Understanding college students’ achievement goals toward using open educational resources from the perspective of expectancy–value theory

Abstract

Evidence that open educational resources (OER) can decrease college students’ educational cost without harm to their course performance on different subjects has been well documented, but student motivation to use OER for learning is underexplored. This study investigated college students’ achievement goals of using OER from the perspective of expectancy–value theory. We recorded the survey responses of 246 college students in an education course at a public university in the southeast of the United States of America for analysis. We established a structural equation model to investigate the relationship between their task value beliefs in using OER and their course achievement goals. The findings show that the students’ perceived usefulness of OER positively predicted their mastery-approach goals but negatively predicted their performance-approach goals. In addition, their self-estimated cost of learning with OER predicted their mastery-avoidance goals. We discuss practical implications for facilitating college students’ motivation toward using OER.

Keywords:

• open educational resources (OER)
• achievement goals
• expectancy–value theory
• task value belief
• college students

Introduction

To date, college students have incurred increasing expenditure on their tuition and educational materials. In parallel with the inflation nationwide, a high educational cost prevents many college-aged students from attending higher education. Student loans may alleviate those students’ financial burdens to afford college tuition, but the high cost of textbooks and other supplemental learning materials can considerably decrease student learning effectiveness (Read et al., 2020).

Open educational resources (OER) can be a low-cost alternative to traditional textbooks (Hilton, 2020; Wiley & Hilton, 2018). OER are freely available online resources that are openly licensed and can be retained, reused, revised, remixed, and redistributed to meet personalized needs (Freeman et al., 2022; Lin & Tang, 2017). Educators can use OER to develop personalized instruction and assessment that caters to student interests, leading to students’ heightened engagement and improved performance in a course (Koseoglu & Bozkurt, 2018; Tang & Bao, 2021, 2023). In addition to a lower cost, the use of OER has no harm to student learning achievement such as course scores and passing rates (Hilton, 2020; Read et al., 2020). The rise of OER has been envisioned as an opportunity for broadening college student access to high-quality educational resources and improving their learning experience at low costs (Hilton, 2016, 2020).

To this end, understanding students’ achievement goals in using OER is critical for instructors to tailor OER to student goals and support their achievement in OER-based courses. Achievement goals are defined as the psychological motives to “develop, attain, or demonstrate competence at an activity” (Harackiewicz & Elliot, 1993, p. 904). In education settings, achievement goals is one of the most important predictors of students’ intrinsic motivation, choices, and academic achievement (Elliot & Thrash, 2001; Harackiewicz & Elliot, 1993; Rawsthorne & Elliot, 1999). Achievement goal theory postulates two dimensions of individual goals toward achievement behaviors: competence (mastery versus performance) and valence (e.g., approach versus avoidance), beyond a traditional mastery-performance dichotomy (Elliot & McGregor, 2001). Research has indicated that achievement goals with negative valence of competence (e.g., avoidance goals) are usually associated with low academic outcomes (Rawsthorne & Elliot, 1999; Sun & Xie, 2020; Van Yperen et al., 2009). To facilitate student acceptance of OER and their academic performance in OER-based courses, understanding the factors that may affect students’ perceived competence in and value of the use of OER is thus needed. However, empirical evidence about students’ achievement goals with regard to OER is scarce, let alone any mention of strategies that may facilitate their achievement goals.

As a preliminary effort to close this gap, this study was thus intended to investigate college students’ achievement goals in using OER through the lens of expectancy–value theory (EVT; Eccles & Wigfield, 1995). EVT suggests that achievement behaviors hinge upon the interplay between one’s beliefs about their competence and perceived value of the tasks. Particularly, task value belief, a construct that describes individuals’ perceived importance and usefulness of a task for their goals, is a predictor of individual intention of performing the task as well as their subsequent task performance (Eccles & Wigfield, 1995). This study thus focused on students’ task value beliefs and then identified the relationship between students’ task value beliefs and their achievement goals toward using OER. By entangling the complex interrelations between those factors, this research aimed to help scholars and educators investing in OER and educational equity to efficiently integrate OER in higher education.

Literature review

OER

OER can be broadly defined as openly licensed educational resources that users have open and free access to utilize, manipulate, and share for individualized purposes (United Nations Educational, Scientific, and Cultural Organization, 2019). OER include a wide range of online and print materials such as open textbooks (e.g., OpenStax), open courseware (e.g., massive open online courses, MIT OpenCourseWare), open repositories (e.g., Merlot, CK-12, EngageNY), open-access journals (e.g., PLOS ONE, arXiv), and even open-licensed videos on YouTube or Khan Academy. Generally, those resources follow Creative Commons licenses that grant users with the permission to retain, reuse, revise, remix, and redistribute them for customizing and reproducing a broader collection of materials (Hilton, 2016, 2020; Lin & Tang, 2017). In particular, retain means users can download and keep a copy of the resources (Tang & Bao, 2020, 2021). Reuse gives users permission to utilize the whole or a portion of OER (Tang, 2021). To satisfy users’ personalized needs, OER also allow users to make changes to the existing OER (revise) or merge any existing OER (remix) (Tang et al., 2020, 2021). Then redistribution allows users to share and disseminate these resources without any restrictions of copyrights.

OER can allow students to maintain a similar level of learning effectiveness at a lower cost than using traditional textbooks. For example, Clinton (2018) found college students using OpenStax textbooks in an introductory psychology course in an American public university performed slightly better than those using traditional textbooks at a lower cost. Grewe and Davis (2017) indicated the use of OER is positively correlated with students’ final grades in an online history course. Ross et al. (2018) also noted no significant difference in the average grade existed between two offerings of a college-level introductory sociology course, one of which used OpenStax textbooks while the other taught with commercial counterparts. In addition, Tillinghast et al. (2020) integrated student engagement as a critical aspect of effectively adopting OER. Research has shown that the use of OER in higher education settings improves student engagement and affords student collaborations. Cooney (2017) found that students perceived a higher level of course engagement using OER in a health psychology course. Kimmel et al. (2022) described that the ease of accessing and using OER promoted student engagement with the content and interactions with the peers in a community college. Trust et al. (2022) applied project-based learning approach to engage students in OER-based projects, which significantly increased student engagement with course materials.

Despite the potential of OER for decreasing college students’ spending for textbook purchases, a majority of college students have not been aware of OER or used OER for learning (Read et al., 2020). Research on student use and perception of OER focuses mainly on measures of learner behavior and performance, but whether students are motivated by this low-cost alternative to traditional textbooks remains unknown. In addition, in OER-integrated learning environment such as MOOCs, students tend to struggle with maintaining their engagement at the same level as the course starts (B. Yang et al., 2022). In an open learning setting, intrinsic motivation matters for students to consistently uphold their course participation and exhibit exceptional performance (Tang et al., 2018, 2019). To increase the impact of OER on student learning, there is a call for research tapping into psychological perspectives of learners’ motivation to use OER and antecedents of motivations.

Achievement goals

Achievement goals account for intentional behaviors conducted by individuals in achievement-related situations, and such goals impact their behavior and performance (Bardach et al., 2020; Rawsthorne & Elliot, 1999). There has been a wealth of research on the topic of achievement goals. The origin of achievement goals can be traced back to a dichotomous model in 1980s that primarily referred to two types of goals related to competence: mastery goals and performance goals (Dweck, 1986; Nicholls, 1984). Specifically, mastery goals focus on developing competence for personal growth or development, while performance goals address exhibiting competence in comparison to others (Elliot & McGregor, 2001; Liu, 2021). Individuals with mastery goals view success as obtaining desired knowledge and/or skills and thus they tend to develop intrinsic motivation toward learning, maintain a positive mindset focused on growth, and invest considerable effort in the attainment of relevant competence (Bong, 2001; Dweck, 2013). In contrast, individuals with performance goals consider success in terms of whether they can outperform others so they are likely to divert their focus from achieving proficiency to comparing their performance with others or standards (Elliot & McGregor, 2001; Elliot & Thrash, 2001). Accordingly, performance goals have been linked with less adaptive learning outcomes such as decreased motivation and impaired performance (Dweck, 1986; Elliot, 1999).

Beyond the dichotomous model, the valence of competence was then added as an additional dimension to formulate a 2 × 2 framework of achievement goals (Elliot, 1999; Elliot & McGregor, 2001). The valence of competence, referring to the degree to which one values the mastery of such competence (Elliot & McGregor, 2001), categorizes achievement goals as approach or avoidance goals in terms of whether the outcome is desirable or not. Individuals with approach goals are driven by seeking out desirable outcomes such as the mastery of competence or the superiority in performance relative to others, whereas those with avoidance goals care about preventing undesirable outcomes such as avoiding failures or errors (Van Yperen et al., 2009). Accordingly, the 2 × 2 framework of achievement goals postulated four types of achievement goals: mastery-approach, mastery-avoidance, performance-approach, and performance-avoidance goals (Elliot & McGregor, 2001). Students with a mastery-approach goal in an OER course may focus on making progress or mastering the course content, for example, whereas those with a performance-approach goal may aim at achieving a high grade, while those endorsing performance-avoidance goals may try to just get by avoiding failure in the class through OER materials.

Achievement goals predict how students apply learning strategies to perform academic tasks (Bardach et al., 2020; Holzer et al., 2022). Specifically, each type of achievement goal contributes to learner achievement and performance in different ways. Mastery-approach goals motivate individuals to invest effort in accomplishing goals with a focus on mastering competence (Bardach et al., 2020). Studies have consistently shown that mastery-approach goals produce favorable changes in competence-related settings such as developing intrinsic motivations and persistent engagement as well as exhibiting willingness to take on and maintain perseverance with challenges (Bardach et al., 2020; Hulleman et al., 2010). Performance-approach goals propel individuals to acquire expertise with the goal of outperforming others (Elliot & McGregor, 2001). Most studies have shown positive relations between performance-approach goals and remarkable achievement (Hulleman et al., 2010), but performance-approach goals may also lead to one’s anxiety about performance and resistance to difficulty tasks and thereby exhibit maladaptive behaviors such as procrastination (Deemer et al., 2018; Sun & Xie, 2020). In comparison with approach goals, avoidance goals (mastery-avoidance and performance-avoidance) are more likely to yield maladaptive behaviors and negative outcomes (Elliot & McGregor, 2001; Van Yperen et al., 2009).

Achievement goals have been well researched in many disciplines including but not limited to language, science, and math, consistently showing the beneficial effects of mastery-approach goals as well as mostly detrimental results of performance-based goals (Elliot & McGregor, 2001; Senko & Dawson, 2017; Y. Yang et al., 2016; Y. Yang & Taylor, 2013). More recently, Darnon et al. (2018) documented the discriminating effects of performance goals on academic performance, showing that mastery-approach goals predicted final grade for lower-class students, whereas performance-approach goals only did so for upper-class students. Similarly, the findings of another recent study (in Peru) favored mastery-approach goals over other goals (Matos et al., 2017). Researchers have also applied achievement goals theory as a lens for investigating learner motivation of engaging in open learning environments, such as MOOCs and other open courseware platforms. de Barba et al. (2016) reported that learners’ mastery-oriented goal positively predicted their frequency of quiz attempts and course grades in a MOOC. Wang and Baker (2018) found that learners’ goal-orientation affected their tendency to completing a MOOC in that learners with mastery-goal orientation are more likely to complete a MOOC and earn a course certificate.

Those findings confirm the importance of cultivating mastery-oriented goals in OER-based classes so that students may focus on improving, making progress, and growing their expertise while discouraging their performance-based goals priming on competition, performance, or fear of failing (Elliot & McGregor, 2001). To promote mastery-approach goals in classes using OER, therefore, it is compelling to find out the antecedents of achievement goals. Several studies have examined motivation variables that might decode the achievement goals that students endorse (Chen & Liu, 2009; Jiang et al., 2018), some of which pointed to the values students attach to a task (Greene et al., 1999).

EVT

EVT attributes individuals’ choice of engaging in achievement-related behaviors to an interplay between their expectation about success based on self-appraised competence (e.g., competence belief) and their perceived value of engaging in such behaviors (e.g., task value belief) (Eccles & Wigfield, 1995; Wigfield & Eccles, 2000). As a theory of motivation, EVT has been widely explored in education research, and evidence about the importance of EVT for learners’ academic achievement has been widely documented (Joo et al., 2013; Robinson et al., 2022; Wigfield & Eccles, 2000). Joo et al. (2013) found college students’ task value belief significantly predicted their course satisfaction, performance, and persistence in an online course. Robinson et al. (2022) commented that the relation between students’ self-appraised competency beliefs and their course achievement is recursive in that students’ prior high achievement leads to high competence beliefs and, in turn, such high competence beliefs result in their high level of engagement in learning. Although both EVT components are positively associated with their academic achievement, Bissell-Havran and Loken (2009) argued that task value beliefs tend to correspond to individual choices of academic behaviors but competence beliefs attend to learner performance in academic settings.

Task value belief, described as one’s value appraisal of engaging in an achievement-related task (Eccles et al., 1983), explains why learners undertake academic tasks and relates to their performance and achievement in this setting (Eccles & Wigfield, 1995; Joo et al., 2013). For instance, Lee and Song (2022) identified learners’ task value belief as a significant predictor of their consistent engagement in a MOOC. de Barba et al. (2016) found that learners’ task value belief predicted their frequency of video hits and course grades in MOOCs. To facilitate learner choices over academic tasks, understanding their task value belief toward learning is critical.

Four types of task value beliefs—intrinsic, utility, attainment, and cost value—have been widely investigated in education (Flake et al., 2015; Jiang et al., 2018; Trautwein et al., 2012; Wigfield & Eccles, 2000). Specifically, intrinsic value addresses the sense of enjoyment acquired from performing a task (Bissell-Havran & Loken, 2009). Learners who maintain intrinsic value beliefs toward a task choose to perform the task because of interests or enjoyment internal to the task itself (Spinath et al., 2006). In OER-based learning, students with intrinsic values may enjoy a variety of OER course materials and assignments instead of merely focusing on achieving a satisfactory score. Berweger et al. (2022) indicated that students’ intrinsic value toward learning led to their positive emotions in an online learning environment.

Attainment value is described as one’s subjective judgement of the importance of a task, so learners with such a valuation toward a task tend to complete the task because of its importance for their sense of identity (Arens et al., 2019). Even for learners with a low level of competence belief, attainment value motivates them to behaviorally engage in learning and thereby improve their performance (Putwain et al., 2019). Likewise, students endorsing attainment values in an OER-based learning setting may find it important to review no-cost materials and complete the course assignments, attaching self-appraised importance to doing so.

Utility value implies individuals’ perceived usefulness of completing a task for accomplishing their goals, either long-term or short-term (Wigfield & Eccles, 2000). Putwain et al. (2021) found a positive relationship between utility value and learners’ enjoyment in completing a task but, on the other hand, learners’ utility value was negatively correlated with their feeling of boredom about the task. Students with utility value, as the term suggests, may value taking an online OER course as a useful means to an end, that is, earning course credits and eventually an academic degree. For instance, learners with advanced academic degrees and prior experience in online learning exhibited a significant preference for obtaining MOOC certificates, and thereby their utility value toward MOOC certificates motivated them to complete the MOOCs (Rõõm et al., 2023).

In addition, cost value describes one’s subjective estimation of the physical and mental sacrifice required to perform a task (Flake et al., 2015). For example, costs may include the amount of time and effort invested in a specific task and discomfort and anxiety from taking on the task (Jiang et al., 2018; Wigfield & Eccles, 2000). In addition, the opportunity cost for choosing to complete the task rather than other incidents or plans is another part of costs (Flake et al., 2015; Jiang et al., 2018). In the context of OER-based online learning, students who prioritize cost value tend to find it appealing that they no longer need to pay for commercial textbooks; however, spending additional time and effort on adapting to the OER-based course materials and novel course assignments may be discouraging for them. Lee and Song (2022) found that students’ perceived cost for taking a free online course (e.g., MOOCs) negatively affected their persistence, accounting for their dropout or failure in this course.

Model and hypothesis

The importance of achievement goals for learning has been well evidenced, so understanding the antecedents of students’ achievement goals is critical. EVT provides a basis for examining the predictors of students’ achievement goals from a task value perspective. For instance, intrinsic value is related with motivations inherent to competence or learning itself, but utility and attainment value are considered as extrinsic value (Putwain et al., 2021). In addition, costs are mostly negative valence of a task in that high perceived costs lead to low motivation of performing the task (Flake et al., 2015; Jiang et al., 2018) and avoidance goals (Bong, 2001). However, literature provides relatively limited insights on facilitating achievement goals in contexts of using OER for teaching and learning. Among limited evidence, Tang et al. (2020) found that learners’ perceived usefulness and perceived ease of using OER are predictors of their acceptance of OER, but this finding stemmed from learners’ valuation of OER from a technology acceptance perspective (e.g., the intention of using OER rather than the reason to use OER).

As EVT is domain-specific (Wigfield & Eccles, 2000), it is unclear whether the established relationships between task value beliefs and motivation would translate to courses using OER, and it is unknown what task value beliefs are endorsed by college students enrolled in OER courses. As a primary effort to close the gap, this study aimed to establish a model of examining how task value beliefs toward learning with OER predict their achievement goals toward learning in a college-level education course. Specifically, we had the following research questions for the study:

1. What were the achievement goals of the students taking the OER course?

2. What were the task values student held toward the OER course?

3. How did students’ task value beliefs predict their achievement goals in the OER course?

Based on the literature, we investigated the following hypotheses (H):

• H1. Students’ intrinsic value toward learning with OER significantly predicts their achievement goals toward learning.

• H2. Students’ attainment value toward learning with OER significantly predicts their achievement goals toward learning.

• H3. Students’ utility value toward learning with OER significantly predicts their achievement goals toward learning.

• H4. Students’ cost value toward learning with OER significantly predicts their achievement goals toward learning.

Methods

Participants and settings

This study took place in an OER-based online course for senior undergraduates with an education major at a public university in the southeast of the United States of America. This study involved a total of six offerings of this course that spanned over six semesters from Summer 2019 to Fall 2021. All those offerings were taught by the same instructor and the length of each offering was equal (e.g., 16 weeks). All the course materials used in this course were OER compiled by the instructor in collaboration with two colleagues through an OER grant project. Specifically, OER used in this course covered a wide range of open-license resources with Creative Commons licenses including those from library collections, magazines, websites, and book chapters or journal articles. The instructor made intentional effort to highlight the applicability of the course content to their personal and professional lives in addition to reinforcing the need to review the course materials thoroughly to earn satisfactory grades. Students taking this course did not need to pay to access OER-based course materials (except for printing costs if they chose to have the materials printed out). Further, all students taking this course received the same instruction, course assignments, and evaluation criteria from the start to the completion. At the beginning of each offering, the instructor sent a welcoming announcement including an introduction of OER and the no-cost nature of the course. As part of the scaffoldings for students’ familiarity with OER, the instructor offered weekly virtual class meetings to help students navigate through various OER course materials, content modules and projects and answer students’ questions about logistics and content. Additionally, students resorted to discussion forums on frequently-asked-questions as well as instructor’s emails for any assistance or clarification they might need.

We used convenience sampling to recruit students from each offering of this course. An Institution Review Board approval was granted before recruiting the participants. Enrolled students in each offering were recruited by an email from the instructor. Besides an overview of this study, the recruiting email stated that each participant of this study was given a small portion of extra points and they held the right to withdraw from the study at any point without penalty on their course grade. In the end, a total of 246 college students with an education major opted to have their data included in the study. Among these participants, 39 of them were male, 206 of them were female, and one was genderqueer. Moreover, 173 participants were Caucasian, 43 were African American, 2 were Native American, 1 was Asian American/Oriental, 8 were Hispanic, 9 were biracial/multiracial, 4 chose “others” without specification, and 6 did not answer this question. Participants’ age ranged from 18 to 44 with most (N = 197, 80%) aged between 20 and 22.

Measures

The instruments used in this study included the Expectancy–Value Scale (Trautwein et al., 2012) and the 2 x 2 Achievement Goal Questionnaire (Elliot & McGregor, 2001).

We adapted the expectancy–value scale (Trautwein et al., 2012) to assess participants’ task value beliefs toward using OER for learning in this course. We revised all the original 12 items from the Expectancy–Value Scale to examine domain-specific task value beliefs. We added another four items (two about cost and two about utility value) to ensure at least three items assessed each construct of task value beliefs (Raubenheimer, 2004). All the items used a 7-point Likert-type scale to indicate the degree to which each item was true to the participants, ranging from not at all true of me (1) to very true of me (7). In the end, we used a total of 15 items: three questions gauging attainment value, five questions evaluating intrinsic value, four questions assessing utility value, and three questions estimating the perceived costs. Sample items included “Learning through OERs in this class is important to me personally” (attainment value) and “The amount of time I spend on learning this OER class keeps me from doing other things I would like to do” (cost). Particularly, all the items regarding costs were reverse coded to keep uniform directionality. The Cronbach’s alpha for the Expectancy–Value Scale was .998.

We modified the 2 x 2 Achievement Goal Questionnaire (Elliot & McGregor, 2001) to evaluate participants’ achievement goals toward learning in this domain (e.g., education). We retained and rephrased a total of 12 items from the original questionnaire to make each item specific to the domain. Each type of achievement goals was gauged by three items, each of which followed a 7-point Likert-type scale ranging from not at all true of me (1) to very true of me (7). Sample items included “It is important for me to do better than other students” (performance-avoidance goals) and “It is important for me to understand the content of this course as thoroughly as possible.” (mastery-approach goals). The Cronbach’s alpha for the Achievement Goal Questionnaire was .997.

Structural equation modeling

We established a multiple indicators multiple causes (MIMIC) model within the framework of structural equation model (SEM) to understand how the four task value beliefs predict each type of achievement goals. SEM is a statistical technique that can be used to test the relationships between a set of independent variables and dependent variables in a causal model. We first examined the construct validity of the achievement goals questionnaire using confirmatory factor analysis (CFA) and then conduced the MIMIC analysis to regress the exdogenous variables (four achievement goals) on the endogenous variables (task value beliefs). Figure 1 and Figure 2 illustrate the path diagrams for the CFA and MIMIC models. We chose the robust maximum likelihood (MLR) method for the parameter estimation to help mitigate the potential deleterious effect caused by the non-normality of the survey item responses (Curran et al., 1996; Finney & DiStefano, 2006). Furthermore, we examined the model fit of the CFA model and the SEM model by evaluating the indices, such as ${\chi }^2$ test, comparative fit index (CFI), Tucker-Lewis fit index (TLI), root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR). And then we used the SEM model to investigate the relationship between the four types of task value beliefs and the four types of achievement goals in this specific domain.

Figure 1. Results for CFA. Note. pap: performance approach; pav: performance avoidance; map: mastery approach; mav: mastery avoidance.

Figure 2. MIMIC for achievement goals with expectancy values as covariates. Note. pap: performance approach; pav: performance avoidance; map: mastery approach; mav: mastery avoidance.

Results

Descriptive statistics

Table 1 illustrates the mean, variance, skewness, and kurtosis for each item and covariates. We first conducted the statistical test of normality for each item since the maximum likelihood method implemented in the MIMIC analysis assumes all the observed variables follow normal distributions. The skewness for each item was located between -1.96 and 1.96 (Field, 2013), and the kurtosis for each item was smaller than 3, indicating all item responses can be considered as normally distributed (Westfall & Henning, 2013).

Table 1. Descriptive statistics for each item or covariate.

Items or covariates	Mean	Variance	Skewness	Kurtosis	Minimum	Maximum
I1	4.10	2.73	−0.18	−0.55	1	7
I2	4.59	2.81	−0.55	−0.29	1	7
I3	4.26	2.90	−0.27	−0.56	1	7
I4	4.24	2.70	−0.32	−0.64	1	7
I5	4.84	2.27	−0.65	0.08	1	7
I6	4.16	2.74	−0.23	−0.66	1	7
I7	5.99	1.21	−0.93	0.26	2	7
I8	5.79	1.41	−0.87	0.55	1	7
I9	5.52	1.62	−0.73	0.18	1	7
I10	5.85	1.62	−1.08	0.68	1	7
I11	5.96	1.56	−1.50	2.54	1	7
I12	5.27	2.28	−0.84	0.33	1	7
Attainment	14.60	13.11	−0.21	0.11	3	21
Intrinsic	22.24	30.23	−0.06	0.28	5	35
Utility	21.44	19.82	−0.42	0.00	4	28
Cost	13.55	15.11	−0.13	−0.09	3	21

CFA

We first conducted CFA to check the construct validity of the achievement goal questionnaire. The model fit indices were reported via the following indexes: chi square to the degree of freedom χ²(48) = 96.98; p < .001; CFI = .96; TLI = .95; RMSEA = .05; and SRMR = .06. The results revealed that the proposed SEM model showed an acceptable fit for the data (Bentler, 1990). All the factor loadings were significantly larger than 0, and the factor correlations ranged from .22 to .41.

SEM

For the MIMIC analysis, the model fit indices generally demonstrated a good model-data fit. To be specific, CFI (0.95) and TLI (0.92) were larger than the suggested value .90; RMSEA (0.06) and SRMR (.05) were smaller than the suggested value .80 (Bentler, 1990). Chi square to the degree of freedom (χ²/df = 2.68 < 5.0) was smaller than the common benchmark of 5.0. The results revealed a good fit for the model (Bentler, 1990).

The path significance, the path coefficient (β), and the standard error (in parenthesis) by each path are shown in Figure 2. Particularly, the paths between utility values and approach goals (mastery-approach and performance-approach) were significant at alpha level of .05 with coefficients but the relations were in reversed directions. Specifically, the utility values were positively related to mastery-approach goals (β = .143, p < .001) but negatively related to performance-approach goals (β = -.074, p = .01). In addition, the path between costs and mastery-avoidance goals was significant at alpha level of .05. Costs were positively related to mastery-avoidance goals (β = .093, p < .001).

Discussion

This study investigated college students’ achievement goals toward using OER in an education course from an expectancy–value perspective, as part of the effort to fill the gap in relevant empirical evidence. The SEM analysis confirmed that college students’ task value beliefs were antecedents of their achievement goals in the specific context of using OER for learning. Research has shown that approach goals, especially mastery-approach goals, lead to a higher level of course achievement than avoidance goals (Elliot, 1999; Elliot & McGregor, 2001; Sun & Xie, 2020). The findings of this study provide implications for facilitating student motivation toward using OER and maintaining a high level of course performance in OER-based courses.

First, the study found that college students’ utility values toward using OER for learning positively predicted their mastery-approach goals but was negatively correlated with their performance-approach goals. This finding confirmed the importance of reinforcing students’ perceived usefulness of OER for their learning goals. Prior findings (Tang, Lin et al., 2020, 2021) attributed students’ perceived usefulness of OER as a predictor of their intention of continuing to use OER. Beyond merely the acceptance of using OER, this study extends the investigation focused on students’ valence of competence and the antecedents of their motivation to learn with OER. Specifically, when students find using OER beneficial for their learning, they tend to develop positive valence of competence and thereby strive for obtaining competence and producing positive academic outcomes. In addition, students with a higher level of utility value toward using OER are more likely to invest effort in developing proficiency of competence rather than merely striving for excellence in course performance. It is inferred that, by providing free educational resources, OER allow students to concentrate on learning without being concerned about financial burden or the return of investment (e.g., textbook spending, tuition), reiterating the benefits of OER described by Wiley and Hilton (2018). This finding accentuates the need to develop college students’ awareness of OER and the usefulness of OER for their learning. For example, strategies such as open educational practices may be considered as students can gain such awareness in an authentic experience of using OER (Tang, 2020, 2021). It is also noteworthy that the instructor compiles OER-based materials with a focus on the connection between course content with their future careers. The findings of the study add to the importance of continuing such application-based efforts in improving the utility values of an online OER course.

In addition, the study indicates that students’ self-estimated cost of using OER in learning predicted their mastery-avoidance goals, echoing prior findings (Bong, 2001; Flake et al., 2015). Students’ high perceived cost toward a task is associated with their negative feelings such as discomfort and anxiety about completing the task as they are concerned about the time and effort invested required for the task (Flake et al., 2015; Jiang et al., 2018). In addition, mastery-avoidance goals divert student attention from elaborate effort in developing competence to concerns over avoiding failure or errors, which may lead to maladaptive outcomes such as anxiety (Sun & Xie, 2020; Van Yperen et al., 2009). It is thus speculated that students with a high perceived cost about using OER tend to focus on developing competence but for the purpose of avoiding failure, especially given such a high level of time and effort required for mastering the competence. Such a valence of academic tasks then leads to the development of students’ mastery-avoidance goals. As the detrimental effects of mastery-avoidance goals are well documented, it is essential to reduce the cost of taking courses using OER. As Flake et al. pointed out, there are different types of cost when engaged in a task, such as task effort cost, outside effort cost, loss of valued alternatives cost, and emotional cost. While courses using OER are generally less costly financially, it may cost students more time and efforts to organize and master the course materials and complete various assignments. Therefore, it is essential that instructors reduce the cognitive and organizational load for students so they can be at least equivalent to, if not more, time- and energy-efficient than courses using traditional textbooks. Compared with well-developed and well-organized commercial textbooks, OER may appear less-organized, messier, and more daunting, hence increasing the effort and emotional cost (Flake et al., 2015) for students. Therefore, the findings of this study highlight the importance of providing scaffolding for students to help reduce these types of costs. The instructor in the study offered such initiatives, including synchronous virtual class meetings, course frequently-asked-questions discussion forums, one-on-one guidance, and peer support. The finding of the study reinforces the significance of providing such scaffolding to improving students’ value belief about OER by reducing the cost for students to use OER materials.

While the finding that intrinsic values and attainment values did not predict achievement goals is astonishing, it seems plausible. Intrinsic values are usually associated with motivation internal to learning whiles attainment values are relevant to perceived importance of a learning task (Arens et al., 2019; Putwain et al., 2019; Putwain et al., 2021). This finding may be due to the drawbacks of OER as the design of OER fails to develop students’ interests in learning or foster their perceived importance of using OER for learning in this specific context. Research has shed light on the deficiency of OER in providing trustworthy and engaging content (Hilton, 2016, 2020; Tang, 2020). To develop students’ intrinsic values and utility values toward using OER for learning, strengthening the design of OER is critical. When adapting or curating OER for a course, instructors should consider the alignment between the content of OER and instructional objectives of the course. In addition, the manner in which instructors integrate OER in the course may affect student perception of OER. It is necessary to draw an explicit link between OER content and course achievement outcomes, rather than just embedding OER content in a course (Tang, 2020; Wiley & Hilton, 2018).

This study also provides practical implications for supporting teaching and learning with OER. First, instructors need to integrate task-value activities and deliver task-value messages to reinforce students’ perceived usefulness of OER. For instance, task-value activities and messages that can build an explicit link between using OER and course outcomes are recommended (Putwain et al., 2019). Second, it is critical to reinforce the design of OER, especially in relation to providing high-quality learning content rather than just free access to educational resources. This will hopefully help students develop intrinsic interests in learning and identify the importance of using OER in their course activities. Third, instructors may assure students that in addition to no expense for using OER, the amount of effort and time invested in learning is comparable to that using traditional textbook. Helping students reduce their worries about the cost of using OER is important to motivate them to develop competence with OER.

This study is also limited by several constraints. First, this study solely relied on self-reported instruments to collect students’ subjective perception of using OER, but students’ responses may be subject to bias and may not accurately assess their perception throughout the course. Second, the analysis presents a descriptive account of students’ task value beliefs and their achievement goals in an education course. Given that students’ perceptions of their task value belief and achievement goals are domain specific, the generalization power of the findings of this study is relatively limited. Third, this study only tapped into college students’ task value beliefs in online OER courses, but their competence beliefs were not addressed. Future research may consider college students’ competence beliefs to develop a unified understanding of student motivation toward OER from the lens of EVT. For future research, we also recommend adopting multimodal measures of students’ perceptions and beliefs toward learning in order to provide a validated account of the relationship between students’ value beliefs and their achievement goals. In addition, future research may consider validating the findings in a different subject or context to increase the generalizability of the findings.

Acknowledgments

We wish to express our sincere gratitude to Professor Anne Barnhart and Associate Professor C. J. Ivory for their involvement in and contribution to the grant project which made this study possible. We also sincerely appreciate the support from Dr. Virginia Clinton-Lisell and the Open Education Group.

Disclosure statement

No potential conflict of interest was declared by the author(s).

Data availability statement

The participants of this study did not give written consent for their data to be shared publicly, so due to the sensitive nature of the research supporting data is not available.

Additional information

Funding

This study was partially funded by Open Educational Resources Grant (ALG#380) through Georgia Affordable Learning. We also appreciate the support from the Open Education Group.

Notes on contributors

Hengtao Tang

Hengtao Tang is an associate professor of learning design and technologies in the Department of Leadership, Learning Design, and Inquiry at the University of South Carolina. His research interests include multimodal data analytics, self-regulated learning, AI-scaffolded learning, and open educational resources.

Yan Yang

Yan Yang is a professor of educational psychology at the University of West Georgia. Her major research interests are motivation in distance learning and diversity education.

Yu Bao

Yu Bao is an assistant professor in the Department of Graduate Psychology and an assessment specialist in the Center for Assessment and Research Studies at James Madison University. Her research interests include psychometric modeling and higher education.