被忽略的負向動機構念：以期望──價值理論發展臺灣國中生成本量表與信、效度檢驗

The Neglected Negative Motivational Construct: Development and Validation of a Cost Scale for Taiwanese Junior High School Students Based on Expectancy-Value Theory

儘管期望—價值理論是解釋學習者動機的重要框架之一，成本概念在過去的研究中卻相對被忽略。成本——作為一個多面向的負向動機構念，能有效解釋學習者的逃避動機及其負向的學習行為與學業成就。基於此，本研究編製一套特定於臺灣國中生數學領域的成本量表，並對其信度與效度進行考驗。研究以679位國中生為樣本，透過結構方程模式進行資料分析，結果顯示：一階四因素的成本模式具有良好的適配度、幅合效度及區別效度，顯示該量表具有良好的信度與效度，能有效解釋國中生的成本信念。其次，成本能顯著預測動機涉入及學業成就，但不同成本對學業結果的影響存在差異。具體而言，情緒成本顯著預測非適應性動機涉入，並負向預測適應性動機涉入及學業成就，最能代表成本此一構念的負向動機特質。機會成本能顯著預測非適應性動機涉入，而自我成本則負向預測部分非適應性動機涉入（即逃避求助），但卻正向預測學業成就。與本研究預期相反的是，努力成本正向預測適應性動機涉入，負向預測非適應性動機涉入，但對學業成就無顯著預測效果。本研究根據結果進行討論，強調了不同成本在教育實踐中的應用性，並進一步針對未來研究提出建議。

Motivation in learning has long been a critical issue in educational psychology, as it directly influences students’academic achievement and learning behaviors (Lin, 2022; Schunk et al., 2008; Shih, 2006). Among the various theoretical frameworks in motivation research, the contemporary expectancy-value theory (EVT) has emerged as one of the most influential models, as it explains the extent of learners’motivation and their academic-related choices and behaviors. EVT posits that learners’motivation depends on their expectations of success and their subjective valuation of a given task (Eccles et al., 1983). When learners perceive a task as highly important and valuable and believe in their ability to succeed, they are more likely to engage in learning and achieve higher academic performance (Wigfield & Eccles, 2000).
Although the constructs of expectancy and value have been extensively examined in numerous studies and have been established as key predictors of academic outcomes, the construct of cost has received relatively less attention. Cost is a multidimensional concept that describes the sacrifices learners must make to complete a learning task, such as time, effort, and psychological pressure (Eccles & Wigfield, 2002). In recent years, researchers have increasingly recognized that cost, as a negative motivational construct, plays a crucial role in explaining avoidance motivation and maladaptive learning behaviors and outcomes (Jiang et al., 2018).
Given that previous empirical studies have primarily focused on the effects of expectancy and value on learning behaviors, research on cost remains limited. This has led to an insufficient understanding of avoidance behaviors and negative motivation among learners. Moreover, conflating cost with other values (e.g., intrinsic, attainment, and utility value) may obscure its unique influence on learning behaviors, thereby preventing educators from effectively addressing students’challenges and resistance in learning (Wigfield et al., 2017). Furthermore, most existing cost measurement instruments have been developed in Western cultural contexts, and research on cost beliefs in Taiwan remains scarce. Only a few studies (e.g., Goh, 2021; Huang, 2024) have adopted measurement tools developed abroad, and the reliability and validity of cost scales have yet to be fully examined in the local context.
To address this research gap, the present study develops a cost belief scale specifically for junior high school students in Taiwan based on Jiang’s (2015) cost classification framework. This scale aims to more accurately capture students’motivational states and provide valuable insights for future research and educational practices.
In summary, this study focuses on the mathematics domain among junior high school students in Taiwan, aiming to develop a reliable and valid cost belief scale and examine the impact of different types of cost on students’motivational engagement and academic achievement. It is expected that this scale will effectively measure the multidimensional nature of cost beliefs in junior high school students and further explore the effects of various cost factors on learning processes and outcomes. In essence, the development of this cost belief scale contributes to future research and educational practices by enhancing educators’understanding of students’perceived costs in learning tasks and their implications, ultimately leading to the design of effective instructional strategies and support systems.
The contemporary expectancy-value theory (EVT) is a theoretical framework for explaining achievement motivation, originally proposed by Eccles et al. (1983) to account for individuals’effort, choices, and outcomes in achievement-related activities. According to EVT, achievement motivation is composed of an individual’s expectancy for success in a given task and their evaluation of the task’s value. Students’success expectancies and task value assessments influence their achievement and decision-making behaviors through either additive or interactive mechanisms (Eccles, 2009).
Expectancy refers to individuals’beliefs about their abilities in a particular task or domain, addressing the question,“Can I do this”? Eccles et al. (1983) conceptualized expectancy as comprising both expectancies for success and ability beliefs. On the other hand, value describes individuals’assessment of the importance of a particular task or activity, addressing the question,“Do I want to do this”? Eccles et al. identified value as a multidimensional construct encompassing four types: Intrinsic value, attainment value, utility value and cost. Furthermore, a substantial body of empirical research has demonstrated that expectancy and value positively predict various adaptive outcomes, including students’course selection, effort investment, persistence in learning, and academic achievement (e.g., Eccles & Wigfield, 2002, 2020; Part et al., 2020; Peng, 2021; Wigfield & Cambria, 2010).
Eccles et al. (1983) introduced the concept of“cost”in EVT, defining it as a negative motivational construct that describes the adverse consequences associated with engaging in a task or activity. As the theory evolved, cost was classified as a component of task value (Eccles, 2005; Wigfield & Eccles, 1992), and subsequent studies (e.g., Conley, 2012; Meyer et al., 2019; Trautwein et al., 2012; Wu & Fan, 2017) have largely followed this perspective. However, many researchers have continued to exclude the cost construct when assessing task value, relying solely on expectancy and the three positive value components to explain individuals’learning behaviors and outcomes (e.g., Guo et al., 2015; Jacobs et al., 2002; Kosovich et al., 2017; Nuutila et al., 2018; Putwain et al., 2019). Alternatively, some studies have aggregated cost with the other three value components to obtain an overall task value score (e.g., Buehl & Alexander, 2005). These approaches limit the explanatory power of EVT in fully capturing students’adaptive and maladaptive engagement in learning tasks, as they fail to account for the unique influence of cost on achievement behaviors and outcomes (Jiang et al., 2018).
With the increasing research on cost, Jiang’s (2015) four-category framework has been widely applied in studies conducted in East Asian countries. For example, Lee and Song (2022) explored the effort, opportunity, and emotional costs experienced by Korean students in Massive Open Online Courses (MOOCs). Similarly, Chung and Kim (2022) investigated the emotional and effort costs of female middle school students in mathematics, as well as the relationship between these costs and other motivational and learning outcomes. In Taiwan, Goh (2021) and Huang (2024) revised the Chinese version of the cost scale based on Jiang’s four-category framework and applied it to the measurement of mathematics and English subjects. Wu’s study found that a first-order four-factor model best fit the observed data, indicating that cost can be clearly categorized into four types. Huang’s study further supported the classification of these four types of costs into two broader categories: Effort cost and sacrifice cost (including opportunity, self, and emotional costs).
In summary, this study adopts Jiang’s (2015) theoretical framework for cost classification, which was developed with consideration of the specific cultural context of East Asian students. This framework explicitly incorporates self-cost as a crucial component, providing a clear and culturally appropriate theoretical foundation for explaining the learning motivation of East Asian students. Furthermore, studies by Flake et al. (2015), Perez et al. (2014), Jiang et al. (2018), and Huang (2020) have further expanded the measurement and application of the cost construct, providing empirical support and reference for the development of the present study’s measurement tool. Accordingly, this study builds upon Jiang’s cost classification framework, integrating previously well-developed measurement instruments to develop a cost scale specifically for Taiwanese junior high school students, allowing for a more effective assessment of students’perceived cost beliefs in the learning process.
The objectives of this study are as follows: To develop a cost scale comprising four dimensions, assess its reliability and validity, and examine the current status of junior high school students’perceived cost beliefs. To explore the relationship between the four types of costs and adaptive/non-adaptive motivational engagement, as well as academic achievement.
Data analysis was conducted using IBM SPSS Statistics 25 and LISREL 8.50. To address the first research objective, exploratory factor analysis (EFA), internal consistency analysis, confirmatory factor analysis (CFA), discriminant validity, and convergent validity were used to examine the reliability and validity of the cost scale. For the second research objective, latent variable regression analysis was employed to test the predictive effects of the four types of costs on adaptive motivational engagement, non-adaptive motivational engagement, and academic achievement.
Using a sample of 679 students, structural equation modeling was employed to analyze the data. The findings revealed that the proposed four-factor cost model exhibited good model fit, convergent validity, and discriminant validity, supporting the scale’s reliability and validity in capturing students’cost beliefs. Furthermore, cost significantly predicted motivational engagement and academic achievement, with notable differences in the effects of various cost types. Specifically, emotional cost strongly predicted maladaptive motivational engagement, and negatively predicted adaptive motivational engagement and academic achievement, underscoring its negative motivational characteristics. Opportunity cost was found to significantly predict maladaptive motivational engagement, while ego cost negatively predicted certain maladaptive behaviors (e.g., help-seeking avoidance) but positively predicted academic achievement. Contrary to expectations, effort cost positively predicted adaptive motivational engagement, negatively predicted maladaptive motivational engagement, and had no significant effect on academic achievement. These findings are discussed in relation to their implications for educational practice, with suggestions for future research.