Background and Objectives: This study's primary objective was to employ canonical analysis to examine the relationship between behavioral, normative, and control beliefs and their influence on the propensity to share knowledge with attitudes, subjective norms, behavioral intentions, and perceived behavioral control among middle school teachers in the city of Karaj.
Methodology: This study adopted a descriptive correlational design. The statistical population for this research comprised all secondary school teachers in the city of Karaj. To this end, a sample of 350 individuals was randomly selected from this population using a simple random sampling technique.  The Knowledge Sharing Questionnaire developed by Chennamaneni and Raja (2012) and the Reasoned Action Questionnaire by Fishbein and Ajzen (2002) were used to evaluate the variables. Canonical correlation analysis served as the primary statistical technique for the data analysis.
Findings: Following the collection of research data and subsequent analysis using canonical correlation analysis, a significant linear relationship, with a 5% significance level, was observed between all variables examined. These findings supported the research hypothesis by demonstrating the predictability and significance of the relationship between behavioral, normative, and control beliefs and the propensity to share knowledge with attitudes, subjective norms, behavioral intentions, and perceived behavioral control. Among the criterion variables, behavioral intention (0.67) emerged as the most influential, whereas normative beliefs (0.41) indicated the highest significance among the predictor variables. The analysis of Wilks' lambda value (p < 0.001, F = 18.868, λ = 0.461) revealed a significant canonical correlation between the two sets of variables, with a 99% probability. However, it should be noted that Wilks' lambda represented the unexplained variance in the model. In this analysis, the effect size of the two canonical functions was 0.539, representing the shared variance between the two sets of variables the model could explain. With an effect size of 0.539, the obtained model suggested the existence of a common source of variance that explains the overlap between the two variable sets. Furthermore, the results of the F-test indicated that only one of the canonical functions was statistically significant (F = 18.868, p < 0.001, λ = 0.461). The second function (F = 0.503, p > 0.873, λ = 0.987), third (F = 0.456, p > 0.768, λ = 0.955), and fourth function (F = 0.072, p > 0.789, λ = 0.999) were not statistically significant. This study considered and calculated four canonical dimensions, but only one demonstrated statistical significance.
Discussion: This study found a significant canonical correlation and a substantial overlap (53%) between behavioral, normative, and control beliefs and the propensity to share knowledge with attitudes, subjective norms, behavioral intentions, and perceived behavioral control. These findings indicated that attitudes, subjective norms, behavioral intentions, and perceived behavioral control could be predicted and have a significant relationship based on behavioral, normative, and control beliefs and the propensity to share knowledge. Specifically, In the first canonical function, all four predictor variables (behavioral, normative, and control beliefs, as well as the propensity to share knowledge) were identified as significant contributors to the relationship. These variables were more predictive of attitudes, subjective norms, behavioral intentions, and perceived behavioral control than those in the second function. The study's findings suggested that teachers' knowledge-sharing behavior was influenced by their propensity to share knowledge. This propensity was influenced more by teachers' normative and control beliefs than by their behavioral beliefs and attitudes. Drawing upon the results of this study, several key individual motivations for teachers to engage in knowledge-sharing with their colleagues can be identified. These motivations include the enjoyment of helping others, group participation and collaboration, personal responsibilities, encouragement, and rewards.  Considering the influence of individual factors on the formation of behavioral beliefs, it can be inferred that the positive or negative perception of knowledge sharing (behavioral beliefs) among teachers is influenced by several key factors. These variables include their enjoyment of helping others, their understanding of the benefits associated with knowledge sharing, and their perception of the difficulty or ease associated with knowledge sharing. When considering the role of behavioral, normative, and control beliefs in determining the propensity to share knowledge, it can be concluded that teachers who hold positive behavioral beliefs about knowledge-sharing are more likely to engage in knowledge-sharing activities. Furthermore, teachers who demonstrated significant self-control were more likely to share their knowledge. Moreover, the research findings indicated a positive correlation between teacher attitudes and knowledge sharing.