al, 2012) indicated that age and gender might moderate some of the relationships in the original UTAUT, we conducted multiple post-hoc tests to check for moderating effects. First, as recommended by Pedhazur (1982), we reestimated the model via multiple regression runs. Our findings from these regressions were consistent with the

European Journal of Information Systems

Tim Barnett et al

results provided by the SEM. Second, we tried to assess if the potential interactions of gender and age with the UTAUT variables influenced our results. Thus, we reestimated two hierarchical models with perceived use and actual use as the dependent variables. We first entered all controls, with the interaction terms entered in the second step. In the final step, the FFM personality traits were added. With actual use of the course management system as the dependent variable, only one interaction was significant, that between age and effort expectancy (p<0.05). However, the overall interactions block in the regression model did not add significantly to the R2, with the change in R2 being non-significant, indicating that the interactions did not explain significantly more variance in actual system use. When we added the FFM personality traits in the final step, there was a significant increase in R2, indicating that significantly more variance in actual use of the system was explained. This result was consistent with our SEM results on the effect of the FFM traits on the actual use of the course management system. With perceived use as the dependent variable, two interaction effects were significant, that between age and effort expectancy (p<0.05) and that between gender and behavioral intentions (p<0.05). However, the overall interactions block in the regression model did not result in a significant increase in the model R2, indicating that the interactions did not explain significantly more variance in perceived use. Further, when we added the personality traits in the final step, there was also not a significant change in R2, indicating that, in the case of perceived use, the FFM personality traits did not explain significantly more variance in perceived use in the regression model. This result is also generally consistent with our SEM results, as we observed stronger relationships between the FFM personality traits and actual use, as compared with perceived use, of the course management system.

Discussion and implications To accomplish our primary research aim, we integrated the FFM of personality with the UTAUT model of technology acceptance and tested the resulting research hypotheses. Specifically, we provided a test of the primary relationships between the UTAUT constructs and technology use and then drew upon the interactional psychology perspective to examine the additional predictive ability of users’ personality traits in explaining variance in perceived and actual IT system use. By linking the FFM of personality directly to technology use, we provided additional support for including individual difference variables in the UTAUT. Our findings add to the broad area of research in the IS area tying personality differences to various aspects of technology acceptance and use (Stewart & Segars, 2002; Vishwanath, 2005; Ehrenberg et al, 2008; Hunsinger et al, 2008; Junglas et al, 2008; Korzaan & Boswell, 2008; Lin, 2008; Theotokis et al, 2008; Jacques et al, 2009; Pramatari & Theotokis, 2009; Lin & Ong, 2010; Wilson et al, 2010;

Five-factor model personality traits

Table 3

11

Tim Barnett et al

Summary of hypotheses

Hypotheses

Findings

H1: Behavioral intention will fully mediate the relationship between performance expectancy and IT usage, both actual and perceived. Performance expectancy→Intention Intention→Actual usea Intention→Perceived usea H2: Behavioral intention will fully mediate the relationship between effort expectancy and IT system usage, both actual and perceived Effort Expectancy→Intentiona H3: Behavioral intention will fully mediate the relationship between social influence and IT system usage, both actual and perceived H3a: Socical Influence→Intentiona H4: Facilitating conditions will be positively related to IT sytem usage, both actual and perceived H4 Facilitating conditions→Actual use

Full mediation is supported consistent with UTAUT

H4: Facilitating conditions→Perceived use H5: Conscientiousness→Actual use H5: Conscientousness→Perceived use H5a: Conscientousness→Intentiona H6: Openness→Actual use H6: Openness→Perceived use H6a: Openness→Intentiona H7: Neuroticism→Actual use H7: Neuroticism→Perceived use H7a: Neuticism→Intentiona H8: Extraversion→Actual use H8: Extraversion→Perceived use H8a: Extraversion→Intention H9: Actual use→Perceived use

Standardized path coefficients for final model

0.39 0.20 0.24 Full mediation is supported consistent with UTAUT 0.35 Full mediation is supported consistent with UTAUT 0.15

Direct relationship not supported

ns.

Supported Supported Partial mediation not supported Not supported Not supported Partial mediation not supported Supported Supported Partial mediation not supported Not supported. (Significant in opposite direction) Not supported Partial mediation not supported Supported

ns. 0.28 0.27 ns. ns. ns. ns. −0.30 −0.18 ns. −0.18 ns. 0.04 0.35

Controls and Results for Agreeableness Agreeableness→Actual use Agreeableness→Perceived use Agreeableness→Intentionsa Control GPA→Actual use Control GPA→Perceived use Control- GPA → Intentiona

No relationship was expected No relationship was expected No relationship was expected Significant Significant Significant

ns. ns. ns. 0.12 −0.28 −0.13

a

The relationship between intentions and actual and perceived use does not change and is thus only reported once under H1a. Notes: All significant paths are significant at p<0.05 with critical ratios > ± 1.96

Witt et al, 2010; Al-Natour et al, 2011; Bansal, 2011; Buckner et al, 2012; Zhou & Lu, 2011; Koenigstorfer & Groeppel-Klein, 2012; Lane & Manner, 2012; Terzis et al, 2012; Venkatesh & Windeler, 2012; ; Jackson et al, 2013; Kober & Neuper, 2013; Liu et al, 2013). More specifically, we extend recent work incorporating personality traits (particularly the FFM traits) into TAM and UTAUT-based models of technology acceptance and use (McElroy et al, 2007; Devaraj et al, 2008; Aldas-Manzano et al, 2009; Abu-Shanab et al, 2010; Svendsen et al, 2013). FFM

personality traits had previously been found to moderate the relationship between system characteristics and usage in the TAM (Devaraj et al, 2008), to impact behavioral intentions through the mediating effects of perceived usefulness and ease of use in the TAM (Svendsen et al, 2013), and to affect self-reported internet use directly in a combined TAM/UTAUT model (McElroy et al, 2007). However, our study is the first, to our knowledge, that shows direct effects of three FFM traits on the actual use of a technological system. These significant direct effects were

European Journal of Information Systems

12

Five-factor model personality traits

independent of, and in addition to, the effects of the situational constructs in the UTAUT. Before discussing our results regarding the FFM in more detail, we should note that our study also confirmed three of the basic UTAUT hypotheses (Venkatesh et al, 2003), which link situational constructs to system usage, via behavioral intentions. Specifically, we found that, as expected, performance expectancy, effort expectancy, and social influence had effects on perceived and actual IT system usage that were fully mediated by intentions to use the IT system. By empirically replicating three of the main effect hypotheses of the UTAUT, we contribute to the development of knowledge in social sciences in general (Lamal, 1991). Indeed, it is particularly important to replicate research that is considered a key theory or hypothesis in a particular field of study (Lamal, 1991), as the UTAUT is in the IS literature. Contrary to the UTAUT, however, we found no relationship between facilitating conditions and actual or perceived usage of the IT system. However, this nonsignificant finding may be related to the m-web course management system used in our study, which is a basic and relatively easy to use system. Coupled with the fact that university students are typically routine users of computers and similar technologies, our sample may have had less need for support or guidance in using the system. Thus, facilitating conditions may be more important in studies of more complex technological systems, or of users who are not as technologically savvy. The relationships observed in our study were generally supportive of the interactional psychology perspective, demonstrating an additive explanatory effect of the situational (UTAUT) and personality constructs (FFM). In our SEM analysis, two of the four personality traits (conscientiousness and neuroticism) did have a direct effect on both perceived and actual use of the IT system. Specifically, conscientiousness was positively associated with both perceived and actual IT system use. Although Devaraj et al (2008), in another student sample, found that conscientiousness moderated and strengthened relationships between TAM variables and behavioral intentions and Svendsen et al (2013), in a study of adults, found that the TAM constructs (1) perceived usefulness and (2) ease of use mediated the relationship between conscientiousness and behavioral intentions to use a hypothetical software technology, our study is the first to find the direct effect of conscientiousness on actual system use in a TAM or UTAUT-based context. This finding is consistent with the perspective that conscientiousness, as a personality trait characterized by dependability, responsibility, and a focus on achievement, leads individuals to set goals and engage in behaviors (such as use of a course management system) that help them to succeed (Major et al, 2006). It is also consistent with the notion that conscientiousness is correlated with a learning goal orientation (Payne et al, 2007) and is also linked with the importance placed on productivity-enhancing technological systems (Lane & Manner, 2012) and the

European Journal of Information Systems

Tim Barnett et al

enjoyment of using technological systems (Wang et al, 2012b). Also as hypothesized, neuroticism was negatively associated with both perceived use and actual system usage in our analysis. This finding is consistent with the Devaraj et al (2008) finding that neuroticism was negatively associated with perceived usefulness, and is also consistent with Svendsen et al (2013), who found a negative association between neuroticism and behavioral intentions to use a hypothetical software technology. However, our study is the first to show that neuroticism is directly and negatively linked to the actual use of a technological system. The literature on neuroticism and behavior suggests that individuals high on neuroticism often lack motivation to learn and do not seek out learning situations (Major et al, 2006), that neurotic individuals do not usually have a strong learning goal orientation (Payne et al, 2007), and that neuroticism is unrelated to the importance placed on productivity-enhancing technological systems (Lane & Manner, 2012). Our findings are consistent with these expectations. However, it should be noted that, contrary to the tenor of our results or those of Devaraj et al (2008) and Svendsen et al (2013), McElroy et al (2007) found that neuroticism was positively linked to users’ expressed willingness to buy and sell online. These findings suggest that the negative effect of neuroticism on technology use may be more applicable to learning situations, and not other types of technology use, such as shopping or selling. In our analysis, a third FFM trait, extraversion, had a direct but unexpectedly negative effect on actual use. Extraversion had previously been linked positively to enjoyment of blogging or other technological systems (Wang et al, 2012a, b), the perceived importance of a computer-based acceptance model (Terzis et al, 2012), and trust in a mobile service provider (Zhou & Lu, 2011). TAM and UTAUT-based studies, however, have shown mixed results, with Devaraj et al’s, 2008 TAM study finding that extraversion moderated the subjective norm–behavioral intention relationship, but Svendsen et al (2013) finding that extraversion affected behavioral intention only through its impact on the TAM constructs perceived usefulness and perceived ease of use. In a mixed TAM and UTAUT model, McElroy et al, (2007) found no significant relationships between extraversion and other model constructs. Our finding that extraversion negatively and directly impacted actual use may be explained through a more thorough understanding of the extraversion construct. Extraversion is characterized by low emotional arousal, which, in turn, may lead to a focus on external sources of stimuli, such as interaction with others (Eysenck, 1967, 1973). As such, research has concluded that extraverts perform better in groups or on tasks requiring significant interaction with others (Mount et al, 1998). In a similar fashion, Landers & Lounsbury (2006) found extraversion to be negatively related to internet use. Given the individualized nature of computer systems, in which many tasks require minimal interaction with others (such as the tasks

Five-factor model personality traits

included in this study), perhaps our finding that extraversion was significantly and negatively related to actual use should not be so surprising. If this finding is replicated in future studies, it may be that extraverts perform well in training exercises where interaction with others is included, and then simply choose not to be heavy system users when doing so might cause isolation from others. Neither agreeableness nor openness to experience was related to perceived or actual use in our study. In general, IS studies have linked agreeableness with some aspects related to technology use, such as blogging enjoyment (Wang et al, 2012a), perceived ease of use of a computerbased assessment (Terzis et al, 2012), and trust in a mobile service provider (Zhou & Lu, 2011). However, studies grounded in TAM and/or UTAUT models have found that agreeableness is unrelated to technology acceptance (McElroy et al, 2007; Svendsen et al, 2013), with the exception of the moderating effect it had on the subjective norms–behavioral intentions link in the study by Devaraj et al (2008). Although we did not expect agreeableness to have any direct effects on system use, the finding that openness did not directly affect technology use is inconsistent with our expectation and with previous findings linking openness to self-reported use (McElroy et al, 2007), although it is generally consistent with Svendsen et al (2013), who found that openness did not affect behavioral intentions to use a technology. One possible explanation for our lack of significant results is that openness might not affect use of an IT system either directly or through behavioral intentions, but instead through increased effort expectancy. One could speculate that openness to experience might translate into greater effort expectancy because of a higher disposition to experience all features of the system. Indeed, the observed correlation between the two variables, although modest, was statistically significant (r = 0.12, p<0.05). Moreover, openness to experience might impact user behaviors when the situation calls for broad, general information seeking, such as web exploration (McElroy et al, 2007); however, these effects may not be present in a specific learning-oriented context as was used in this study. Prior studies of personality and technology use have not addressed both perceived and actual usage. Thus, an additional interesting finding of our research was the observed relationship between actual and perceived usage, as the two were correlated, but only moderately at r = 0.33. This leads us to essentially the same conclusion Straub and his coauthors drew in 1995: ‘users are poor estimators of their own behavior’ (Straub et al, 1995, p. 1340). Future research should focus on why perceived and actual usage are not more closely related. Indeed, it has been suggested that this low correlation could be explained by differences in information-processing capabilities of individuals (Straub et al, 1995). A recent study suggested that individual differences could be a source for over/under-predicting usage behaviors (Barnett et al, 2007). Indeed, Barnett et al’s (2007) study found that personality traits affected actual and perceived use differently (i.e., there were

Tim Barnett et al

13

differences in path coefficients). More research is needed to understand the relatively weak relationship between perceived and actual use. A related and intriguing finding of this study relates to the effect of our control variable, GPA, on perceived and actual usage behavior. Although a higher GPA was positively related to actual use, it was negatively related to perceived use, highlighting the need for studies to include both types of usage measures. This finding might also suggest that one’s cognitive ability may systematically lead to an over- or under-estimation of one’s usage behavior. Future research will need to investigate this issue in more detail. We also need to address limitations of our study. Although we replicated three of the four main effects of the UTAUT in our primary study SEM analysis, in our post hoc analyses we did not find all the significant moderator effects for gender and age that were found by Venkatesh et al (2003). While their study found support for gender and age effects, they note that gender appeared ‘to work in concert with age’ (p. 469). For example, they found that the performance expectancy – intention to use relationship was stronger for men and younger workers. Venkatesh et al (2003) go on to explain ‘We interpret our findings to suggest that as the younger cohort of employees in the workforce mature, gender differences in how each perceives information technology may disappear’ (p. 469). Indeed our relatively weak moderator findings for age and gender support this exact contention. Our sample included younger IT users with IT experience. Future research, utilizing subjects with greater variance in age and experience, as well as on other characteristics, may reveal different findings. Future research could investigate the interaction of gender and age not only on the variables in the core UTAUT model, but also on the personality variables in our conceptual model. Although the web-based system in our study is similar to other commonly used web-based systems that are used in business, our sample consisted of undergraduate students, which may limit the generalizability of findings to some extent. This concern is somewhat mitigated by the fact that personality traits are considered relatively stable over time, and that there is no theoretical reason why the effects of personality on intention and usage behavior should differ in similar web-based applications. Our study investigated perceived and actual usage, but did not investigate how effectively the individuals in our study were able to use the information that was obtained through their usage behavior. That is, some individuals may access the system fewer times than others, but get an equal or greater benefit because they are more efficient in its use. Indeed, as Taylor & Todd (1995a) point out, actual time spent on the system lacks practical relevance if the individual cannot translate the time spent with the system into actual performance outcomes. Future research might examine not only potential performance implications of the system, but also the motivation of the individual user’s needs. For example, students in our sample might have

European Journal of Information Systems

14

Five-factor model personality traits

accessed the grades portion of the system not only because of their effort or performance expectancies or their personality traits, but also out of a need for feedback or selfefficacy. Further, our results cannot be generalized to nonusers. Future research might examine the role of personality on non-users’ desires or intentions to use computers for the first time. Finally, our hypotheses were based on research in learning and our model was tested using data from an educational setting, thus, perhaps limiting the generalizability of the study. Future research in employment settings may provide unique additional insights into how personality impacts intentions to use, as well as perceived and actual use. In conclusion, the interactional psychology approach used in our study, while exploratory in nature, provides a useful extension to the technology acceptance literature in

Tim Barnett et al

general, and the UTAUT in particular, by showing that the FFM traits provide additional insights about individuals’ actual use of technology. Our findings combined with those of earlier research (e.g., McElroy et al, 2007; Devaraj et al, 2008; Svendsen et al, 2013), show that the FFM traits are potentially important in developing a more complete model of technology acceptance and use. As noted by Devaraj et al (2008, p. 103), ‘The predictive power of other IS models may be enhanced by incorporating personality variables, and the FFM appears to be a useful framework for identifying the relevant domains of personality’. Our study is, however, only one step. Future research needs to investigate mediating processes between personality traits and usage behavior, as well as moderators of these relationships.

About the Authors Tim Barnett is the Bobby & Barbara Martin Fellow and Professor of Management at Mississippi State University, Starkville, Mississippi, U.S.A. Allison W. Pearson is the Jim and Julia Rouse Professor of Management and a Giles Professor at Mississippi State University, Starkville, Mississippi, U.S.A. Rodney Pearson is the Keil Innovation Fellow and a Professor of Information Systems in the Department of

Management and Information Systems at Mississippi State University, Starkville, Mississippi, U.S.A. Franz W. Kellermanns is the Addison H. & Gertrude C. Reese Endowed Chair and Professor of Management in the Belk College of Business at the University of North Carolina – Charlotte. He holds a joint appointment with the INTES Center at the WHU–Otto Beisheim School of Management (Germany).

References ABU-SHANAB E, PEARSON JM and SETTERSTROM AJ (2010) Internet banking and customers’ acceptance in Jordan: the unified model’s perspective. Communications of the Association for Information System 26(Article 23), 493–524. AGARWAL R and KARAHANNA E (2000) Time flies when you are having fun: cognitive absorption beliefs about information technology usage. MIS Quarterly 24(4), 665–694. AGARWAL R and PRASAD J (1999) Are individual differences germane to the acceptance of new information technologies? Decision Sciences 30(2), 361–391. AJZEN I (1991) The theory of planned behavior. Organizational Behavior and Human Decision Processes 50(2), 179–211. AJZEN I and FISHBEIN M (1980) Understanding Attitudes and Predicting Social Behavior. Prentice-Hall, Englewood Cliffs, NJ. ALDAS-MANZANO J, RUIZ-MAFE C and SANZ-BLAS S (2009) Exploring individual personality factors as drivers of M-shopping acceptance. Industrial Management & Data Systems 109(6), 739–757. ALLPORT GW and ODBERT HS (1936) Trait-names: a psycho-lexical study. Psychological Monographs 47, No. 211. AL-NATOUR S, BENBASAT I and CENFETELLI R (2011) The adoption of online shopping assistants: perceived similarity as an antecedent to evaluative beliefs. Journal of the Association for Information Systems 12(5), 347–374. ANDERSON JC and GERBING DW (1988) Structural equation modeling in practice: a review and recommended two-step approach. Psychological Bulletin 103(3), 411–423. ARMITAGE CJ and CHRISTIAN J (2003) From attitudes to behavior: basic and applied research on the theory of planned behavior. Current Psychology: Developmental, Learning, Personality, Social Behavior 22(3), 187–195. BANSAL G (2011) E-book usage: role of environmental consciousness, personality, and past usage. Journal of Computer Information Systems 52(2), 93–104. BARNETT T, KELLERMANNS FW, PEARSON A and PEARSON RA (2007) Measuring system usage: replication and extensions. Journal of Computer Information Systems 47(2), 76–85.

European Journal of Information Systems

BARRICK MR and MOUNT MK (1991) The big five personality dimensions and job performance: a meta analysis. Personnel Psychology 44(1), 1–26. BLOCK J (1995) A contrarian view of the five-factor approach to personality description. Psychological Bulletin 117(2), 187–215. BOAG S (2011) Explanation in personality psychology: ‘verbal magic’ and the five-factor model. Philosophical Psychology 24(2), 223–243. BOWERS KS (1973) Situationism in psychology: an analysis and a critique. Psychological Review 80(5), 307–336. BROWN SA, DENNIS AR and VENKATESH V (2010) Predicting collaboration technology use: integrating technology adoption and collaboration research. Journal of Management Information Systems 27(2), 9–53. BUCKNER V,JE, CM CAASTILLE and SHEETS TL (2012) The five factor model of personality and employees’ excessive use of technology. Computers in Human Behavior 28(5), 1947–1953. CHAN FKY, THONG J, VENKATESH V, BROWN SA, JEN-HWA HU,P and KAR YT (2010) Modeling citizen satisfaction with mandatory adoption of an E-government technology. Journal of the Association for Information Systems 11(10), 519–549. CHANG L, CONNELLY BS and GEEZA AA (2012) Separating method factors and higher order traits of the big five: a meta-analytic multitraitmultimethod approach. Journal of Personality and Social Psychology 102 (2), 408–426. COSTA PT and MCCRAE RR (1988) Personality in adulthood: a six-year longitudinal study of self-reports and spouse ratings on the NEO Personality Inventory. Journal of Personality and Social Psychology 54(5), 853–863. COSTA PT and MCCRAE RR (1992) Four ways five factors are basic. Personality and Individual Differences 13(6), 653–665. COTE S and MINERS CTH (2006) Emotional intelligence, cognitive intelligence, and job performance. Administrative Science Quarterly 51(1), 1–28. DAVIS F. (1989) Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly 13(3), 319–340.

Five-factor model personality traits

DAVIS FD, BAGOZZI RP and WARSHAW PR (1989) User acceptance of computer technology: a comparison of two theoretical models. Management Science 35(8), 982–1003. DEVARAJ S, EASLEY RF and CRANT JM (2008) How does personality matter? Relating the five-factor model to technology acceptance and use. Information Systems Research 19(1), 93–105. DEVOLDER P, PYNOO B, SIJNAVE B, VOEW T and DUYCK P (2012) Framework for user acceptance: clustering for fine-grained results. Information & Management 49(5), 233–239. DIGMAN JM (1990) Personality structure: emergence of the five-factor model. Annual Review of Psychology 41(1), 417–440. DRISKELL JE, HOGAN J, SALAS E and HOSKIN B (1994) Cognitive and personality predictors of training performance. Military Psychology 6(1), 31–46. DUDLEY NM, ORVIS KA, LEBIECKI JE and CORTINA JM (1996) A meta-analytic investigation of conscientiousness in the prediction of job performance: examining the intercorrelations and the incremental validity of narrow traits. Journal of Applied Psychology 91(1), 40–57. EHRENBERG A, JUCKES S, WHITE KM and WALSH SP (2008) Personality and selfesteem as predictors of young people’s technology use. CyberPsychology & Behavior 11(6), 739–741. EPSTEIN S (2010) The big five model: grandiose ideas about surface traits as the foundation of a general theory of personality. Psychological Inquiry 21(1), 34–39. EYSENCK HJ (1967) The Biological Basis of Personality. Charles C. Thomas, Springfield, IL. EYSENCK HJ (1973) Eysenck on Extraversion. Wiley, New York. FILLION G, BRAHAM H and EKIONEA JPB (2012) Testing UTAUT on the use of ERP systems by middle managers and end-users of medium-to large-sized Canadian enterprises. Academy of Information & Management Sciences Journal 15(2), 1–28. GEORGE J (1992) The role of personality in organizational life: issues and evidence. Journal of Management 18(2), 185–203. GOLDBERG LR (1990) An alternative description of personality: the big-five factor structure. Journal of Personality and Social Psychology 59(6), 1216–1229. GOLDBERG LR (1999) A broad-bandwidth, public domain, personality inventory measuring the lower-level facets of several five-factor models. In Personality Psychology in Europe (MERVIELDE I, DEARY I, DE FRUYT F and OSTENDORF F, Eds) Vol. 7, pp 7–28, Tilburg University Press, Tilburg, The Netherlands. HARRISON AW and RAINER K (1992) The influence of individual differences on skill in end-user computing. Journal of Management Information Systems 9(1), 93–111. HINKIN TR (1995) A review of scale development practices in the study of organizations. Journal of Management 21(5), 967–988. HINKIN TR (1998) A brief tutorial on the development of measures for use in survey questionnaires. Organizational Research Methods 1(1), 104–121. HU LT and BENTLER PM (1995) Evaluating model fit. In Structural Equation Modeling: Concepts, Issues, and Applications (HOYLE RH, Ed), pp 76–99, Sage, Thousand Oaks, CA. HUNSINGER M, POIRIER CR and FELDMAN RS (2008) The roles of personality and class size in student attitudes toward individual response technology. Computers in Human Behavior 24(6), 2792–2798. HURTZ GM and DONOVAN JJ (2000) Personality and job performance: the big five revisited. Journal of Applied Psychology 85(6), 869–879. JACKSON JD, YI MY and PARK JS (2013) An empirical test of three mediation models for the relationship between personal innovativeness and user acceptance of technology. Information & Management 50(4), 154–161. JACQUES PH, GARGER J, BROWN CA and DEALE CS (2009) Personality and virtual reality team candidates: the roles of personality traits, technology anxiety and trust as predictors of perceptions of virtual reality teams. Journal of Business and Management 15(2), 143–157. JOHN OP and NAUMANN LP (2010) Surviving two critiques by block? The resilient big five have emerged as the paradigm for personality trait psychology. Psychological Inquiry 21(1), 44–49. JOHN OP, NAUMANN LP and SOTO CJ (2008) Paradigm shift to the integrative big five trait taxonomy: history, measurement, and conceptual issues. In Handbook of Personality: Theory and Research ( JOHN OP, ROBINS RW and PERVIN LA, Eds), pp 114–158, Guilford Press, New York.

Tim Barnett et al

15

JUDGE TA and ILIES R (2002) Relationship of personality to performance motivation: a meta-analytic review. Journal of Applied Psychology 87(4), 797–807. JUNGLAS IA, JOHNSON NA and SPITZMULLER C (2008) Personality traits and concern for privacy: an empirical study in the context of location-based services. European Journal of Information Systems 17(4), 387–402. KOBER SE and NEUPER C (2013) Personality and presence in virtual reality: does their relationship depend on the used presence measure? International Journal of Human-Computer Interaction 29(1), 13–25. KOENIGSTORFER J and GROEPPEL-KLEIN A (2012) Consumer acceptance of the mobile internet. Marketing Letters 23(4), 917–928. KORZAAN ML and BOSWELL KT (2008) The influence of personality traits and information privacy concerns on behavioral intentions. Journal of Computer Information Systems 48(4), 15–24. LAIDRA K, PULLMANN H and ALLIK J (2007) Personality and intelligence as predictors of academic achievement: a cross-sectional study from elementary to secondary school. Personality and Individual Differences 42(3), 441–451. LAMAL PA (1991) On the importance of replication. In Replication Research in the Social Sciences (NEULIEP JW, Ed), pp 31–35, Sage, London. LANDERS RN and LOUNSBURY JW (2006) An investigation of big five and narrow personality traits in relation to internet usage. Computers in Human Behavior 22(2), 283–293. LANE W and MANNER C (2012) The influence of personality traits on mobile phone application preferences. Journal of Economics & Behavioral Studies 4(5), 252–260. LEWIN K (1951) Formalization and progress in psychology. In Field theory in social science (CARTWRIGHT D, Ed), Harper Publishing, New York. LIM B-C and PLOYHART RE (2006) Assessing the convergent and discriminant validity of Goldberg’s international personality item pool: a multitrait-multimethod examination. Organizational Research Methods 9(1), 29–54. LIN CS (2008) Exploring the personality trait of self-monitoring on technology usage of web portals. CyberPsychology & Behavior 11(2), 235–238. LIN MY and ONG CS (2010) Understanding information systems continuance intention: a five-factor model of personality perspective. PACIS 2010 Proceedings, Paper 52, Taipei, Taiwan. LIU Y., LI H. and HU F (2013) Website attributes in urging online impulse purchase: an empirical investigation on consumer perceptions. Decision Support Systems 55(3), 829–837. MAJOR DA, TURNER JE and FLETCHER TD (2006) Linking proactive personality and the big five to motivation to learn and development activity. Journal of Applied Psychology 91(4), 927–935. MARAKAS GM, JOHNSON RD and PALMER JW (2000) A theoretical model of differential social attributions toward computing technology: when the metaphor becomes the model. International Journal of Human-Computer Studies 52(4), 719–750. MCCRAE RR and COSTA PT (1989) Rotation to maximize the construct validity of factors in the NEO personality inventory. Multivariate Behavioral Research 24(1), 107–124. MCELROY JC, HENDRICKSON AR, TOWNSEND AM and DEMARIE SM (2007) Dispositional factors in internet use: personality versus cognitive style. MIS Quarterly 31(4), 809–820. MOUNT MK and BARRICK MR (1995) The big five personality dimensions: implications for research and practice in human resources management. In Research in Personnel and Human Resource Management (ROWLAND KM and FERRIS G, Eds) Vol. 13, pp 153–200, JAI Press, Greenwich, CT. MOUNT MK, BARRICK MR and STEWART GL (1998) Five factor model of personality and performance in jobs involving interpersonal interactions. Human Performance 11(2/3), 145–165. MULAIK SA, JAMES LR, ALSTINE JV, BENNETT N, LING S and STILWELL CD (1989) Evaluation of goodness-of-fit for structural equation models. Psychological Bulletin 105(3), 430–445. NELSON DL (1990) Individual adjustment to information driven technologies: a critical review. MIS Quarterly 14(1), 79–100. NEUFELD DJ, DONG L and HIGGINS C (2007) Charismatic leadership and user acceptance of information technology. European Journal of Information Systems 16(4), 494–510.

European Journal of Information Systems

16

Five-factor model personality traits

NORMAN WT (1963) Toward an adequate taxonomy of personality attributes: replicated factor structure in peer nomination personality ratings. Journal of Abnormal and Social Psychology 66(6), 574–583. ONES DS and VISWESVARAN C. (1996) Bandwidth-fidelity dilemma in personality measurement for personnel selection. Journal of Organizational Behavior 17(6), 609–626. PAYNE SC, YOUNGCOURT SS and BEAUBIEN JM (2007) A meta-analytic examination of the goal orientation nomological net. Journal of Applied Psychology 92(1), 128–150. PEDHAZUR EJ (1982) Multiple Regression in Behavioral Research: Explanation and Prediction. Holt, Rinehart & Winston, New York. PRAMATARI K and THEOTOKIS A (2009) Consumer acceptance of RFIDenabled services: a model of multiple attitudes, perceived system characteristics, and individual traits. European Journal of Information Systems 18(6), 541–552. SALGADO JF (1997) The five factor model of personality and job performance in the European community. Journal of Applied Psychology 82(1), 30–43. SCHNEIDER B (1983) Interactional psychology and organizational behavior. Research in Organizational Behavior 5(1), 1–32. STEWART KA and SEGARS AH (2002) An empirical examination of the concern for information privacy instrument. Information Systems Research 13(1), 36–49. STRAUB D, LIMAYEM M and KARAHANNA-EVARISTO E (1995) Measuring system usage: implications for theory testing. Management Science 41(8), 1328–1342. SVENDSEN GB, JOHNSEN JK, ALMAS-SORENSEN L and VITTERSO J (2013) Personality and technology acceptance: the influence of personality factors on the core constructs of the technology acceptance model. Behaviour & Information Technology 32(4), 323–334. TAYLOR S and TODD P (1995a) Assessing IT usage: the role of prior experience. MIS Quarterly 19(4), 561–570. TAYLOR S and TODD P (1995b) Understanding information technology usage: a test of competing models. Information Systems Research 6(2), 144–176. TERBORG JR (1981) Interactional psychology and research on human behavior in organizations. Academy of Management Review 6(4), 569–576. TERZIS V, MORIDIS CN and ECONOMIDES AA (2012) How student’s personality traits affect computer based assessment acceptance: integrating BFI with CBAAM. Computers in Human Behavior 28(5), 1985–1996. TETT RP, JACKSON DN and ROTHSTEIN M (1991) Personality measures as predictors of job performance: a meta-analytic review. Personnel Psychology 44(4), 703–742. THATCHER JB and PERREWE PL (2002) An empirical examination of individual traits as antecedents to computer anxiety and computer self-efficacy. MIS Quarterly 26(4), 381–396.

European Journal of Information Systems

Tim Barnett et al

THETOKIS A, VLACHOS PA and PRAMATARI K (2008) The moderating role of customer-technology contact on attitude towards technology-based services. European Journal of Information Systems 17(4), 343–351. THOMPSON RL, HIGGINS CA and HOWELL JM (1991) Personal computing: toward a conceptual model of utilization. MIS Quarterly 15(1), 125–143. VASILOPOULOS NL, CUCINA JM and MCELREATH JM (2005) Do warnings of response verification moderate the relationship between personality and cognitive ability? Journal of Applied Psychology 90(2), 306–322. VENKATESH V and DAVIS FD (1996) A model of the antecedents of perceived ease of use: development and test. Decision Sciences 27(2), 451–481. VENKATESH V and DAVIS FD (2000) A theoretical extension of the technology adoption model: four longitudinal field studies. Management Science 46(2), 186–204. VENKATESH V, MORRIS MG, DAVIS GB and DAVIS FD (2003) User acceptance of information technology: toward a unified view. MIS Quarterly 27(3), 425–478. VENKATESH V, THONG L and XIN X (2012) Consumer acceptance and use of information technology: extending the unified theory of acceptance and use of technology. MIS Quarterly 36(1), 157–178. VENKATESH V and WINDELER JB (2012) Hype or help? A longitudinal field study of virtual world use for team collaboration. Journal of the Association for Information Systems 13(10), 735–771. VISHWANATH A (2005) Impact of personality on technology adoption: an empirical model. Journal of the American Society for Information Science and Technology 56(8), 803–811. WANG YS, LIN HH and LIAO YW (2012a) Investigating the individual difference antecedents of perceived enjoyment in students’ use of blogging. British Journal of Educational Technology 43(1), 139–152. WANG W, NGAI EWT and WEI H (2012b) Explaining instant messaging continuance intention: the role of personality. International Journal of Human-Computer Interaction 28(8), 500–510. WILSON K, FORNASIER S and WHITE KM (2010) Psychological predictors of young adults’ use of social networking sites. Cyberpsychology, Behavior, and Social Networking 13(2), 173–177. WITT EA, MASSMAN AJ and JACKSON LA (2010) Trends in youth’s videogame playing, overall computer use, and communication technology use: the impact of self-esteem and the big five personality factors. Computers in Human Behavior 27(2), 763–769. ZHOU T. and LU Y (2011) The effects of personality traits on user acceptance of mobile commerce. International Journal of Human-Computer Interaction 27(5), 545–561. ZMUD RW (1979) Individual differences and MIS success: a review of the empirical literature. Management Science 25(10), 966–979.

Five-factor model personality traits

17

Tim Barnett et al

Appendix

Table A1 Construct Time 1 Effort expectancy

Performance expectancy

Behavioral intentions to use

Social influence Facilitating conditions

FFM Conscientiousness

Agreeableness

Openness to experience

Extraversion

Neuroticism

Time 2 Perceived frequency of use

Time 3 Actual frequency of use

Control: GPA

Scale Items

Item

Alpha

It is easy to get m-web to do what I want to do M-web is flexible to work with M-web is easy to use Using m-web improves my performance in this course Using m-web increases my productivity in this course Using m-web increases my effectiveness in this course I plan to use m-web very often during the rest of the semester I intend to use m-web frequently during the rest of the semester I plan to use m-web much during the rest of the semester People who influence my behavior think I should use m-web People who are important to me think that I should use m-web I have the knowledge necessary to use m-web The m-web system is not compatible with other systems I have useda A specific person is available for assistance with m-web system difficulties

0.94

I am always prepared I waste my timea I find it difficult to get down to worka I get chores down right away I carry out my plans I shirk on my dutiesa I have a good word for everyone I get back at othersa I make people feel at ease I have a sharp tonguea I cut others to piecesa I insult peoplea I believe in the importance of art I enjoy hearing new ideas I am not interested in abstract ideasa I do not like arta I do not enjoy going to art museumsa I would describe my experiences as somewhat dulla I am skilled in handling social situations I know how to captivate people I have little to saya I don’t talk a lota I am the life of the party I am not easily bothered with things I am often down in the dumpsa I panic easilya I rarely get irritated I seldom feel blue I feel comfortable with myself I have frequent mood swingsa

0.89

0.91

0.90 0.77

0.74

0.76

0.76

0.79

0.71

In a typical week, how many times have you looked up grades? In a typical week, how many times have you accessed the email archive? In a typical week how many times have you accessed the course website?

0.79

Actual grade look-up Actual email archive use Actual course website access Provided by Registrar

a

Indicates reverse scoring.

European Journal of Information Systems