INTRODUCTION

There have been great and rapid improvements in the area of information and communication in recent years. As a result, modern educational institutions have great tendency to employ technology and the Internet, with the hope of boosting the effectiveness and quality of education due to potentialities of the tools. Consequently, we are witnessing an explosion of innovative ideas in the realm of education.

Along with this progress, principals, instructors and students keep on facing new obstacles (Aldammagh et al., 2021; AlOmoush, 2022). Consequently, the role of the instructors has broadened in the process of turning the traditional classes into global classrooms of e-learning inclusive of concepts such as e-content, e-books, e-training, distant learning and virtual classrooms (Farsi et al., 2023; Shaheen et al., 2021; Yeou, 2016; Zandi et al., 2022).

A look at Google scholars reveals that a great number of papers have been already conducted about the theory of planned behavior (TPB) applied to education domain. There still exists a noticeable gap in factors that influence high school English teachers’ intentions of adopting computer-mediated teaching (Liu & Wang, 2024).  However, the TPB’s application to high school context is limited. This study offers an opportunity to fill this gap. It is also a chance to test the sufficiency of the TPB. Thereby it contributes to the advancement of the growing body of scientific knowledge.

LITERATURE REVIEW

Theory of Reasoned Action (TRA)

Theory of reasoned action was designed by Martin Fishbein in 1975. Fishbein and Ajzen (1975) proposed to predict behavioral intention as a strong predictor of actual behavior (Ho et al., 2015). The theory of reasoned action (Fishbein & Ajzen, 1975) posits that behavioral intentions are a function of two sets of beliefs: Attitude and subjective norms. Attitude is an evaluation of performing a future behavior (Fishbein & Ajzen, 2010, p. 78). Subjective norm is a person’s evaluation of how significant others will react to specific behaviors (Fishbein & Ajzen, 1975, 2010; Paek et al., 2015). Both attitude and subjective norms predict and explain the behavioral intention and actual behavior (Davis, 1989; Wayne, 2018). The theory is illustrated in Figure 1.

 Figure 1. Theory of Reasoned Action Model (Fishbein & Ajzen, 1975)

The theory of reasoned action has been widely used in predicting behavior (Madden et al., 1992). This model has been successful in predicting behavior (Albarracín et al., 2001; Armitage & Conner, 2001; Cooke & Sheeran, 2004; Hagger et al., 2002; Sheppard et al., 1998). Reasoned action theory was improved when Ajzen added the perceived behavioral construct and proposed TPB as a new model (Ajzen & Madden, 1986).

Theory of Planned Behavior (TPB)

The TPB is an extension of the theory of reasoned action (Ajzen & Fishbein, 1980; Fishbein & Ajzen, 1975). The concept of perceived behavioral control differentiates TPB from the theory of reasoned action. Thus, the inclusion of a perceived behavioral control variable as an additional determinant of intention and behavior established TPB (Ajzen, 1985). Perceived behavioral control is a person’s belief that he is capable of, or has control over, performing a specific behavior (Ajzen, 2002) and this definition is broadly similar to Bandura’s self-efficacy conception (Fishbein & Ajzen, 2010, p. 155). TPB attempts to predict behavior from subjective norms, attitudes, perceived behavioral control, and intentions (Anderson et al., 2013). TPB has been studied extensively in a variety of fields, including health (Godin & Kok, 1996), pro-environmental behavior (Ho et al., 2015), and philanthropy (Kinnally & Brinkerhoff, 2013). The theory is shown in Figure 2.

Figure 2. Theory of Planned Behavior (TPB) Model (Ajzen, 2000)

Competing Models Review

Social Cognitive Theory (SCT)

Social cognitive theory (SCT) is one of the most frequently applied theories of health and behavior. It posits a reciprocal deterministic relationships between individuals, his or her environment, and behavior (Compeau et al., 1999). They interact with and upon one another to form the basis for the behavior (Bandura, 1986, 1997; Venkatesh et al., 2003). The central tenet of SCT is the concept of self-efficacy. Behaviors are determined by the interaction of outcome expectations and efficacy expectations. SCT has been widely used in different research domains, including computer and Internet utilization (Ratten, 2011). In spite of that, SCT has been critiqued for the high percentage (68%) of unexplained variance in users’ behavior (Compeau & Higgins, 1995), being too comprehensive in formulation, making for difficulty in operationalizing and evaluating the theory in its entirety (Abbasi, 2011).   

Task-technology Fit (TTF) Model and Task Acceptance Model (TAM)

Task-technology Fit (TTF) was proposed by Goodhue and Thompson in 1995. They proposed the model to explain the utilization of technology by examining the fit of technology to users’ tasks or requirements. According to Goodhue and Thompson (1995), the literature on information system management domain was characterized by two streams of research, namely focusing on technology utilization and TTF. The research on technology utilization mostly examined the relationship between attitude, beliefs, their associated factors and the use of information technologies. One stream of research reflected the focus on technology performance and TTF. Technology acceptance model (TAM) originated from the theory of reasoned action along with TPB and postulated that the use of technology raises cognitive evaluation in the form of perceived ease of use and perceived usefulness (Davis, 1986). The TTF theory and TAM have a number of limitations such as complexity of the models, weak predictive power, and lack of focus on situational and personal factors (Lee, 2009; Lee et al., 2003; Taylor & Todd, 1995b). The theory is illustrated in Figure 3.

Figure 3. Extending TAM with TTF Constructs (Dishaw & Strong, 1998)

Innovation Diffusion Theory (IDT)

Innovation diffusion theory (IDT) was developed by Rogers in 1962. It explains how an idea or product diffuses through a specific population or social system (Rogers, 2003). There are four elements in the innovation diffusion model which need more attention: Innovation, communication channels, time and the social system (Estabrooks et al., 2006; Rogers, 2003). Researchers in the field of behavioral change have found that people who adopt an innovation early have different characteristics from those who adopt innovation later (Rogers, 2003). Roger’s diffusion of innovations model has been applied for over 5000 studies. It has been well-accepted and used in agricultural extension, certainly up to the 1980s when the transfer of technology approach was still prevalent (Haidar & Kreps, 2004). However, even with this success, there have been a number of criticisms. Pro-innovation bias, individual-blame bias, recall problem, and the issue of equality are the four main categories of criticisms (El Malouf & Bahemia, 2025). Rogers himself believed that “the progress of a scientific field is helped by realization of its own assumptions, biases, and weaknesses” (Rogers, 2003, p. 106). IDT model is illustrated in Figure 4.

Figure 4. Rogers’ (2003) Innovation Diffusion Theory

The Model of Personal Computer Utilization (MPCU)

The personal computer utilization model was derived from Triandis’ (1977) theory of interpersonal behavior (Khater, 2016). Thompson et al. (1991) developed MPCU to predict personal computer utilization behavior. The nature of this theoretical model is suited to predict individual acceptance and use of several information technologies (Thompson et al., 1991). An important difference between the TPB and MPCU is that the former does not use behavioral intentions as a key variable. Teacher’s intention is the pivotal variable in our study. Thus, for our purpose, the use of this model is invalid. 

The Motivation Model

The motivation model was developed by Davis et al. (1992), who applied this theory to investigate information technology adoption and use. The main premise of this theory is that there are extrinsic and intrinsic motivations that influence the behavior of users (Davis et al., 1992). Extrinsic motivation involves concepts like perceived usefulness, perceived ease of use, and subjective norm (Sharma & Mishra, 2014). The feeling of pressure someone gets from playing with a computer is an example of intrinsic motivation (Davis et al., 1992; Venkatesh & Davis, 2000).

Although this model has been found useful in understanding new technologies acceptance and use (Igbaria et al., 1996), it just explains a low percentage of the variance in question compared with TPB (Davis et al., 1992; Igbaria et al., 1996).

Sufficiency of the TPB

A number of researchers have conducted meta-analyses to assess the theoretical sufficiency of the TPB (Armitage & Conner, 2001; Godin & Kok, 1996; Hagger & Hamilton 2023; Lee & Vincent, 2021; Notani, 1998; Park & Shin 2021; Riebl et al., 2015). These scholars mainly focused on TPB. Other scholars assessed the theory of reasoned action (Hagger et al., 2002; Hausenblas et al., 1997; Schulze & Wittmann, 2003; Sutton, 1998). Most of these meta-analyses demonstrate robust support for the TPB.  In 1996, Godin and Kok (1996) conducted a meta-analysis, which also supports the TPB. These scholars looked at 56 studies that used the model to study health-related behavior. They found TPB did a good job in explaining intention. A similar support for the model was found in Notani’s (1998) meta-analysis. The findings indicated that perceived behavioral control is a stronger predictor of behavior when it is operationalized as a global measure and is conceptualized to reflect control over factors primarily internal to the individual.

In addition to meta-analyses, a number of scholars have compared competing models and tried to determine the theoretical sufficiency of the TPB. Taylor and Todd’s (1995a) work is one of those scholarly contributions. They completed a study which is widely cited in the literature. They compared TPB, TAM, and the theory of reasoned action in a study involving information technology use. The result of the study was that the decomposed model of planned behavior provided a fuller understanding of behavioral intention.

In brief, these meta-analysis and comparative studies demonstrate the theoretical sufficiency of TPB with the majority of these studies explaining health-related applications, such as exercise and physical activity.

In short, in comparison with the above-mentioned models, the TPB offers a more parsimonious, simple, understandable, and general framework for examining teachers’ intentional adoption of computer-mediated English course. 

Technology Adoption in Computer-Mediated Teaching

TPB has received substantial support in a large number of empirical investigations (Fishbein & Ajzen, 2010). According to Ajzen (2020), this theory has been cited over 250,000 times in the diverse fields of psychology, consumer behavior, education, work and leisure, environmental psychology and political behavior, which scrutinize TPB. Some of these studies have used the quantitative research methods and deal with technological innovations (Hsu et al., 2006; Morris & Venkatesh, 2000; Troung, 2009), and some deal with various aspects of online teaching (Amirkhani et al., 2025; Irani & O’Mally, 1998; Lee, 2010, Pooreh & Hosseini Nodeh, 2015; Rakhshani, 2024). As these studies have been successful, the utility of TPB for the research involving computer-mediated courses is reinforced.

PURPOSE OF THE STUDY

The purpose of the study was to examine the extent to which TPB explains high school English teachers’ intentions to adopt computer-mediated teaching. In this study, Ajzen’s TPB has been used because it provides a solid framework for studying high school English teachers’ intentions of adopting computer-mediated teaching. The theory has been successfully applied to studies of different relations in various human domains and has received substantial support from the social scientists.

This study provides another opportunity to test the model and apply it to a new domain. Thus, it can potentially advance scholarly understanding of this domain and at the same time offers an opportunity to test the theoretical sufficiency of the model. To achieve the objectives of this study, the following research questions (RQs) were formulated:  

• What is the relationship between teachers’ attitude, subjective norm, perceived behavioral control and their intentions to adopt computer-mediated teaching?
• What is the goodness of fit for the Theory of Planned Behavior model applied to English teachers’ intentions to adopt computer-mediated teaching?
• What is the relationship between the teachers’ age, gender, years of English teaching and their intentions to adopt computer-mediated teaching?

METHOD

This study was conducted in two stages. The first stage dealt with the development of a questionnaire to accurately measure the constructs in Ajzen’s theory. In the second stage, confirmatory factor analysis was used to determine the composite reliability and construct validity of the final survey questionnaire and evaluate model fit indices.

Participants

The researcher targeted all of the 700 high school English teachers subscribed to Iranian social network (Shad) as well as Telegram platforms. Four hundred forty teachers voluntarily participated in the survey. This is what is recently called network sampling method. The participants consisted of 221 male and 219 female high school English teachers. The minimum and maximum of their age were 28 and 55, respectively. Years of experience in teaching English ranged from three to twenty-nine, and their education degrees completed by respondents spanned from bachelor’s degree to Ph.D.

Table 1. Descriptive Statistics of the Participants

Design

For the purpose of this study, a survey method as the most conventional available to the researchers was used (Farhady, 2022). More precisely, this is a cross-sectional quantitative survey study.

Instrumentation

To collect data from participants “there is no single reasoned action questionnaire” (Fishbein & Ajzen, 2010, p. 456).  Thus, the researchers developed a questionnaire based on the guidelines provided by Ajzen (2006).  The researchers began with the focus groups. Through convenience sampling, 15 high school English teachers (five males and ten females) were chosen as participants in three focus groups to identify the most salient behavioral, normative and control factors about computer-mediated teaching. The findings resulting from the focus groups helped shape the pilot test, which was sent to 30 high school English teachers from around the country representing the target population. Before implementation, five experts in language teaching were requested to study the final survey questionnaire. This resulted in a shorter questionnaire of high quality. Then, the researchers examined the reliability of the Likert-scale to develop the final self-administered questionnaire.

Data Collection Procedure

The data collection process was launched in fall 2024 and took six months. The final survey was conducted through the Iranian social network (Shad) as well as Telegram platforms on high school English teachers’ channels. The participants were informed about the aims of the study, ensured about anonymity, and were requested to voluntarily fill out the questionnaire which was accessible in Google Forms format link. At the time this study was being conducted, there were approximately 700 high school English teachers subscribed to the above-mentioned channels. Out of the 700 eligible high school teachers, 440 returned the self-administered questionnaire. In the final survey, all constructs of Ajzen’s model had an acceptable alfa values. Analysis study was based on the recommendations of Francis et al. (2004). Recoding was done in a way that the higher numbers reflected a positive attitude, great social pressure and great level of control.

Data Analysis

The data for this study were obtained from two sections of the questionnaire: (a) demographic information section and (b) thirty-five 7-point bipolar Likert-scale items section. Recoding ranged from +1 to +7. It was in a way that higher numbers reflected a positive attitude, a higher level of social pressure and a higher level of control. All the data were analyzed using IBM SPSS Statistics (Version 27) and IBM SPSS Amos (Version 260. Before the path analysis was run, the assumptions of the test (outliers, normality, linearity, homoscedasticity, multicollinearity, and independence of errors) were checked in SPSS. Confirmatory factor analysis was performed in Amos to assess the validity and reliability of the final survey questionnaire and evaluate the goodness of fit indices of the Ajzen’s model.

RESULTS

Table 2 demonstrates the descriptive analysis of the data.

Table 2. Descriptive Analysis of the Items and Constructs

Table 3 below indicates the validity and reliability of the questionnaire.

Table 3. Validity Analysis

Note. Significance of correlations: P< 0.100, * P< 0.050, ** P< 0.010, *** P< 0.001, Criteria for convergent validity: CR>.7, AVE>.5, CR>AVE, Criteria for divergent validity: AVE>MSV, AVE>ASV

Figure 5 below shows the convergent and divergent validity of the questionnaire.

Figure 5. Validity Analysis

As Table 3 and Figure 5 show, the reliability and validity of the questionnaire is acceptable. We move on to test the hypotheses in the structural model.

RQ1: What is the relationship between teachers’ attitude, subjective norm, perceive behavioral control and their intentions to adopt computer-mediated teaching?

Figure 6. The Path Coefficients of the Structural Model

The above Path diagram clearly shows that 44 percent of the intent variance can be explained by attitude, subjective norm and PBC.

Table 4. Pearson’s linear correlation of TPB Constructs

As shown in Figure 6 and Table 4, there is a significant correlation between the teachers’ attitude (β= .34, P< .000), subjective norm (β = .50, P < .000), perceived behavioral control (β= -.18, P < 0.05) and their intentions to adopt computer-mediated teaching.

RQ2: What is the goodness of fit for the TPB model applied to English teachers’ intentions to adopt computer-mediated teaching?

Table 5. Model Fit Measures

Table 5 indicates the goodness of fit indices for the model. All the indices are inadequate, implicating a poor model fit for predicting the teachers’ intentions to adopt computer-mediated teaching.

 RQ3: What is the relationship between the teachers’ age, gender, years of English teaching and their intentions to adopt computer-mediated teaching?

Table 6. Correlation of Intent with Demographic Variables

**Correlation is significant at the 0.01 level (2-tailed).

*Correlation is significant at the 0.05 level (2-tailed).

As Table 6 shows, there is no significant correlation between the teachers’ age, gender, years of English teaching and their intentions to adopt computer-mediated teaching.

DISCUSSION

This research study was conducted to explore the determinants of the adoption of computer-mediated courses by high school English teachers in Iran. Drawing on the TPB, the researcher proposed a conceptual model assuming a positive association between the intention of teachers and their attitudes, subjective norms, and perceived behavioral control. The researcher conducted a survey for collecting primary data to empirically test the model throughout Iran and found support for her hypotheses. The results have revealed that when a teacher has a favorable attitude toward the course, they have a higher intention to develop it, which later becomes the reason for their adoption of the computer-mediated course. Subjective norm is another determinant. This finding shows that when their significant others’ opinions are also favorable toward such courses, it becomes a psychological reason for their adoption intention.    

In the educational domain as in the other diverse fields, the influence of perceived behavioral control has substantively gained support (Akour, 2021; Al Breiki et al., 2023; Garcia & Oducado, 2021). The existence of significant relationship between perceived behavioral control and intention in the present study gives further support to the studies mentioned before.

When a path is drawn in Amos diagram software from behavioral beliefs to sub-norm, the total variance explained goes up from 0.44 to 0.47. This implies what a person believes has a significant influence on the perception of expectations by important others, which conceptually does not make sense. A path drawn from subjective norm to attitude increases the total variance from 0.44 to 0.49. It would mean that a person’s attitude is affected by others’ expectations of them to adopt computer-mediated courses, which makes sense. This finding corroborated the findings of Knabe (2012).

Notwithstanding all the features aligned with previous studies, some anomalies were present in the results. First, there was a poor model fit. This could be due to contextual limitations or cultural differences. Some unknown features such as institutional constraints may be at work. This offers a new venue for further future investigations. Second, the negative PBC coefficient may be interpreted as the lack of interest to do the behavior when there is too much familiarity or frequent repetitions of the behavior over a long period of time so that the behavior does not offer a new challenge or excitement.  

CONCLUSION AND IMPLICATIONS

In sum, all the constructs of the Ajzen’s theory have significant predictive power in shaping the intention of high school English teachers to adopt and develop computer-mediated courses. According to Fishbein and Ajzen (2010), the origin of behaviors lies in beliefs. Beliefs are defined as the subjective probabilities that performing a behavior leads to certain outcomes. Once beliefs related to a particular behavior have been formed, they provide the basis for attitudes, subjective norms, and perception of control, which, in turn, leads to intentions and actions. With regard to changing interventions, TPB can guide researchers by identifying the behavioral, normative and control beliefs to target in the intervention. In-service teaching programs can help teachers to change their intentions by targeting those beliefs. Bringing about the desired changes in the teachers’ intentions can make learning more interactive, tailor learning materials and activities to individual student’s needs and learning styles, enable students for real-time communication between students and teachers, and facilitate quick feedback and support.    

This study carries important practical implications for principals, and other educational authorities involved. They can launch in-service programs to positively enhance the teachers’ attitude, subjective norm and perceived behavioral control in favor of the adoption of computer-mediated English courses.

In spite of all the above-mentioned strengths and implications, this study should be seen in light of its limitations. For example, the data were cross-sectional, so future researchers may conduct experimental or longitudinal study to validate the model. All our respondents reside in Iran; hence, the generalizability of our finding is limited. Future research studies can thus be conducted in other countries.       

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID