Theory Extensions

The Decomposed Theory of Planned Behaviour

Understanding individuals’ acceptance and use of information technology has become a popular research area since the 1980s. As stated by Taylor and Todd (1995), one important stream of this area is to identify determinants of behavioural intention to predict usage, whereas another stream uses a diffusion of innovation perspective. Within the first stream, TPB as an intention-based model has been well-applied. Although the three determinants influencing intention were clear, their corresponding belief structures were not well understood in the Information Systems (IS) research domain. To address this limitation, as well as to operationalize TPB in the context of users' acceptance of technology, Taylor and Todd (1995) decomposed the three belief structures by referring to other technology acceptance theories and the stream of innovation diffusions, naming the resultant model the decomposed Theory of Planned Behaviour (DTPB).

Derived from the perceived characteristics of an innovation (Rogers, 1983) and also considering the technology acceptance model (Davis, 1989), attitudinal (behavioural) belief structures were decomposed into perceived usefulness (i.e., perception that the usage of a technology would enhance job performance), ease of use (i.e., perception of the ease or difficulty of using a technology), and compatibility (i.e., the degree to which the technology fits with someone’s values, experiences, and needs). Focusing on the context of technology acceptance in organisations, normative belief structures were decomposed into peer influence and superiors' influence. Control belief structures were decomposed into self-efficacy, resource facilitating conditions, and technology facilitating conditions. Self-efficacy, which is related to perceived ability, stands for someone’s internal constraints to use the technology. The resource facilitating condition (e.g., availability of time and money) and technology facilitating condition (e.g., compatibility with current technologies in use) reflect the external constraints. The model was validated and tested with a sample of 786 students in a university using a computing resource centre as the technology example. Taylor and Todd (1995) also compared DTPB with TPB and TAM, with the results indicating that DTPB explained both behaviour intention and usage behaviour better than the other two theories.

DTPB has been applied to investigate user acceptance of various technologies since its formation. For instance, Shih and Fang (2004) employed DTPB to understand people’s adoption of internet banking and found that 66% of intention could be explained by the decomposed factors. Garay et al. (2019) used DTPB as the framework to explore accommodation managers’ intention to adopt sustainability-oriented innovations. The three factors with their decomposed structures informed 56% of their behavioural intentions. Although DTPB was proposed in the IS area, it has been used in many other disciplines. For example, in education, DTPB has been used as the framework to investigate teachers' and/or students’ adoption of different teaching techniques, such as digital literacy (Sadaf & Gezer, 2020), STEM education (Wu et al., 2022), and e-learning (Santos & Okazaki, 2013).

The Reasoned Action Approach

The reasoned action approach (RAA) was proposed based on TPB by Fishbein and Ajzen (2015) with the aim of maximizing the precision of behaviour prediction. RAA retained the structure of TPB but broke the three determinants of intention down into two related but distinct subcomponents.

Attitude towards behaviour includes experiential and instrumental attitudes as its subcomponents. Experiential attitude is rooted in an individual’s subjective evaluation that performing the behaviour of interest would involve a certain experience, which reflects the more affective aspect of attitude (Hagger et al., 2020). In contrast, instrumental attitude focuses more on the cognitive aspect, and it is based on an individual’s evaluation of the outcome occurring after performing a behaviour. To illustrate, “interesting-boring” and “valuable-worthless” are example descriptors of experiential and instrumental attitudes respectively. It has been suggested that experiential attitude is more closely linked to intention and behaviour compared with instrumental attitude (Tenenbaum & Eklund, 2020).

Subjective norm is reconceptualised as perceived norm in RAA, and it contains injunctive and descriptive norms. Injunctive norm concerns an individual’s perception regarding whether the targeted behaviour would be approved by important others, while descriptive norm is about whether the important others themselves perform the behaviour (Hagger et al., 2020).

Capacity and autonomy are the subcomponents representing the perceived behavioural control construct in RAA. Capacity reflects an individual’s perception of the ease or difficulty of conducting a behaviour, which is noted to be overlapping with self-efficacy (Tenenbaum & Eklund, 2020). Autonomy refers to the extent to which an individual believes the performance of the behaviour is up to them (Hagger et al., 2020).

RAA has attracted a great deal of attention particularly in the field of health behaviours (McEachan et al., 2016). For instance, it was found that various human health behaviours can be understood by RAA, such as sleeping (Branscum, Fay & Senkowski, 2020), smoking (Dobbs et al., 2019), doing sports (Branscum & Fairchild, 2019), and COVID-19 prevention (Norman, Wilding & Conner, 2020). In addition, it has been demonstrated to have the ability to predict health protection and risk behaviours (Conner et al., 2017). Therefore, RAA provides a powerful tool to promote health behaviours (Fishbein, 2008).

Applications

Rooted in psychology, TPB has been applied across numerous disciplines where the understanding and prediction of human behaviour are needed. In an editorial, Bosnjak et al. (2020) conducted a review based on more than 4,200 articles that referenced TPB in the Web of Science bibliographic database. The results revealed that TPB has received considerable attention in a range of areas, particularly business and management (e.g., IS, entrepreneurship, and tourism and hospitality), and psychology (e.g., environmental behaviour, health behaviour, and education research).

In the business and management domains, the crucial and distinctive role of TPB is widely acknowledged in many research areas. For instance, in the IS research field, TPB has been an important lens to understand people’s adoption of technologies in both organisation and consumer contexts. First and foremost, TPB influenced and inspired the development of other important technology acceptance theories as a fundamental theory (Venkatesh & Davis, 2000; Venkatesh et al., 2003). Furthermore, empirical studies have been employing TPB as the theoretical framework to investigate people’s intention to use and purchase, and their actual usage of, technologies. Pavlou and Fygenson (2006) extended TPB to understand and predict people’s adoption of electronic commerce. This piece of work focused on users’ behaviours when getting information and purchasing products from online vendors. Apart from identifying beliefs affecting the three determinants of intentions and examining the relationships proposed within the original TPB framework, they also explored the relationships between intention to get information and intention to purchase, as well as getting information behaviour and purchasing behaviour. The results suggested that attitude, subjective norm, and perceived behavioural control explained more than half of the variances of intention to get information and to purchase. With the rapid development of artificial intelligence (AI) and machine learning, TPB has been used to investigate people’s adoption of products equipped with those technologies. Taking automated vehicles as an example, Kaye et al. (2020) used TPB to explore drivers’ intention to operate conditional and fully automated vehicles and found that TPB is able to explain more than 60% of variance in both scenarios.

Apart from IS, TPB exhibits relevance and robustness in the prediction of entrepreneurial intentions and actions. Using longitudinal survey data collected from Austrian and Finnish citizens aged from 20 to 64 years old, Kautonen et al. (2015) supported all the proposed relationships in TPB, with subjective norm being the most important determinant of intention. Su et al. (2021) employed TPB as the framework to investigate Chinese university students’ entrepreneurial intention, and it was found that 74% of intention can be explained by attitude and perceived behavioural control. However, subjective norm was shown to be a non-significant factor. TPB has also been integrated with other theories to predict entrepreneurial intentions and actions. For instance, in a meta-analysis study (Schlaegel & Koenig, 2014), TPB was integrated with the Entrepreneurial Event Model to predict entrepreneurial intent. It was found that apart from being significant direct predictors of entrepreneurial intent, attitude, subjective norm, and perceived behavioural control were all significant determinants of perceived desirability and perceived feasibility – the key factors of the Entrepreneurial Event Model.

In the field of tourism and hospitality research, TPB also appears to be influential. The theory has been used to investigate and understand people’s willingness to visit or their actual visits to certain destinations. For instance, Quintal et al. (2010) extended TPB with perceived risk and perceived uncertainty to investigate Chinese, Japanese, and South Korean residents’ intention to visit Australia. The results suggested that the explanatory power of the model, and the significance and importance of the determinants, varied observably in those three country settings. This research also found that perceived risk can significantly affect attitude, and perceived uncertainty can significantly influence attitude and perceived behavioural control. Soliman (2021) employed TPB to understand visitors’ intention to revisit a destination, and they also considered tourist motivation, electronic word of mouth (e-WOM), destination familiarity, and destination image as predictors. All of the proposed relationships were supported and the extended TPB explained 69% of the variance. The relationships between e-WOM and the three determinants in TPB were further explored by Jalilv and Samiei (2012) in the context of visiting an Iranian city. It was found that e-WOM generated considerable positive influence on all three determinants of intention.

TPB plays a key role in psychology studies. In the environment research area, TPB was employed to investigate intervening in and changing people’s behaviours to act and consume more sustainably. This is known as promoting consumers’ green behaviours in social marketing (White, Habib & Hardisty, 2019). For instance, Mak et al. (Mak et al., 2018) extended TPB to understand how to promote food waste recycling in both commercial and industrial contexts. Si et al. (2020) extended TPB to explore people’s intention and behaviour in terms of bike usage and sharing. Becker-Leifhold (2018) used TPB to understand collaborative fashion consumption in the clothing industry. In a review of TPB in environmentally focused studies, Yuriev et al. (2020) witnessed an increasing number of studies employing TPB as their theoretical framework. Although the guideline of application of TPB has been respected in this discipline, it has also been noticed that researchers tend to extend TPB to understand the phenomena better with additional constructs such as moral norm, past behaviour, anticipated emotions, environmental awareness, environmental values, and sense of community. Interestingly, some studies reported that the original factors all hold up when TPB is extended with other factors (Blok et al., 2015; Chan & Bishop, 2013; Chu & Chiu, 2003), while other research provided evidence that some of the original three predictors become non-significant. For example, Fielding et al. (2008) found that perceived behavioural control became non-significant, and the importance of subjective norm dropped greatly, when TPB was extended with group membership and self-identity. Shi et al. (2017) found that subjective norm became non-significant when moral norm and descriptive norm were added as the antecedents of intention.

TPB also serves as an important tool to understand health behaviours. For instance, it has been applied to investigate mental disorder prevention (Shi & Kim, 2020), condom usage (Guan et al., 2016), organ donation (Bresnahan et al., 2007), childhood obesity prevention (Andrews, Silk & Eneli, 2010), and healthy eating behaviours (Grønhøj et al., 2012). In addition, TPB offers implications for more effective communications to promote healthy behaviours. Anderson et al. (2013) extended TPB to predict young adults’ routine dental check-ups. They developed messages based on TPB and noted that subjective norm-based messages could prompt dental check-ups significantly. Wang (2009) focused on promoting regular physical activity participation using TPB. Wang suggested that greater attention should be paid to attitude constructs and targets’ personality traits when designing messages.

In the education research domain, Ajzen and Madden (1986) is a landmark paper, which emphasised the role of perceived behavioural control in predicting goal-directed behaviour. It was found that students' perception of their ability to perform a behaviour can affect both their intentions and the behaviour itself. TPB has also been applied to understand both students’ and teachers’ intentions and behaviours in education studies. Martin and Kulinna (2004) employed TPB to investigate teachers’ intention to teach physical education classes that required engaging in moderate to vigorous physical activities. The results demonstrated the significance of all three determinants of intention, accounting for 59% of the variance. Regarding student behaviours, TPB has been used to investigate students’ intention to complete education (Davis et al., 2002) and to withdraw from education (Dewberry & Jackson, 2018). TPB also facilitates the understanding of learning and teaching methods or technique adoption. For instance, Cheon et al. (2012) used TPB to explore students’ readiness to use mobile learning in the higher education sector and all three determinants were again found to positively affect usage intention. A remarkable 87.2% of intention was explained, with perceived behavioural control being the most crucial factor.

Limitations

Having only four concepts to explain human behaviour, TPB has been questioned regarding its trade-off between parsimony and validity. One criticism in this regard is that it is “not taking sufficient account of cognitive and affective processes that are known to bias human judgments and behaviour” (Ajzen, 2011:p1115-1116). For instance, TPB does not include unconscious, irrational, and emotional influences in the theory.

Regarding intention, it has been argued that individuals’ behavioural, normative, and control beliefs are easily influenced and changed by many intervening events, which can lead to changes in their attitudes, subjective norms, and perceived behavioural control (Sniehotta, Presseau & Araújo-Soares, 2014). As a result, their intentions can eventually be affected. However, TPB cannot address this issue since it assumes the formation of intention and the performance of human behaviour as a linear decision-making process. In addition, scholars have been doubtful about the notion that beliefs only lead to one of the three determinants rather than intention directly (Araujo-Soares et al., 2013).

With regard to actual behaviour, the majority of variance is not explained by the concepts included in TPB in many studies, and the validity of TPB varies considerably when moderators, such as demographics, are applied (Sniehotta, Presseau & Araújo-Soares, 2014). TPB also has the limitation that it can only predict one behaviour at one time, while multiple human behaviours may occur simultaneously or sequentially under complicated situations (Yuriev et al., 2020).

Another key criticism of TPB lies in its inability to fully address the intention-behaviour gap: the phenomenon where strong behavioural intentions do not always translate into corresponding actions (Sheeran & Webb, 2016). While TPB emphasises intentions as a central predictor of behaviour, it underestimates the influence of situational and contextual factors which can hinder action despite favourable intentions. Additionally, the model assumes a linear and rational progression from intention to behaviour, and it assumes that humans are rational decision-makers, like most theories that predict behaviour. However, many behaviours are influenced by emotional, habitual, or impulsive factors, which TPB does not adequately account for (Sniehotta, Presseau & Araújo-Soares, 2014). These limitations reduce the predictive power of TPB in real-world contexts, particularly for behaviours that require sustained effort or are influenced by dynamic external factors. Despite its limitations, it is still considered to be one of the most fundamental theories to predict and understand human behaviour.