Web Systems Design and Online Consumer Behavior [Electronic resources]

Theoretical Foundations

Over the last two decades, a significant body of research has focused on identifying various factors that influence user-acceptance behaviour, putting forward several theoretical models. In particular, the Technology Acceptance Model (TAM), introduced by Davis and his colleagues (Davis, 1989; Davis et al., 1989), has received considerable attention (see Lucas and Spitler, 1999, for a review). Several researches have demonstrated the validity of this model across a wide variety of IS (see Moon and Kim, 2001). Specifically, the model was shown to have good predictive validity for the use of several ITs including e-mail and the Web (Fenech, 1998; Gefen and Straub, 1997).

It has thus become established as a parsimonious yet powerful model for (1) explaining attitude towards using IS, and (2) predicting usage intentions and its adoption. In other words, to understand (1) the causal link between external variable and user acceptance of PC-based applications (Fenech, 1998); and, more recently, (2) human Web acceptance (Johnson and Hignite, 2000; Lin and Lu, 2000).

Technology Acceptance Model

Davis adapted the Theory of Reasoned Action (TRA) to TAM by developing two key beliefs (i.e., usefulness and ease of use) that specifically account for IS usage as a basic dependent variable of IS. TAM adopts the well-established causal chain of beliefs ( attitudes ( intention ( behaviour (TRA) that has been put forward by social psychologists, Fishbein and Ajzen (Ajzen, 1991; Fishbein and Ajzen, 1975). Consistent with TRA, both users’ beliefs determine the attitudes toward using the system. Behavioural intentions to use, in turn, are determined by these attitudes toward using the system. Finally, behavioural intentions to use lead to actual system use (see Figure 9-1).

Figure 9-1: TAM

The first of these main beliefs is perceived usefulness. It is defined as “the degree to which a person believes that using a particular system would enhance his or her performance” (Davis, 1989). Perceived Usefulness was originally seen as a fairly simple concept including components of effectiveness and efficiency (that are mainly related to extrinsic motivation in work contexts). As shown by Davis (1989), perceived usefulness affects usage of computers. Specifically, Teo et al. (1999) found that perceived usefulness has a strong significant relationship with the Web usage. For example, e-shoppers will use the Web sites more if they find them useful for shopping offering quality information helpful for shopping as well as useful functionality (such as online order status tracking capability, Baty and Lee, 1995; Bellman et al., 1999). In short, individuals will use IS if they perceive that such usage would help them to achieve and enhance the desired task performance, even if it is at first difficult to use (Eid and Trueman, 2002).

The second belief is perceived ease of use, defined as “the degree to which a person believes that using a particular system would be free of effort” (Davis, 1989), being determined by the users’ skills and the usability of the system (Venkatesh and Davis, 1996). Perceived ease of use has been (1) used as a measure of system quality in studies of IS success (Seddon, 1997); (2) considered a component of a Web site’s system quality (Liu and Arnett, 2000); and (3) found to influence computer usage and the Internet usage indirectly via (3.1) perceived usefulness (Davis, 1989; Teo et al., 1999) and (3.2) perceived enjoyment (Igbaria et al., 1995; Teo et al., 1999). Thus, ease of use is an important component when measuring user satisfaction with a Web site (Wang et al., 2001) and its usage (Elliot and Fowell, 2000).

In this context, as perceived ease of use has an inverse relationship with the perceived complexity of use of the technology, it can affect perceived usefulness. A system that is difficult to use is less likely to be perceived as useful. Assuming other things being equal, users consider a system more useful when it is more effort-free. These relationships have been examined and supported by many prior studies (Davis, 1989; Davis et al., 1989; Venkatesh and Davis, 1996; Venkatesh and Davis, 2000). Moreover, if the challenges of an activity are beyond the individual’s skill level, demanding more than the individual can handle, a state of anxiety ensues and users (1) interpret challenges as simply functional complexity or obscurity (and not as opportunities for action), (2) do not perceive the system as useful, and thus (3) tend to use the system sporadically. Likewise, a system perceived as difficult to use is less likely to be perceived as enjoyable, leading to decreased usage (Lim, 2002). Thus, from a causal perspective, ease of use may be an antecedent to usefulness rather than a parallel determinant of usage.

In short, as Hubona and Geitz (1997) reported, perceived usefulness and perceived ease of use have sound theoretical foundations. They are therefore widely accepted as valid and predictive measures of future Web usage levels. TAM yields highly consistent results in the acceptance behaviour of the users towards new systems.

However, most of the TAM research has only been conducted from an extrinsic motivation perspective (Igbaria et al., 1996). Researchers have become increasingly aware of the relevance of the non-cognitive aspects of use motives such as emotions, symbolism and hedonistic desires in understanding facets of behaviour. For example, user behaviour-based findings in the intrinsic motivation and self-efficacy research indicate that emotional responses play important roles in determining (1) a person’s attitude towards using the Web, (2) a behavioural intention, and (3) an actual behaviour.

Following Human-Computer Interaction (HCI) Research, several researchers propose the need for incorporating intrinsic human factors or integrating other theories in a specific study to improve its particular and explanatory value (Hu et al., 1999; Legris et al., 2003; Venkatesh and Davis, 2000). For example, psychologists have proposed a variety of theories explaining how behavioural reactions are influenced both by cognition and affect (Berkowitz, 1993; Epstein, 1994; Leventhal, 1984; Zajonc, 1980). Specifically, one of the intrinsic human motives related to prior factors is “flow.” Next we analyse the proposed flow models.

Flow Theory

Flow has been particularly studied in the context of ITs and hypermedia computermediated environments (CMEs, defined by Hoffman and Novak, 1996b, as a distributed computer network used to access and provide hypermedia content). Flow, defined as “the holistic sensation that people feel when they act with total involvement” (Csikszentmihalyi, 1975), has been recommended as a possible metric of the online consumer experience (Agarwal and Karahanna, 2000; Ghani et al., 1991; Ghani and Deshpande, 1994; Hoffman and Novak, 1996b; Novak et al., 2000; Trevino and Webster, 1992; Webster et al., 1993). Therefore, we suggest that flow-based theory could contribute partly to explain attitudes towards using the Web-based technologies and behaviours.

Although a body of research suggests that flow on the Web is fleeting, rarely experienced, associated with the increases in depression and loneliness (see Kraut et al., 1998), and mostly by novice Web users, the growing research concerning theory of optimal flow has been proposed (1) as a useful framework for identifying the factors that influence this experience and, in turn, (2) as a way of defining the nature of compelling online experiences (Table 9-1).

Next we analyse the main factors that influence this optimal experience. We distinguish two submodels for better understanding of the relationships between TAM-based beliefs and flow state.

Submodel 1

Researchers have maintained that involvement is a major socio-psychological variable that explains individual differences (Petty et al., 1981) that impact on attitude and individual behaviour. Following a review of the construct of involvement in psychology, organisational behaviour, and marketing, Barki and Hartwick (1989) conclude that these disciplines have converged in a definition of involvement “as a subjective psychological state, reflecting the importance and personal relevance of an object or event.” In this cognitive-processing context, involvement has also been argued to have a significant effect on consumer subjective perception of how much they think they know about products (Zinkhan and Muderrisoglu, 1985). In turn, Houston and Rothschild (1978) and others have found that involvement increases with familiarity with the stimulus or individual’s prior knowledge (i.e., ability) (Figure 9-2.)

Figure 9-2: Extending model (I)

On the other hand, involvement contributes to the attention focused on the stimulus (Zaichkowsky, 1986) and it is considered as a prerequisite for flow (Hoffman and Novak, 1996b). Ghani and Deshpande (1994) emphasise that the total concentration in an activity is the key characteristic of flow (Figure 9-2). According to Csikszentmihalyi and Csikszentmihalyi (1988), when one is in flow “one simply does not have enough attention left to think about anything else.” The computer screen functions as a limited stimulus field. Moreover, involved users report being mesmerised during their computer interactions. Accordingly, Park and Young (1986) note that users -for whom extrinsic motives are salient-focus their attention on utilitarian cues and evoke cognitive responses. In turn, users — for whom intrinsic motives are salient — focus their attention on symbolic or experiential cues and evoke emotional responses.

Involvement can be thus understood by distinguishing the types of involvement according to the motives underlying involvement (Park and Young, 1986). Particularly, the distinction between extrinsic and intrinsic motives of behaviour suggests two types of involvement:

• Situational involvement arises from several transitory factors that affect the relationship between the individual and the stimulus (Celsi and Olson, 1988). It is externally motivated and it is thus more likely to result in a goal-directed behaviour (Hoffman and Novak, 1996b). Extrinsic motives-based involvement (i.e., SI) is related to the performance of an activity. In this sense, the activity is perceived to be instrumental in achieving valued outcomes that are distinct from the activity itself.

• Enduring involvement is an intrinsically-motivated individual difference variable that is relatively long-lasting. In this context, intrinsic motivation refers to the performance of an activity for no apparent reinforcement other than the process of performing the activity per se. According to no apparent reinforcement, the concept of perceived enjoyment is defined as “the extent to which the activity of using the computer is perceived to be enjoyable in its own right, apart from any performance consequences that may be anticipated” (Davis et al., 1992). In short, intrinsic motives to use the Web could be associated with frequent Web use for intrinsically enjoyable purposes.

Therefore, situational involvement reflects temporary feelings of involvement that accompany a particular situation, whereas enduring involvement is an individual difference variable representing the general, long-run concern with a stimulus that a consumer brings to a situation (Richins et al., 1992).

According to intrinsic motives related to enduring involvement, enjoyment has been identified as an important motivational factor in computer use, (1) contributing towards creativity and exploratory use behaviour (Ghani, 1991), as well as (2) being a major dimension of optimal experience or flow, which has been above described as an intrinsically enjoyable experience (Csikszentmihalyi, 1975). Specifically, research on the use of the Web has found empirical support for enjoyment as a driver of Web usage (e.g., Atkinson and Kydd, 1997; Moon and Kim, 2001; Teo et al., 1999). If individuals like and enjoy their Web browsing experience, it is likely that they are going to (1) involve in browsing and, in turn, (2) enhance their online service perceptions (e.g., perceived usefulness and ease of use). Use of the Web may therefore evoke emotional values that are not only captured by ease of use or usefulness (Hoffman and Novak, 1996ab; Singh and Nikunj, 1999). Use of the Web goes beyond utilitarian aspects to include intrinsic enjoyable experience (Berthon et al., 1996; Pine and Gilmore, 1998). For example, Davis et al. (1992) argued that “while usefulness will once again emerge as a major determinant of intentions to use a computer in the workplace, enjoyment will explain significant variance in usage intentions beyond that accounted for by usefulness alone.”

Moreover, HCI-based research using the TAM model has found that perceived enjoyment of using a system (e.g., Web) has a relationship with a perceived ease of use (Venkatesh, 1999; Venkatesh, 2000; Moon and Kim, 2001) and perceived usefulness of the system (Agarwal and Karahanna, 2000) (Figure 9-2).

On the one hand, Agarwal and Karahanna (2000) found a multi-dimensional construct called cognitive absorption (similar to flow state) which had a significant influence on usefulness over and above ease of use. Venkatesh (2000) showed that enjoyment influenced usefulness via ease of use without assessing its direct effect on usefulness over and above ease of use (Yi, 2003). Likewise, several researchers note that when the usage experience is more enjoyable the impact of perceived usefulness on Web usage could be relatively lower. This prior phenomenon is based on a cognitive consistency argument in which the underlying rule is that when usage is emotional, instrumental issues - such as perceived usefulness - ought not to come into one’s main decision making criteria for future usage (Chin et al., 1996). However, the effect of enjoyment on perceived usefulness is still relatively unknown.

On the other hand, Csikszentmihalyi (1975) argued that flow could be enhanced when an individual perceived an activity to be executed easily. Empirical research has also found support for this relationship in traditional settings (Igbaria et al., 1996). It is conceivable that a Web site that is easier to use provides better feedback to a visitor’s stimuli, and consequently, leads to increased enjoyment and flow. Moreover, Venkatesh (1999) compared two training methods (traditional training vs. game-based training) and found that the training method with a component aimed at enhancing intrinsic motivation induced higher ease of use perceptions. Later, as we commented above, Venkatesh (2000) conceptualised enjoyment as an antecedent of ease of use, whose effect increases over time as users gain more experience and perceived control with the system (adapted from Hwang and Yi, 2002).

Finally, studies applying the perspective of flow have shown that to provide intrinsic motivation, some services must represent a certain challenge to the user as antecedent of emotional arousal. It is probably that excessive ease of use that reduces the sense of accomplishment, (1) negatively influences on perceived enjoyment and (2) leads to boredom states. In other words, ease of use is not the only key criterion for Web site design, as Web site usage would decrease with time. On the contrary, a main determinant must be its stimulating use, so that it evokes compelling experiences and therefore increases profitable Web site use (Sánchez-Franco and Rodríguez-Bobada, 2004). According to Csikszentmihalyi (1996), a Web site must be challenging, competitive, and provide feedback to its users in order to encourage the occurrence of flow. As Ginzberg (1978) recommended, system success must be evaluated in terms of the way it is used rather than just the extent of use. Therefore, an important prerequisite for this rewarding experience is that an individual is able to accomplish the task (i.e., ability). But it is equally important for it to be experienced as a challenge and the individual gets (1) stimulation (i.e., arousal) and (2) unambiguous feedback (i.e., perceived control) inherent in the performance of the activity. To complete the global model, we thus introduce a second submodel based on users’ ability, challenge, arousal and perceived control.

Submodel 2

Looking at earlier definitions of flow (Table 9-1), in order to experience flow while engaged in an activity, users must perceive a balance between their abilities (defined as the skill or proficiency) and the challenges of the activity (defined as their opportunities for action on the Internet) (Novak et al., 2000). Particularly, challenges are related to a sense of accomplishment rather than simply functional complexity or obscurity. Both their abilities and challenges must be above a critical threshold (Massimini and Carli, 1988).

The balance facilitates the experience of arousal, perceived control and flow.

• If the challenges of an activity (i.e., those opportunities which provoke users to further explore Web sites) demand more than the individual can transitorily handle, a state of stimulation ensues (i.e., arousal: high challenge/ moderated skill). Users become aroused until they are familiarised with the system through more frequent system usage –e.g., practice or training (Gardner et al., 1989) (Figure 9-3). However, too much stimulation will lead users to making errors and feel out of control (i.e., anxiety as a negative affective reaction toward Web use).

• On the contrary, when the challenges are lower than the individual’s skill level, perceived control (moderated challenge/high skill) may be the result (Figure 9-3). However, if the challenges are too low, users lose interest and tend to use the Web sporadically (i.e., boredom or apathy).

  Figure 9-3: Extending model (II)

According to arousal as an emotional response, it reflects a user’s concern about having the ability to succeed with a new perceived challenge. Arousal can be thus considered as an involvement-based response. However, the user’s concern must be perceived as moderate, important and relevant (1) to acquire Web-based skills and (2) to match the skill level and perceived challenge. On the contrary, too much concern will lead users to feel out of control.

Moreover, if users return to the same Web site over time, it is reasonable to expect (1) learning to occur, (2) perceived challenges to decrease, and (3) session lengths to decline (see, for example, Johnson et al., 2003, on “the power law of practice”). A main recommendation must be thus to promote its stimulating use, so that it permanently evokes (1) arousal and compelling experiences, and (2) more frequent and longer visits. For example, CMC (Computer-Mediated Communication) technologies can stimulate cognitive curiosity and the desire to attain competence with the technology by providing options such as menus that encourage exploration (Malone and Lepper, 1987) and competence attainment. In this sense, arousal, as a consequence of perceived task challenge, is a key factor in the experience of flow.

According to perceived control, it has been studied in the context of electronic commerce and found to have a positive effect on customer attitudes and behaviour (Ghani et al., 1991; Novak et al., 2000; Koufaris et al., 2001-2002). Specifically, it refers to users’ perception of their capabilities to interact in CME. Perceived control comes from: (1) the users’ perception of their ability to adjust the CME; (2) their perception of how the CME responds to their input; plus (3) an environment where challenges are relatively moderate. These users thus believe their actions and abilities determine their successes or failures (Sánchez-Franco and Rodríguez-Bobada, 2004).

Therefore, users with a high level of ability and, consequently, perceived control (Ghani and Despandhe, 1994; Novak et al., 2000): (1) are likely to feel more able to perform the activity, and (2) show a high comfort level. They would be more inclined to feelings of enjoyment while become involved in the activity and, in turn, to use the Web more frequently. Likewise, users become more playful (Lieberman, 1977) and experiential, positively affecting Web exposure length. As Bandura (1982) suggested, “people do not perform maximally, though they possess the constituent skills.” He suggests that “the reason people enjoy challenging tasks is that by testing the upper limits of their competencies, they find out what they are able to do, thereby increasing their feelings of self-efficacy.” On the contrary, those with low self-efficacy expectations in a particular situation will experience unpleasant feelings, such as anxiety, and will behave in unproductive ways, such as avoiding work, and may lack persistence (Bandura, 1977).

Perceived control is thus similar to Bandura’s self-efficacy (1982) defined as “judgements of how well one can execute courses of action required to deal with prospective situations.” It is (1) specific to an action and it can be different across situations or actions; (2) facilitated by the medium adapting to feedback from the individual, and by providing explicit choices among alternatives (Webster et al., 1993), and (3) considered - by several researchers - as an antecedent of perceived behavioural control (i.e., “perceptions of internal and external constraints on behaviour,” Taylor and Todd, 1995).

Finally, perceived control (or self-efficacy) can be related to perceived ease of use. Users regard the system easier to use when their conviction in their own efficacy regarding the particular system is higher (Agarwal et al., 2000; Venkatesh and Davis, 1996; Venkatesh, 2000).

Implications for Web Usage

Based on the above theoretical development (Figure 9-3), two main Web user-types can be theoretically evidenced on a continuum from “pure browsing” to “pure seeking.” The distinction is a continuum rather than a dichotomy. Individual differences drive a person’s information and entertainment consumption processes. Individuals shift from one mode to the other.

When users show ritualised orientations exploring the Web (experiential behaviour), they are moved by an intrinsic motive: “to feel pleasure and enjoyment from the activity itself” (Bloch et al., 1986). Users find the interaction intrinsically interesting. They are involved in the activity for the emotional responses it provides rather than for utilitarian purposes. Thus, a main objective is that a Web site is designed to be stimulating to use and thus to evoke compelling user experiences related to playfulness, exploratory behaviour and positive affects.

According to playfulness and positive affects, Atkinson and Kydd (1997) examined the influence of playfulness on the use of the Web, defined as the degree of cognitive spontaneity in microcomputer interactions (Webster and Martocchio, 1992). They found that both playfulness and usefulness affect its use in different ways, depending on its use for entertaining or for work. Likewise, they found that playfulness is significantly associated with total Web use. Those who are more playful with computers tend to indulge in using a new system just for the sake of using it. Therefore, they in general underestimate the difficulty associated with using a new system (Venkatesh, 2000). Previous computer adoption studies have verified that if users are more playful with computer systems, they are more willing to use the systems (Igbaria et al., 1994; Webster and Martocchio, 1992). In turn, Webster et al. (1993) note that research has suggested that “higher playfulness results in immediate subjective experiences such as positive mood and satisfaction” (Levy, 1983; McGrath and Kelly, 1986; Sandelands et al., 1983). Furthermore, previous research on human-computer interactions (Sandelands and Buckner, 1989; Starbuck and Webster, 1991; Webster and Martocchio, 1992) has shown that higher degrees of pleasure and involvement during computer interactions lead to concurrent subjective perceptions of positive affect and mood (Hoffman and Novak, 1996b).

Also, Amabile (1988) noted that “only the intrinsically motivated person (…) who is motivated by the interest, challenge, and the enjoyment of being in the maze (…) will explore, and take the risk of running into a dead-end here and there.” In this sense, Ghani and Deshpande (1994) examined flow in the context of individuals who used computers in their daily work and found that it had a significant impact on exploratory use of the computer which, in turn, had a significant effect on the extent of computer use.

When users show an instrumental orientation to the Web (goal-directed behaviour), they search for contents adapted to their needs and goals and leave the Web after an active and efficient search. Pure seekers use the Web less for experiential activities and more for goal-directed activities based on perceived usefulness. Thus, an objective is that a Web site is designed to be easy to use and useful to increase profitable Web site usage. Likewise, goal-directed users are generally involved in activities that already have a high extrinsic motivating potential. Such individuals are less likely to seek challenges and evoke arousal in Web use (Ghani and Deshpande, 1994). As users become more skilful, their information search shifts from an extensive manner to a simplified one. Web users can evidence opportunity costs of time and confront a variety of time constraints. An idea put forward in some early empirical research on the Web holds that Web users will continue to browse as long as the expected benefit or value of an additional page view exceeds the cost (Sánchez-Franco and Rodríguez-Bobada, 2004).

In short, as Novak et al. (2000) suggest, because the Web mixes experiential and goal-directed behaviours, the model can be used as a first step in evaluating Web sites in terms of the extent they deliver these two types of experience while users browse.