Motivation to learn and to succeed: A path analysis of the CANE model of cognitive motivation

Commitment and Necessary Effort (CANE) model of motivation, a modification of M. Ford's Motivation Systems Theory (MST), posits two stages in motivated behavior. In stage one, positive or strong levels of self-efficacy, emotions, and values interact multiplicatively to predict goal commitment (or persistence). In stage two, self-efficacy and persistence predict the exertion of mental effort (the use of non-automatic elaborations). Together, persistence and mental effort are hypothesized to influence achievement.; 343 urban high school students representing a variety of ability, ethnicity, and program groups were surveyed on their perceptions of self-efficacy, emotionality, task value, persistence, and mental effort relative to their performance on a college-level laboratory assignment in their biology or chemistry class. Responses were analyzed using traditional multivariate statistical techniques and a structural equation modeling path analysis.; Self-efficacy was found to be highly correlated with persistence. Emotional state (mood) correlated significantly with persistence as well, but emotional trait (personality) did not. Values correlated significantly with persistence with the stronger values correlating more strongly than the weaker ones (i.e., importance correlated more strongly than interest, and interest than utility). The path analysis showed that general science self-efficacy (as compared to the lab-specific self-efficacy) had the strongest relationship with persistence, both through its direct effect and its indirect effects through specific self-efficacy, importance, interest, and utility. Emotionality was found to be a non-factor in the path analysis.; The path analysis revealed no significant contributing factor to mental effort. The inverted-U nature of this factor may preclude any significant correlation with motivational variables such as self-efficacy and persistence. It is concluded that better instruments need to be developed to measure students' use of non-automatic elaborations.; Overall fit of the CANE model was good (CFI = .908). The model's predictive and explanatory validity can be strengthened by refining the measure of task-related emotionality, improving the measure of mental effort, and in having a more uniform, broad-based measure of task achievement.