the role of beliefs and motivation in asynchronous online learning in
Transcription
the role of beliefs and motivation in asynchronous online learning in
J. EDUCATIONAL COMPUTING RESEARCH, Vol. 50(3) 315-341, 2014 THE ROLE OF BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING IN COLLEGE-LEVEL CLASSES KUI XIE The Ohio State University KUN HUANG Mississippi State University ABSTRACT Epistemic and learning beliefs were found to affect college students’ cognitive engagement and study strategies, as well as motivation in classroom settings. However, the relationships between epistemic and learning beliefs, motivation, learning perception, and students’ actual learning participation in asynchronous online settings have been under-studied and under-theorized. In this study, 132 students participated in collaborative learning activities through asynchronous online discussions in a college-level online course. The results from correlation analysis and structural equation modeling indicate that epistemic and learning beliefs have significant effects on students’ learning participation and perceived learning, through the mediation of achievement goals. The findings provide theoretical and practical implications for the design of online learning. 1. INTRODUCTION Research has demonstrated the interactive relationships between epistemic and learning beliefs, motivation, and learning (Braten & Strømsø, 2004; DeBacker & Crowson, 2006; Kardash & Howell, 2000; Ravindran, Greene, & DeBacker, 315 Ó 2014, Baywood Publishing Co., Inc. doi: http://dx.doi.org/10.2190/EC.50.3.b http://baywood.com 316 / XIE AND HUANG 2005). The few existing attempts also examined how these variables interact with each other to collectively impact learning in classroom settings (Chen & Pajares, 2010; Kizilgunes, Tekkaya, & Sungur, 2009; Ravindran et al., 2005). As today’s education moves toward online settings, students’ online learning experience is significantly different from that in face-to-face settings. They have to rely on computer-mediated communications and exercise self-regulation to achieve success in online learning. Investigations become necessary on the interplay between learners’ epistemic and learning beliefs, motivation, and learning in online learning environments. In addition, previous studies on epistemic and learning beliefs relied mostly on self-reported measures of students’ cognitive engagement (e.g., DeBacker & Crowson, 2006; Ravindran et al., 2005) or study strategies (e.g., Kardash & Howell, 2000; Schommer, Crouse, & Rhodes, 1992). Recent research has suggested that students’ actual participation and learning behavior can greatly reflect their motivational characteristics (Fredericks, Blumenfeld, & Paris, 2004; Reeve, 2006). While student participation data are difficult to obtain in face-to-face settings (Dennen, 2008), most Learning Management Systems (LMS; e.g., WebCT®) have the capability to track a variety of students’ participation records in online learning activities (Xie, 2013). The learning participation data, also being referred to as learning analytics (Ferguson, 2012), become valuable resources for educational research. More importantly, these data may provide researchers with the opportunity to better understand online learning. Using a combination of self-reported and learning analytics data, the present study built and tested a model among epistemic and learning beliefs, motivation, and student participation and perception in collaborative, asynchronous online discussions in college-level online classes. 2. THEORETICAL FRAMEWORK 2.1 Relationship Between Beliefs and Learning Epistemic beliefs refer to an individual’s implicit beliefs about knowledge and knowing (Hofer & Pintrich, 2002). Research on epistemic beliefs started with a unidimensional model with “dualist, absolutist, and objectivist” on one end, and “relativist, subjectivist, contextual, constructivist, and evaluative” on the other (Stathopoulou & Vosniadou, 2007, p. 146). Later, researchers advanced multidimensional models in examining the relationship between epistemic beliefs and learning (e.g., Schommer, 1990; Schommer-Aikins, Duell, & Hutter, 2005; Schraw, Bendixen, & Dunkle, 2002; Wood & Kardash, 2002). Five general dimensions of epistemic beliefs have been articulated, including the structure of knowledge (ranging from the belief that knowledge can be described as isolated pieces to the belief that knowledge structure is complex and highly interrelated), the certainty of knowledge (ranging from the belief that knowledge is certain and BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING / 317 unchanging to the belief that knowledge is tentative and evolving), the source of knowledge (ranging from the belief that knowledge comes from omniscient authorities to the belief that knowledge emerges from personal construction), the nature of ability to learn (ranging from the belief that ability to learn is innate to the belief that learning ability can be acquired with effort), and the speed of learning (ranging from the belief that learning takes place quickly or not at all to the belief that learning is a gradual process) (Braten & Strømsø, 2005; Schommer, 1990; Schraw et al., 2002). The five dimensions were later questioned by Hofer and Pintrinch (1997), who argued that the nature of ability and the speed of learning do not conceptually belong to the construct of epistemic beliefs. Schommer-Aikins (2004) later theorized that the two sets of beliefs are in fact beliefs about learning. She further advocated the need to investigate epistemic beliefs in light of other closely related belief systems such as beliefs about learning. Research has found that epistemic beliefs (e.g., simple and certain knowledge) and beliefs about learning (e.g., quick learning) could influence academic performance (e.g., Qian & Alvermann, 1995; Schommer, 1993; Schommer-Aikins & Easter, 2006; Schommer-Aikins et al., 2005; Windschitl, 1997), cognitive engagement (e.g., DeBacker & Crowson, 2006; Ravindran et al., 2005), and study strategies (e.g., Kardash & Howell, 2000; Schommer et al., 1992). While previous research demonstrated a clear impact of epistemic and learning beliefs on learning, it is unclear whether the relationship is direct, indirect through other mediating variables, or both. Some studies found that epistemic and learning beliefs directly influenced learning (e.g., Cano, 2005; Kizilgunes et al., 2009; Schommer, 1993; Schommer, Calvert, Gariglietti, & Bajaj, 1997; Schommer-Aikins et al., 2005). Others suggested that epistemic and learning beliefs affected learning indirectly through latent variables such as learning approaches (e.g., Cano, 2005; Schommer et al., 1992), regulation of cognition (e.g., Greene, Muis, & Pieschl, 2010; Muis, 2007), and motivation (e.g., DeBacker & Crowson, 2006; Hofer & Sinatra, 2010; Kizilgunes et al., 2009). As a result, in this study we modeled the relationship between beliefs and students’ participation and perception in online learning both directly and indirectly through the mediation of motivation. 2.2 Relationship Between Motivation and Learning Students’ motivation in achievement settings is often elaborated through goal orientations, which emphasize the reasons for engaging in academic activities (Eccles & Wigfield, 2002). Earlier research distinguishes between two types of goals: mastery goals (or learning/task goals) focusing on increasing competence, and performance goals (or ego goals) focusing on gaining favorable or avoiding negative judgment of competence (Ames, 1992; Dweck, 1986; Nicholls, Cobb, Yackel, Wood, & Wheatley, 1990). In resolving controversial findings regarding performance goals, Elliot and Church (1997) proposed a 318 / XIE AND HUANG trichotomous achievement goal framework, which divided performance goals into performance-approach (outperform others in a desirable event to show competency) and performance-avoidance goals (avoid demonstration of incompetence in an undesirable situation). Achievement goals have important bearings on learning. Studies have reliably shown that individuals with mastery goals are likely to choose challenging tasks, show interest, persistence and effort, demonstrate self-regulated learning, adopt meaningful study strategies, and achieve better learning outcomes (Elliot & Dweck, 1988; Elliot, McGregor, & Gable, 1999; Eppler & Harju, 1997; Graham & Golan, 1991; Licht & Dweck, 1984). Further, mastery goals were found to be more beneficial in social settings (in comparison to isolated settings when individuals perform tasks alone) (Utman, 1997). Comparatively, individuals with performance-avoidance goals use shallow processing strategies, demonstrate less intrinsic motivation and interest, and show poor academic performance (Church, Elliot, & Gable, 2001; Elliot & Church, 1997; Elliot & McGregor, 1999; Greene & Miller, 1996; Pintrich & Garcia, 1991). With regard to performance-approach goals, research has shown their positive effects on academic performance and affect, especially in competitive college classrooms (Church et al., 2001; Elliot & Church, 1997; Harackiewicz, Barron, Tauer, & Elliot, 2002). On the other hand, performanceapproach goals may also lead to negative outcomes such as shallow cognitive strategies (Elliot et al., 1999; Greene & Miller, 1996; Midgley, Kaplan, & Middleton, 2001). Another important motivational construct, self-efficacy, is defined as “the conviction that one can successfully execute the behavior required to produce the outcomes” (Bandura, 1977, p. 193). It is a person’s inner belief of his or her own competency. Compared with their less self-efficacious peers, the students with higher levels of self-efficacy are likely to initiate effort when faced with challenging tasks, show more persistence and effort, adopt meaningful learning strategies, and achieve better academic performance (Greene & Miller, 1996; Pajares & Miller, 1994; Walker, Greene, & Mansell, 2006; Zeldin & Pajares, 2000; Zimmerman & Bandura, 1994). Achievement goal theories suggest that students’ goal adoption has strong relations with their self-efficacy beliefs. Research has identified positive correlations between self-efficacy and mastery goals (Greene & Miller, 1996; Sins, van Joolingen, Savelsbergh, & van Hout-Wolters, 2008). High self-efficacy was found to predict mastery and performance-approach goals (Elliot & Church, 1997; Greene, Miller, Crowson, Duke, & Akey, 2004; Walker et al., 2006), whereas low competency expectancies predicted performance-avoidance goals (Elliot & Church, 1997). In light of the findings about the relationship between motivation and learning, our model includes relational paths between motivation variables (i.e., mastery goals, performance-approach goals, performance-avoidance goals, and selfefficacy) and learning variables (i.e., learning participation and perception of BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING / 319 learning). Our model also includes relational paths between achievement goals and self-efficacy variables. 2.3 The Relationship Between Beliefs and Motivation Epistemic and Learning Beliefs and Goal Orientation Hofer and Pintrich (1997) posited that epistemic beliefs might influence academic performance through their impact on motivation, which includes achievement goals. They further suggested that individuals with more sophisticated epistemic beliefs would be more likely to adopt mastery goals. DeBacker and Crowson (2006) also argued that particular epistemic beliefs could create tendencies to adopt certain achievement goals. Although research has not yielded consistent results on how different dimensions of epistemic and learning beliefs might influence students’ adoption of achievement goals, recent studies generally suggested that students who held less mature beliefs were less likely to adopt mastery goals but more likely to adopt performance goals, and those who held more mature beliefs were on the contrary. For example, studies have repeatedly shown that students’ adoption of mastery goals was negatively related to beliefs in certain knowledge (Braten & Strømsø, 2005), omniscient authority (Kizilgunes et al., 2009), quick learning (Braten & Strømsø, 2004, 2005), and fixed ability (Cury, Elliot, Da Fonseca, & Moller, 2006; Hong, Chiu, Dweck, Lin, & Wan, 1999; Kray & Haselhuhn, 2007). The adoption of performance-avoidance goals was positively predicted by beliefs in certain knowledge and fixed ability, while the adoption of performance-approach goals was positively predicted by beliefs in omniscient authority (Chen & Pajares, 2010; Muis & Franco, 2009), simple knowledge and fixed ability (Cury et al., 2006; Ravindran et al., 2005). Epistemic and Learning Beliefs and Self-Efficacy Would certain epistemic and learning beliefs give rise to certain levels of self-efficacy beliefs? Research has been done to investigate whether epistemic and learning beliefs would be possible antecedents for self-efficacy beliefs. Paulsen and Feldman’s (1999) correlation study showed that beliefs in simple knowledge and quick learning were related to lower self-efficacy. In Kizilgunes et al.’s (2009) study, path analysis found significant associations between three dimensions of epistemic beliefs (source, development, and justifications) and self-efficacy. Specifically, mature beliefs about the source and development of knowledge were associated with a high level of self-efficacy; on the other hand, those with mature beliefs about the justification of knowledge tended to be less self-efficacious. Different from Kizilgunes et al.’s (2009) findings, Chen and Pajares (2010) used the same instrument to measure epistemic beliefs and found that beliefs about the certainty and justification of knowledge predicted selfefficacy, with more advanced beliefs leading to higher self-efficacy. In Neber and 320 / XIE AND HUANG Schommer-Aikin’s (2002) study, path analysis found that naïve beliefs that success is unrelated to work predicted low levels of self-efficacy. Braten and Strømsø (2005) found three significant predictors of Norwegian student teachers’ self-efficacy: beliefs about the speed of learning, beliefs about control of knowledge acquisition, and implicit theories of intelligence. In a recent study by Komarraju and Nadler (2013), less self-efficacious students were found to be more likely to hold naïve theories of intelligence. Overall, while the findings provided some evidence about the relationship between epistemic and learning beliefs and academic self-efficacy, the findings lack consistency with regard to the impact of specific dimensions of epistemic and learning beliefs on self-efficacy. In light of previous research findings, our study intended to find out whether and how specific dimensions of beliefs interact with students’ motivation. Therefore, our model depicts the relationship between the five dimensions of epistemic and learning beliefs and motivation variables (achievement goals and self-efficacy). 3. TOWARD A THEORETICAL MODEL IN ONLINE LEARNING ENVIRONMENT Online learning environments have brought great benefits to education. Among them, asynchronous online discussion is the most common approach to facilitating knowledge construction and collaborative learning. Therefore, we chose asynchronous online discussion as the learning context in the present study. The recent trend of research in asynchronous online learning focuses on the emerging interaction, discourse, and participation processes among members of a learning community (Lipponen, Hakkarainen, & Paavola, 2004). Students’ participation in asynchronous online discussions was found to be a strong indicator of learning (Xie & Ke, 2011). Students exhibit two types of participation behaviors in asynchronous online discussions: posting and non-posting behaviors. Posting behaviors include posting new messages and replying to others’ messages, both of which leave visible textual input in an LMS. As an indicator of learning, students’ posting behaviors have been investigated in many studies (e.g., content of messages, number of posted or responded messages, length of message, etc.) (e.g., Cheung, Hew, & Ling-Ng, 2008). Students demonstrate non-posting behaviors when they visit a discussion forum to simply read or evaluate others’ messages without actually posting a message. Several recent studies argued that non-posting behaviors should also be studied as an indicator of online learning (Beaudoin, 2002; Dennen, 2008; Xie, 2013). Furthermore, learning analytic data tracked by LMSs involve both quantitative data, which may reflect the frequency of student learning participation, and qualitative data, which may reflect the quality of student learning participation. The body of research has extensively used quantitative data to investigate students’ learning engagement in online discussions. For example, posting BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING / 321 frequency has been investigated as an engagement variable in numerous studies. Learning analytics data of non-posting behavior have also gained attention in recent studies (e.g., Wise, Speer, Marbouti, & Hsiao, 2013; Xie, Yu, & Bradshaw, 2014). On the other hand, qualitative data of student participation, the content of discussions, may be the most valuable source to reflect student learning in online discussions. However, it is still considered to be a challenging task in current research to analyze discussion content and interpret complex natural language in a meaningful way in order to reveal latent constructs (e.g., motivation). Researchers have also recognized technical limitations of content analysis in terms of inter-rater reliability, unit of analysis, generalizability, etc. (Lombard, Snyder-Duch, & Bracken, 2002; Rourke, Anderson, Garrison, & Archer, 2001; Wever, Schellens, Valcke, & Keer, 2006). Therefore, quantitative data, which can be collected and transformed automatically for analyses, provide a reliable and feasible means of evaluating student performance in online learning. The present study collected quantitative data as indicators of student participation in online discussions. Research investigating the relationships among epistemic and learning beliefs, motivation, and learning has been largely focused on classroom situations. A number of studies have begun to examine the relationship among these variables in the online learning environment. For example, Tu, Shih, and Tsai (2008) found that students who had more advanced epistemic beliefs, concurring with a constructivist perspective, had richer learning experience in an open-ended Internetbased learning context. Wu and Hiltz (2004) identified the positive impact of motivation and enjoyment on students’ perceived learning from asynchronous online discussions. Chen and Wu (2012) studied Taiwanese students’ motivation and learning strategies in online learning and found that students’ learning goals and cognitive preferences predicted their use of metacognitive strategies and later influenced their performance. Xie and his colleagues found that posting participation (i.e., messages posted in an asynchronous online discussion system) had a significant relationship with students’ motivation, which suggested that students with higher levels of intrinsic motivation had higher participation rates than those with lower levels of intrinsic motivation (Xie, DeBacker, & Ferguson, 2006; Xie, Durrington, & Yen, 2011). Further, Xie (2013) found significant influences of motivation on the frequency of students’ posting and non-posting participations in asynchronous online learning. The findings also suggested that students’ posting participations were more influenced by extrinsic motives (e.g., course requirements), and non-posting participations were more voluntary in nature and might be a better indicator of students’ intrinsic motivation. In an effort to understand how epistemic and learning beliefs, motivation, and student participation and perception simultaneously interact with each other in asynchronous online learning activities, the present study built and tested a model among the variables by utilizing both self-reported measures and learning analytics. The study also sought to investigate the direct or indirect influence of 322 / XIE AND HUANG epistemic and learning beliefs on online learning. According to the aforementioned theoretical framework, we built a hypothetical model (Figure 1) in which epistemic and learning beliefs (i.e., simple knowledge, certain knowledge, omniscient authority, fixed ability, and quick learning) are related to online learning participation (i.e., posting and non-posting participation) and perception both directly, and indirectly through the mediation of motivation variables (i.e., mastery goals, performance-approach goals, performance-avoidance goals, and self-efficacy). 4. Method 4.1 Participants and Context One hundred thirty-two undergraduate students (32 males and 100 females) at a large university in the southeastern United States participated in this study. The participants’ ages ranged from 19 to 61. Proportion of participants by academic level was: freshmen 5 (3.8%), sophomores 16 (12.1%), juniors 39 (29.5%), and seniors 72 (54.5%). They reported their ethnicity as follows: White 75 (56.8%), African American 50 (37.9%), and others 7 (5.3%). The participants were recruited from eight sections of an undergraduate online course in instructional technology, four of which were offered in the spring semester and four in the fall. Each section had 14 to18 students who chose to participate in this study. Figure 1. Relations between Epistemic and Learning Beliefs, Motivation and Online Learning. Note: Several variables were grouped into categories in order to simplify the presentation of the full model. Epistemic & Learning Beliefs have five subscales: simple knowledge, certain knowledge, omniscient authority, fixed ability, and quick learning. Achievement Goals have three subscales: mastery goals, performance-approach goals, and performance-avoidance goals. Online learning participation has two subscales: posting participation and non-posting participation. BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING / 323 All the sections were taught by the same instructor and followed identical learning procedures. One hundred nineteen participants (90.2%) rated their confidence level as high in the use of technology to complete the coursework. 4.2 Measurement Four groups of variables were measured in this study: epistemic and learning beliefs, motivation (i.e., achievement goals and self-efficacy), online learning perception, and online learning participation (i.e., posting and non-posting participations). Epistemic and Learning Beliefs Epistemic beliefs inventory (EBI) was used to assess epistemic beliefs (Schraw et al., 2002). The reason for choosing EBI is because of its better psychometric quality in assessing epistemic and learning beliefs (DeBacker, Crowson, Beesley, Thomas, & Hestevold, 2008). EBI consists of 23 items measuring five dimensions. Three of the dimensions measure epistemic beliefs: simple knowledge indicating the belief about the structure of knowledge (four items), certain knowledge indicating the belief about the certainty of knowledge (six items), omniscient authority indicating the belief about the source of knowledge (five items). The other two dimensions probe learners’ beliefs about learning: innate ability indicating the belief about the nature of ability to learn (five items), and quick learning indicating the belief about the speed of learning (three items). For each dimension, high scores indicate naïve beliefs. In their validity study, Schraw et al. (2002) documented that the instrument’s subscale a ranged from 0.50–0.60s. Motivation Achievement goals questionnaire developed by Elliot and Church (1997) was adopted to assess students’ goal orientations. This questionnaire is composed of six items for each of the three achievement goals in the trichotomous framework (i.e., mastery goal, performance-approach goal, and performance-avoidance goal). Several studies have provided evidence for the reliability, construct validity, and predictive validity of these measures, where the subscale a’s typically ranged from 0.70–0.90s (see Elliot & Church, 1997; Elliot & McGregor, 1999; Elliot & Thrash, 2001). Dweck’s (1999) confidence measure was used to assess students’ self-efficacy in these online classes (e.g., “I usually think that I’m good in this online class”). The scale is composed of six items and a previously reported coefficient a was .90 (Cury et al., 2006). Perceived Learning Perception of learning from an online discussion scale (Wu & Hiltz, 2004) was administered, which has 10 questions inquiring about students’ perceived 324 / XIE AND HUANG learning in online discussion activities (e.g., “The online discussions helped me to think more deeply”). This scale has been adopted in numerous studies and shown high internal reliability. Online Learning Participation Two components of students’ online learning participation were measured: 1. posting participation index including two subscales: standardized score of the total number of posts and standardized score of the total length of posts by individual students; and 2. non-posting participation index including two subscales: standardized score of the total number of times logged in without posting and standardized score of the total length of logins by students. 4.3 Procedure In this study, students participated daily in online discussion activities in a 16-week duration of an instructional technology course. They were assigned to small groups of 8 to 10. The online discussion activities invited the students to share information and contribute to knowledge construction. The instructor monitored and facilitated the online discussion activities. The online course was delivered in the WebCT® LMS. Asynchronous online discussions were facilitated by an online discussion system (iDiscuss) developed by the researchers. One special feature of this system is its easy-to-access data collection and management functions, which has the capability to automatically track and record students’ posting and non-posting participation data. Instructors and researchers can easily download students’ online discussion data for analysis (i.e., time of login, total message posts, total topics read, etc.) (Xie, 2013; Xie, Miller, & Allison, 2013). At the end of the semester, the students were asked to complete a survey voluntarily. The survey was delivered online, consisting of measures of epistemic and learning beliefs, goal orientations, self-efficacy, and perceived learning toward the online discussion activities. Students’ learning analytics data were exported at the end of the semester. Posting and non-posting participation indices were calculated from the data downloaded from iDiscuss for further statistical analyses. 5. RESULTS Three steps of analysis were performed in order to test the relationships among epistemic and learning beliefs, achievement goals, self-efficacy, students’ perceived learning, and their actual online participations in the asynchronous online learning activities. The first step calculated descriptive statistics and reliability coefficients for the instruments measuring each of the variables. In the second step, a correlation matrix was generated to examine the relationships among BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING / 325 variables. The third step utilized Structural Equation Modeling (SEM), using maximum likelihood techniques with AMOS® version 18. 5.1 Descriptive Statistics and Reliability Coefficients Cronbach a reliability coefficients were computed for each of the selfreported subscales. These values were consistent with the reliability values reported in the literature, providing evidence of internal consistency. For each of the subscales, Table 1 shows a sample item, descriptive statistics, and its Cronbach a reliability coefficient. 5.2 Subscale Inter-Correlations An examination of zero-order correlations, shown in Table 2, provided the validity of composite variables in epistemic and learning beliefs, goal orientations, self-efficacy, online learning participation, and online learning perception. Consistent with theory, the results suggested that epistemic and learning belief variables were significantly correlated with each other, except for the relationship between omniscient authority and quick learning. Among the goal orientation variables, performance-approach goal was significantly correlated with both mastery goal and performance-avoidance goal. In addition, all inter-correlations between the online learning participation and perception variables were significant at .01 level. The correlations among epistemic and learning beliefs, goals, and self-efficacy suggested that there could be potential influential effects of epistemic and learning beliefs and self-efficacy on achievement goals. Omniscient authority, quick learning, and simple knowledge had a significantly positive correlation with performance-avoidance goal. Also, simple knowledge had a positive correlation with performance-approach goal, and quick learning had a negative correlation with mastery goal. Self-efficacy had a positive correlation with mastery goal and performance-approach goal, but a negative correlation with performanceavoidance goal. The correlations between goals, self-efficacy, and online learning variables suggested that goals and self-efficacy might contribute to students’ online learning engagement. The correlation matrix showed that both mastery goal and self-efficacy significantly correlated with all online learning variables. Yet performance-approach goal and performance-avoidance goal did not have a significant correlation with any of the online learning variables. In addition, the correlation matrix did not show significant correlations between epistemic and learning beliefs and online learning variables. The result might be due to the existence of mediators or latent variables between these two groups of variables, which necessitated further analysis using a more advanced statistical technique. 4.09 3.07 4.91 3.80 2.46 5.96 4.38 4.61 5.20 5.26 Simple knowledge: Too many theories just complicate things. Certain knowledge: What is true today will be true tomorrow. Omniscient authority: People who question authority are troublemakers. Fixed ability: How well you do in school depends on how smart you are. Quick learning: Students who learn things quickly are the most successful. Mastery goals: I want to learn as much as possible from this class. Performance-approach goals: It is important for me to do better than the other students. Performance-avoidance goals: My goal for this class is to avoid performing poorly. Self-efficacy: I usually think that I’m good in this online class. Perceived learning: The online discussions helped me to think more deeply. Note: SD = standard deviation; a = Cronbach a coefficient. Mean Variable: Sample Item .90 .95 .92 .91 .82 .75 SD 1.31 1.03 1.24 1.49 Table 1. Sample Items, Descriptive Statistics, and Reliability Coefficients 1.00-7.00 1.67-7.00 1.00-7.00 1.00-7.00 2.50-7.00 1.00-5.40 1.71-6.00 2.40-6.80 1.50-5.25 1.86-6.14 Min-Max .95 .86 .82 .91 .88 .69 .71 .61 .58 .62 a 326 / XIE AND HUANG *p < .05 significant level (two-tailed). **p < .01 significant level (two-tailed). 12. Perceived learning 11. Non-posting participation 10. Posting participation 9. Self-efficacy 8. Performance-avoidance goals 7. Performance-approach goals 6. Mastery goals 5. Quick learning 4. Fixed ability 3. Omniscient authority 2. Certain knowledge 1. Simple knowledge 1.00 1 .33** 1.00 1.00 .24** 3 .18* 2 1.00 .22* .18* .32** 4 6 .16 1.00 1.00 –.22* .55** –.04 .08 .26** –.11 .33** –.01 5 1.00 .21* .15 .16 .06 –.00 .21* 7 1.00 .24** –.05 .22* .16 .22* .12 .22* 8 –.06 10 –.05 .25** –.11 –.12 .13 1.00 1.00 .25** –.27** –.12 .18* .42** –.05 –.15 –.13 –.35** –.06 –.04 9 Table 2. Correlation among Epistemic and Learning Beliefs, Motivation, Online Participation, and Learning Perception 1.00 .62** .27** –.17 –.02 .23* –.08 –.15 .05 –.05 –.13 11 1.00 .22** .29** .21* –.04 .11 .35** –.11 –.09 .09 –.06 .01 12 BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING / 327 328 / XIE AND HUANG 5.3 Structural Equation Modeling In the third step of data analysis, SEM was used to test the relationships among epistemic and learning beliefs, achievement goals, self-efficacy, and online learning participation and perception from a perspective of structured framework. SEM takes into account the modeling of latent variables and nonlinearities to test the model from an integrated perspective, namely to test whether the overall web of causal relationships adequately describes the data (Fox, 2002; Hatcher, 1994). SEM method has been widely used in educational research (e.g., DeBacker & Crowson, 2006; Greene et al., 2004). To evaluate the hypothetical model, a mixture of recommended fit indices was used. The value of the likelihood ratio chi-square statistic (c2) indicates whether the hypothetical model deviates from the data. The root mean square of approximation (RMESA) estimates the lack of fit compared to the saturated model. The goodness of fit index (GFI) represents the proportion of the variance in the sample variance-covariance matrix accounted for by the model. According to previous research (e.g., Bollen & Curran, 2006; Byrne, 1998; Kline, 2005; MacCallum, Browne, & Sugawara, 1996; Tabachnick & Fidell, 2001), models with c2/df less than 2, GFI greater than 0.90, an RMSEA less than or equal to 0.05 are considered to be a reasonable fit to the data. For the purpose of this study, one hypothetical model was built (Figure 1) in which epistemic and learning beliefs were related to students’ online learning participation and their perceived learning, both directly and indirectly through the mediation of motivation variables. The results indicate that the model explained the data well and fit indices provided evidence of adequate model-to-data fit: the Chi-square goodness-of-fit indices (c2 = 2828, c2/df = 1.693) being less than 2, indices GFI (.911) being greater than 0.9, and RMSEA (.049) falling below 0.05. With respect to the predicted paths, our model was partially supported. Many, but not all, of the proposed paths were supported by the data. The significant relationship among variables in the final model (Figure 2) supported that the influence of epistemic and learning beliefs on learning was significant but was mediated through motivation variables. Due to the complexity of the original model, Figure 2 depicts only the significant paths found in the SEM. Both learning beliefs significantly predicted achievement goals: quick learning (b = –.51, p = .026) had significantly negative effects on mastery goal, while fixed ability significantly predicted both performance-approach goal (b = .28, p = .043) and performance-avoidance goal (b = .12, p = .029). Out of the three epistemic beliefs, omniscient authority (b = .33, p = .009) significantly and positively predicted mastery goal. No other relationship was found in the final model between epistemic and learning beliefs and motivation variables. Within the motivation variables, the relationships between self-efficacy and goals were significant in the resultant model. Self-efficacy significantly and positively predicted both mastery goal (b = .39, p < .001) and performance-approach BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING / 329 Figure 2. Final model with significant paths. Note: By default, path coefficients were tested using a two-tailed test. goal (b = .23, p = .041). It also negatively predicted performance-avoidance goal (b = –.42, p < .001). The results were echoed by the correlation results, which denoted that self-efficacy was positively correlated with mastery goal and performance-approach goal, while negatively correlated with performanceavoidance goal. With respect to the relationships between motivation and online learning participation and perception, several significant paths were found in the final model. Mastery goal significantly predicted all of the online learning variables: perceived learning (b = .43, p < .001), online posting participation (b = .51, p = .004), and online non-posting participation (b = .46, p = .020). All the effects were positive. In addition, performance-avoidance goal had a significantly negative relationship to perceived learning (b = –.14, p = .041) and non-posting behavior (b = –.22, p = .044). The relationships between self-efficacy and online learning participation were also significant. Self-efficacy significantly predicted both posting participation (b = .36, p = .05) and non-posting participation (b = .50, p = .018). No significant relationship was found between self-efficacy and perceived learning. The SEM final model suggested that epistemic and learning beliefs did not have direct effects on online learning participation and perceptions. The result corresponded to the correlation results in Table 2, which found no significant correlation between epistemic and learning beliefs and online learning variables. 6. DISCUSSION This study examined the relationship between epistemic and learning beliefs, motivation, learning perception, and students’ actual learning participation in 330 / XIE AND HUANG asynchronous online learning. The SEM results suggested that the data provided good support for the overall model, in that the fit statistics were moderately strong and most of the proposed paths were supported by the empirical findings. More specifically, the model indicated the following significant causal relationships: beliefs in omniscient authority, quick learning, and self-efficacy can influence mastery goals; beliefs in fixed ability and self-efficacy can influence both performance-approach and performance-avoidance goals; mastery goals can positively influence all online learning variables (i.e., posting participation, nonposting participation, and perceived learning); performance-avoidance goals can negatively influence non-posting online participation and perceived learning; self-efficacy can influence both posting and non-posting participation. 6.1 Indirect Influence of Epistemic and Learning Beliefs on Online Learning Previous research found that epistemic and learning beliefs were directly and indirectly related to learning (Cano, 2005; DeBacker & Crowson, 2006; Greene et al., 2010; Kizilgunes et al., 2009; Schommer-Aikins et al., 2005). The body of research mainly focused on academic performance that was often reflected by students’ learning outcomes (e.g., average grades) or self-reported learning approaches, yet little attention was paid to students’ actual learning behavior. This study made use of learning analytics as a means to examine the relationship between epistemic and learning beliefs and students’ learning behaviors in an online learning context. The results indicated that epistemic and learning beliefs did not directly influence online learning participation and perception, but indirectly through the mediation of achievement goals. These results are supported by the existing literature. For example, Buehl and Alexander (2005) suggested that learner’ epistemic beliefs can be indirectly associated with their achievement and academic performance through motivation. Hofer and Pintrich (1997) and Muis (2007) further contended that epistemic beliefs may function as implicit theories that influence the adoption of goals for learning, and these goals can serve to mediate the relations between epistemic beliefs, cognition, and learning performance, as well as learning approach and achievement (Kizilgunes et al., 2009; Muis & Franco, 2009). Therefore, this study suggested that the influence of epistemic and learning beliefs on online learning participation is indirect and implicit, while achievement goals may play more direct roles in students’ participation in online learning. It is important to note that the learning participation measured in this study does not necessarily reflect the quality of students’ participation. Nonetheless, the participation data served as quantitative indicators of students’ learning engagement in online learning settings. Hence, examining students’ participation behaviors and patterns provides researchers with a new angle for looking into a facet of online learning which may have important bearings on students’ performances. BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING / 331 6.2 Influence of Epistemic and Learning Beliefs on Motivation Previous studies examining the relationship between epistemic and learning beliefs and motivation were mainly conducted in the context of face-to-face settings. The findings of this study provide insights into the contribution of epistemic and learning beliefs to students’ adoption of achievement goals in online collaborative learning settings. The results denoted that both epistemic beliefs (i.e., source of knowledge) and learning beliefs (i.e., the speed of learning, the nature of the ability to learn) significantly predicted students’ achievement goal orientations in online learning. These results are partially in support of previous research suggesting that naïve beliefs may prevent learners from investing effort on the mastery of knowledge. Instead, they pay more attention to satisfying performance requirements (e.g., DeBacker & Crowson, 2006). Regarding the relationship between specific learning beliefs systems and achievement goals, our study suggests that students’ naïve beliefs about quick learning and fixed ability to learn often lead to the deviation from mastery goals and the adoption of performance goals. For the beliefs about quick learning, the results indicate that students who believed that learning occurs quickly were less likely to adopt mastery goals. The results are in line with previous studies (e.g., Braten & Strømsø, 2004). Therefore, it appears that the impact of beliefs about quick learning on achievement goals is similar across face-to-face and online settings. If a student holds the belief that learning has to take place quickly or it will never occur, that student is less likely to set a thorough understanding and mastery of content as his or her goal. Regarding the beliefs about fixed learning ability, the results suggest that, when students believe that their learning ability is innate and stable, they will focus more outwardly on performance goals demonstrating to others or themselves (performance-approach goal), or showing ego-protective behaviors, such as avoiding situations in which the lack of knowledge would be revealed (performance-avoidance goal). This finding is consistent with achievement motivation theories, which suggest that a mastery goal is characterized by the belief that academic effort will result in an incremental achievement. In contrast, a performance goal reflects the belief that innate ability alone leads to success (Topping & Ehly, 1998). Regarding the epistemic beliefs about the source of knowledge, the results indicate that students who believe that knowledge comes from authority are more likely to adopt mastery goals in online learning. The results seem to contradict the general findings in literature that naïve beliefs are often less likely to result in mastery goal adoption (e.g., DeBacker & Crowson, 2006). The common agreement in the literature generally holds that the omniscient authority beliefs about the source of knowledge tend to be naïve while the beliefs that knowledge emerges from personal construction are more mature and sophisticated (e.g., Hofer & Pintrich, 1997; Schommer-Aikins et al., 2005). The study 332 / XIE AND HUANG results challenge this general assumption and suggest that not all naïve epistemic beliefs will lead to lower levels of motivation. This finding supports Greene et al.’s (2010) suggestion that the utility of beliefs may be context-specific. In some cases “naïve” beliefs about the source of knowledge may be helpful to learning. Although the tradition of epistemic beliefs research usually associates naïve beliefs with negative learning approaches and outcomes, this study suggests that in the context of online learning, a belief that knowledge comes from authority may actually lead students’ motivation toward mastery and learning, rather than performance. To explain this finding, we make the following propositions: as students’ beliefs about the source of knowledge grow from naïve to mature, it is possible that their interpretation of the meaning of mastery also evolves from simple to complex, thereby the mastery goals they adopt also vary from simple to complex. When students hold the belief that knowledge is handed down from authority, their understanding about mastery can be simple. In the case of this study, that is, as students learn in online classes, the beliefs of omniscient authority may have promoted students’ perception that mastery means learning from authority sources. Adopting a mastery goal with such a simple understanding, students participated in the class activities because they expected to learn from their instructor or the learning materials (e.g., the book). Together, the naïve omniscient authority belief and the associated simple mastery goal may have made the relationship between these two constructs relatively simpler, clearer, stronger, and easier to establish. On the other hand, when students hold the belief that knowledge comes from personal experience and experiment, their understanding about mastery can become sophisticated. In the case of this study, that is, with a more sophisticated mastery goal, students participated in the class activities because they expected to gain competence and master the learning content from their own knowledge construction by interacting with peers in class. The mature epistemic belief and the associated sophisticated mastery goal orientation may have made the relationship between these two constructs relatively more complicated and also ambiguous. Our propositions raise some questions: Does mastery goal orientation always lead to deep learning strategies and positive learning performance and outcomes? Will simple or sophisticated mastery goals impact learning differently? These propositions warrant investigation in future research on the effects of simple and complex mastery goals on learning variables. In addition, our study did not find any significant relationship between epistemic and learning beliefs and self-efficacy, although literature suggests that several epistemic belief dimensions can influence students’ self-efficacy (e.g., Paulsen & Feldman, 1999). A possible explanation lies in the difference in the extent to which the beliefs were contextualized. In the case of this study, students’ epistemic and learning beliefs were measured at a general level (i.e., students’ general beliefs about knowledge and learning without a connection to specific domains or tasks), while self-efficacy measures were tied to the context of particular academic tasks (i.e., their confidence in completing learning BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING / 333 tasks in the online class). This difference may have caused the non-significant results. Previous studies that found associations between epistemic and learning beliefs and self-efficacy were both tied to a particular domain, for instance, science learning (Chen & Pajares, 2010; Kizilgunes et al., 2009). Yet in this study students’ general epistemic and learning beliefs might not be powerful enough to directly influence their self-efficacy in the specific learning activities in these online classes. 6.3 Influence of Motivation on Online Learning The study results suggested that motivation, being manifested through achievement goals and self-efficacy, played a significant role in predicting students’ perceived learning and their actual learning participation in asynchronous online discussions. Mastery goals positively predicted all of the learning participation and perception variables. Previous research has well documented the evidence that mastery goals can have positive influences on students’ learning strategies, cognitive engagement, learning outcomes, academic achievement (e.g., Eppler & Harju, 1997; Graham & Golan, 1991; Licht & Dweck, 1984). This study found new evidence that students with mastery goals were motivated to participate in asynchronous online learning activities. They had frequent participation in online discussions (including both posting and non-posting activities), and perceived that they had learned a great deal from the online learning activities. On the other hand, performance-avoidance goals negatively predicted students’ non-posting behavior and perceived learning in online classes. Students with performance-avoidance goals tend to focus on avoiding normative incompetence and its associated unfavorable judgment of potential failure (Elliot & Church, 1997). Research shows that performance avoidance is negatively associated with academic achievement, and can often lead to avoidance behaviors such as strategic withdrawal of effort, self-handicapping, and procrastination (Rhodewalt, 1990; Rothblum, 1990). Therefore, in the case of this study, students who were driven by the fear of performing poorly in the class would only actively engage in posting participations so that they wouldn’t show their incompetence or receive a bad grade in their classes, yet they were less likely to actively engage in non-posting activities (e.g., reading, evaluation, etc.) because such participations were not required, not visible to, nor could be graded by teachers or peers. Further, participating in non-posting activities would not help them to avoid normative judgment. Consequently, they reported less perceived learning in these online classes. The study results revealed a significant impact of beliefs on motivation, therefore suggesting that research should examine the collective impact of beliefs and motivation on learning, instead of studying them in isolation. In addition, self-efficacy beliefs played a critical role in these online classes. They influenced students’ goal adoptions as well as their learning participation. 334 / XIE AND HUANG This study found that self-efficacy beliefs were positively connected to both mastery and performance-approach goals while negatively connected to performance-avoidance goals. The finding is in line with the existing literature, which suggests that the same relationship between self-efficacy and goal orientations holds true in both face-to-face and online learning settings. Further, the SEM model enabled the examination of the relationship between self-efficacy and goal orientations in light of students’ beliefs about fixed ability. Particularly, for those who believed that one’s ability is fixed and effort does not make a difference, if they held high self-efficacy beliefs, then they tended to adopt performance approach goals because they were confident in their capability to outperform others. Conversely, if their perceived competence was low, they tended to adopt performance avoidance goals since they were reluctant to demonstrate their incompetency in front of others. This study also found that self-efficacy positively predicted both posting and non-posting participations. Similar findings can be found in both classroom research (e.g., Elliot & Church, 1997; Greene & Miller, 1996; Pintrich & Garcia, 1991) and online learning research (e.g., Xie, 2013; Xie & Ke, 2011). With online learning analytic data, the present study provided new evidences in supporting such relationships. 7. IMPLICATIONS AND LIMITATIONS This study contributes to the literature by moving beyond the face-to-face classrooms to investigate the relationships among critical constructs of epistemic and learning beliefs, motivation, and students’ participation and perception in authentic online learning settings. Further, this study moved beyond self-reported methods in educational research and incorporated students’ actual participation data in the modeling through analytical methods supported by technology. Based on the findings, we offer a few suggestions for online learning and teaching. First, online instructors should take care to foster appropriate epistemic and learning beliefs to help students develop productive goals for learning. In their interactions with students, instructors should communicate a clear message that learning does not take place quickly, especially in online learning where it takes time and active participation to acquire knowledge and master skills. Instructors should also try to provide affirming feedback to recognize the impact of students’ effort on their performance. Further, online instructors should not assume a definite connection between learners’ naïve epistemic beliefs (e.g., the source of knowledge) and poor motivation and learning participation. Instead, they should recognize that specific learning contexts may lead to different relationships between epistemic beliefs and learning. Second, given the findings that epistemic and learning beliefs influenced students’ online learning participation through their impact on motivation, it may be more efficient for online instructors and instructional designers to develop interventions that foster students’ motivation. Epistemic beliefs of college students are often moderately well established and consequently it is relatively BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING / 335 difficult to intervene and promote changes in their beliefs within a class session, a week, or even a semester (Kienhues, Bromme, & Stahl, 2008; Muis, Bendixen, & Haerle, 2006). On the other hand, students’ motivation can be contextualized and malleable. For example, research suggests that students’ motivation toward academic activities often changes across a semester or academic year (e.g., Stipek & Ryan, 1997; Wigfield & Eccles, 2000; Xie et al., 2006), and motivation generally is specific to learning tasks (e.g., Anderman & Wolters, 2006; Schraw & Lehman, 2001). Hence, while recognizing that students often hold different beliefs about knowledge and learning, instructors can focus on fostering students’ motivation, which may be more efficient and effective in achieving positive impact on student learning. This study has a few limitations. First, in alignment with other similar studies (e.g., Braten & Strømsø, 2005; DeBacker & Crowson, 2006; Kizilgunes et al., 2009), this study adopted the trichotomous framework of achievement goals. Elliot and McGregor (2001) further differentiated mastery goals according to their valence, thereby creating a 2 × 2 achievement goal framework with approach and avoidance dimensions for both mastery and performance goals. With the 2 × 2 framework, a replication of this study might be able to explain the relationship between beliefs and motivation in some learning contexts where mastery approach and avoidance goals were more prominent. Secondly, this study represents an initial exploration to utilize learning analytics to explain learning behaviors. This preliminary exploration only involved the frequencies of posting and non-posting participation as indicators of students’ learning engagement. Future research may involve more dimensions of learning analytics (e.g., sequential information and interaction quality) in order to identify and model learning engagement patterns. Finally, the learning analytics data from online discussion forums could only partially reflect students’ interactions yet did not represent their engagement through face-to-face meetings or offline interactions. Future research should examine the role of beliefs and motivation in both online and offline engagements in asynchronous learning settings. 8. REFERENCES Ames, C. (1992). Classrooms: Goals, structures, and student motivation. Journal of Educational Psychology, 84, 261-271. Anderman, E. M., & Wolters, C. (2006). Goals, values, and affect. In P. Alexander & P. Winne (Eds.), Handbook of educational psychology (2nd ed., pp. 369-390). Mahwah, NJ: Lawrence Erlbaum Associates Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215. Beaudoin, M. F. (2002). Learning or lurking? Tracking the “invisible” online student. Internet and Higher Education, 5(2), 147-155. Bollen, K. A., & Curran, P. J. (2006). Latent curve models: A structural equation approach. Hoboken, NJ: Wiley. 336 / XIE AND HUANG Braten, I., & Strømsø, H. (2004). Epistemological beliefs and implicit theories of intelligence as predictors of achievement goals. Contemporary Educational Psychology, 293, 371-388. Braten, I., & Strømsø, H. (2005). The relationship between epistemological beliefs, implicit theories of intelligence, and self-regulated learning among Norwegian postsecondary students. British Journal of Educational Psychology, 75, 539-565. Buehl, M. M., & Alexander, P. A. (2005). Motivation and performance differences in students’ domain-specific epistemological belief profiles. American Educational Research Journal, 42(4), 697-726. Byrne, B. M. (1998). Structural equation modeling with Lisrel, Prelis, and Simplis: Basic concepts, applications, and programming. Mahwah, NJ: Lawrence Erlbaum Associates. Cano, F. (2005). Epistemological beliefs and approaches to learning: Their change through secondary school and their influence on academic performance. British Journal of Educational Psychology, 75, 203-221. Chen, C. H., & Wu, I. (2012). The interplay between cognitive and motivational variables in a supportive online learning system for secondary physical education. Computers & Education, 58(1), 542-550. Chen, J., & Pajares, F. (2010). Implicit theories of ability of Grade 6 science students: Relation to epistemological beliefs and academic motivation and achievement in science. Contemporary Educational Psychology, 35, 75-87. Cheung, W. S., Hew, K. F., & Ling-Ng, C. S. (2008). Toward an understanding of why students contribute in asynchronous online discussions. Journal of Educational Computing Research, 38(1), 29-50. Church, M., Elliot, A., & Gable, S. (2001). Perceptions of classroom environment, achievement goals, and achievement outcomes. Journal of Educational Psychology, 93, 43-54. Cury, F., Elliot, A., Da Fonseca, D., & Moller, A. C. (2006). The social cognitive model of achievement motivation and the 2 × 2 achievement goal framework. Journal of Personality and Social Psychology, 90(4), 666-679. DeBacker, T., & Crowson, M. (2006). Influences on cognitive engagement: Epistemological beliefs and need for closure. British Journal of Educational Psychology, 76, 535-551. DeBacker, T., Crowson, M., Beesley, A., Thomas, S., & Hestevold, N. (2008). The challenge of measuring epistemic beliefs: An analysis of three self-report instruments. The Journal of Experimental Education, 76(3), 281-312. Dennen, V. P. (2008). Looking for evidence of learning: Assessment and analysis methods for online discourse. Computers in Human Behavior, 24(2), 205-219. Dweck, C. (1986). Motivational processes affecting learning. American Psychologist, 41(10), 1040-1048. Dweck, C. (1999). Self-theories: Their role in motivation, personality, and development. Philadelphia, PA: The Psychology Press. Eccles, J., & Wigfield, A. (2002). Motivational beliefs, values, and goals. Annual Review of Psychology, 53, 109-132. Elliot, A., & Church, M. (1997). A hierarchical model of approach and avoidance achievement motivation. Journal of Personality and Social Psychology, 72(1), 218-232. BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING / 337 Elliot, A., & McGregor, H. (1999). Test anxiety and the hierarchical model of approach and avoidance achievement motivation. Journal of Personality and Social Psychology, 76, 628-644. Elliot, A. J., & McGregor, H. A. (2001). A 2 × 2 achievement goal framework. Journal of Personality and Social Psychology, 80, 501-519. Elliot, A., McGregor, H., & Gable, S. (1999). Achievement goals, study strategies, and exam performance: A mediational analysis. Journal of Educational Psychology, 91(3), 549-563. Elliot, A., & Thrash, T. (2001). Achievement goals and the hierarchical model of achievement motivation. Educational Psychology Review, 13, 139-156. Elliot, E., & Dweck, C. (1988). Goals: An approach to motivation and achievement. Journal of Personality and Social Psychology, 54(1), 5-12. Eppler, M., & Harju, B. (1997). Achievement motivation goals in relation to academic performance in traditional and nontraditional college students. Research in Higher Education, 38(5), 557-573. Ferguson, R. (2012). The state of learning analytics in 2012: A review and future challenges. The Open University, UK: Knowledge Media Institute. Fox, J. (2002). An R and S-PLUS companion to applied regression. Thousand Oaks, CA: Sage. Fredericks, J. A., Blumenfeld, P. C., & Paris, A. H. (2004). School engagement: Potential of the concept, state of the evidence. Review of Educational Research, 74, 59-109. Graham, S., & Golan, S. (1991). Motivational influences on cognition: Task involvement, ego involvement, and depth of information processing. Journal of Educational Psychology, 83, 187-194. Greene, B., & Miller, R. (1996). Influences on achievement: Goals, perceived ability, and cognitive engagement. Contemporary Educational Psychology, 21, 181-192. Greene, B., Miller, R., Crowson, M., Duke, B., & Akey, K. (2004). Predicting high school students’ cognitive engagement and achievement: Contributions of classroom perceptions and motivation. Contemporary Educational Psychology, 29, 462-482. Greene, J. A., Muis, K. R., & Pieschl, S. (2010). The role of epistemic beliefs in students’self self-regulated learning with computer-based learning environments: Conceptual and methodological issues. Educational Psychologist, 45(4), 245-257. Harackiewicz, J., Barron, K., Tauer, J., & Elliot, A. (2002). Predicting success in college: A longitudinal study of achievement goals and ability measures as predictors of interest and performance from freshman year through graduation. Journal of Educational Psychology, 94, 562-575. Hatcher, L. (1994). A step-by-step approach to using the SAS® system for factor analysis and structural equation modeling. Cary, NC.: SAS Institute, Inc. Hofer, B., & Pintrich, P. (1997). The development of epistemological theories: Beliefs about knowledge and knowing and their relation to learning. Review of Educational Research, 67(1), 88-140. Hofer, B., & Pintrich, P. (2002). Personal epistemology: The psychology of beliefs about knowledge and knowing. Mahwah, NJ: Lawrence Erlbaum Associates. Hofer, B., & Sinatra, G. (2010). Epistemology, metacognition, and self-regulation: Musings on an emerging field. Metacognition Learning, 5, 113-120. 338 / XIE AND HUANG Hong, Y.-y., Chiu, C.-y., Dweck, C., Lin, D., & Wan, W. (1999). Implicit theories, attributions, and coping: A meaning system approach. Journal of Personality and Social Psychology, 77(3), 588-599. Kardash, C., & Howell, K. (2000). Effects of epistemological beliefs and topic-specific beliefs on undergraduates’ cognitive and strategic processing of dual-positional text. Journal of Educational Psychology, 92(3), 524-535. Kienhues, D., Bromme, R., & Stahl, E. (2008). Changing epistemological beliefs: The unexpected impact of a short-term intervention. British Journal of Educational Psychology, 78, 545-565. Kizilgunes, B., Tekkaya, C., & Sungur, S. (2009). Modeling the relations among students’ epistemological beliefs, motivation, learning approach, and achievement. The Journal of Educational Research, 102(4), 243-255. Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New York, NY: Guilford. Komarraju, M., & Nadler, D. (2013). Self-efficacy and academic achievement: Why do implicit beliefs, goals, and effort regulation matter? Learning and Individual Differences, 25, 67-72. Kray, L. J., & Haselhuhn, M. P. (2007). Implicit negotiation beliefs and performance: Experimental and longitudinal evidence. Journal of Personality and Social Psychology, 93(1), 49-64. Licht, B., & Dweck, C. (1984). Determinants of academic achievement: The interaction of children’s achievement orientations with skill area. Developmental Psychology, 20, 628-636. Lipponen, L., Hakkarainen, K., & Paavola, S. (2004). Practices and orientations of CSCL. In J.-W. Strijbos, P. A. Kirschner, & R. L. Martens (Eds.), What we know about CSCL: And implementing it in higher education (pp. 31-50). Boston, MA: Kluwer Academic Publishers. Lombard, M., Snyder-Duch, J., & Bracken, C. C. (2002). Content analysis in mass communication: Assessment and reporting of intercoder reliability. Human Communication Research, 28, 587-604. MacCallum, R. C., Browne, M. W., & Sugawara, H., M. (1996). Power analysis and determination of sample size for covariance structure modeling. Psychological Methods, 1(2), 130-149. Midgley, C., Kaplan, A., & Middleton, M. (2001). Performance-approach goals: Good for what, for whom, under what circumstances, and at what cost? Journal of Educational Psychology, 93, 77-86. Muis, K., Bendixen, L. D., & Haerle, F. C. (2006). Domain-generality and domainspecificity in personal epistemology research: Philosophical and empirical reflections in the development of a theoretical framework. Educational Psychology Review, 18, 3-54. Muis, K. R. (2007). The role of epistemic beliefs in self-regulated learning. Educational Psychologist, 42(3), 173-190. Muis, K. R., & Franco, G. M. (2009). Epistemic beliefs: Setting the standards for selfregulated learning. Contemporary Educational Psychology, 34, 306-318. Neber, H., & Schommer-Aikins, M. (2002). Self-regulated science learning with highly gifted students: The role of cognitive, motivational, epistemological, and environmental variables. High Ability Studies, 13, 59-74. BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING / 339 Nicholls, J., Cobb, P., Yackel, E., Wood, T., & Wheatley, G. (1990). Students’ theories of mathematics and their mathematical knowledge: Multiple dimensions of assessment. In G. Kulm (Ed.), Assessing higher order thinking in mathematics (pp. 137-154). Washington, DC: American Association of Advancement in Science. Pajares, F., & Miller, D. (1994). Role of self-efficacy and self-concept beliefs in mathematical problem solving: A path analysis. Journal of Educational Psychology, 86, 193-203. Paulsen, M., & Feldman, K. (1999). Student motivation and epistemological beliefs. New Directions for Teaching and Learning, 78, 17-25. Pintrich, P., & Garcia, T. (1991). Student goal orientation and self-regulation in the college. In M. Maehr & P. Pintrich (Eds.), Advances in motivation and achievement: Goals and self-regulatory processes (Vol. 7; pp. 271-402). Greenwich, CT: JAI Press. Qian, G., & Alvermann, D. (1995). Role of epistemological beliefs and learned helplessness in secondary school students’ learning science concepts from text. Journal of Educational Psychology, 87(2), 282-292. Ravindran, B., Greene, B., & DeBacker, T. (2005). Predicting preservice teachers’ cognitive engagement with goals and epistemological beliefs. The Journal of Educational Research, 98(4), 222-232. Reeve, J. (2006). Extrinsic rewards and inner motivation. In C. M. Evertson & C. S. Weinstein (Eds.), Handbook of classroom management: Research, practice, and contemporary issues (pp. 645-664). Mahwah, NJ: Lawrence Erlbaum Associates. Rhodewalt, F. (1990). Self-handicappers: Individual differences in the preference for anticipatory self-protective acts. In R. Higgins, C. R. Snyder, & S. Berglas (Eds.), Self-handicapping: The paradox that isn’t (pp. 69-106). New York, NY: Plenum. Rothblum, E. D. (1990). The psychodynamic, need achievement, fear of success, and procrastination models. In H. Leitenberg (Ed.), Fear of failure (pp. 497-537). New York, NY: Plenum. Rourke, L., Anderson, T., Garrison, D. R., & Archer, W. (2001). Methodological issues in the content analysis of computer conference transcripts. International Journal of Artificial Intelligence in Education, 12(1), 8-22. Schommer, M. (1990). Effects of beliefs about the nature of knowledge on comprehension. Journal of Educational Psychology, 82(3), 498-504. Schommer, M. (1993). Epistemological development and academic performance among secondary students. Journal of Educational Psychology, 85, 406-411. Schommer, M., Calvert, C., Gariglietti, G., & Bajaj, A. (1997). The development of epistemological beliefs among secondary students: A longitudinal study. Journal of Educational Psychology, 89(1), 37-40. Schommer, M., Crouse, A., & Rhodes, N. (1992). Epistemological beliefs and mathematical text comprehension: Believing it is simple does not make it so. Journal of Educational Psychology, 84(4), 435-443. Schommer-Aikins, M. (2004). Explaining the epistemological belief system: Introducing the embedded systemic model and coordinated research approach. Educational Psychologist, 39(1), 19-29. Schommer-Aikins, M., Duell, O., & Hutter, R. (2005). Epistemological beliefs, mathematical problem-solving beliefs, and academic performance of middle school students. The Elementary School Journal, 105(3), 289-303. 340 / XIE AND HUANG Schommer-Aikins, M., & Easter, M. (2006). Ways of knowing and epistemological beliefs: Combined effect on academic performance. Educational Psychology, 26(3), 411-423. Schraw, G., Bendixen, L. D., & Dunkle, M. E. (2002). Development and validation of the Epistemic Belief Inventory. In B. K. Hofer & P. R. Pintrich (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 103-118). Mahwah, NJ: Lawrence Erlbaum Associates. Schraw, G., & Lehman, S. (2001). Situational interest: A review of the literature and directions for future research. Educational Psychology Review, 13, 23-52. Sins, P., van Joolingen, W., Savelsbergh, E., & van Hout-Wolters, B. (2008). Motivation and performance within a collaborative computer-based modeling task: Relations between students’ achievement goal orientation, self-efficacy, cognitive processing, and achievement. Contemporary Educational Psychology, 33, 58-77. Stathopoulou, C., & Vosniadou, S. (2007). Conceptual change in physics and physicsrelated epistemological beliefs: A relationship under scrutiny. In S. Vosniadou, A. Baltas, & X. Vamvakoussi (Eds.), Reframing the conceptual change approach in learning and instruction. Oxford, UK: Elsevier. Stipek, D., & Ryan, R. (1997). Economically disadvantaged preschoolers: Ready to learn but further to go. Developmental Psychology, 33(4), 711-723. Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th ed.). Boston, MA: Allyn & Bacon. Topping, K., & Ehly, S. (1998). Peer-assisted learning. Mahwah, NJ/London, UK: Lawrence Erlbaum Associates. Tu, Y., Shih, M., & Tsai, C.-C. (2008). Eighth graders’ web searching strategies and outcomes: The role of task types, web experiences and epistemological beliefs. Computers & Education, 51(3), 1142-1153. Utman, C. (1997). Performance effects of motivational states: A meta-analysis. Personality and Social Psychology Review, 1, 170-182. Walker, C., Greene, B., & Mansell, R. (2006). Identification with academics, intrinsic/ extrinsic motivation, and self-efficacy as predictors of cognitive engagement. Learning and Individual Differences, 16, 1-12. Wever, B. D., Schellens, T., Valcke, M., & Keer, H. V. (2006). Content analysis schemes to analyze transcripts of online asynchronous discussion groups: A review. Computers & Education, 46(1), 6-28. Wigfield, A., & Eccles, J. S. (2000). Expectancy-value theory of achievement motivation. Contemporary Educational Psychology, 25, 68-81. Windschitl, M. (1997). Student epistemological beliefs and conceptual change activities: How do pair members affect each other? Journal of Science Education and Technology, 6(1), 37-47. Wise, A. F., Speer, J., Marbouti, F., & Hsiao, Y. (2013). Broadening the notion of participation in online discussions: Examining patterns in learners’ online listening behaviors. Instructional Science. 41(2), 323-343. Wood, P., & Kardash, C. (2002). Critical elements in the design and analysis of studies of epistemology. In B. Hofer & P. Pintrich (Eds.), Personal epistemology: The psychology of beliefs about knowledge and knowing (pp. 231-260). Mahwah, NJ: Lawrence Erlbaum Associates. Wu, D., & Hiltz, S. R. (2004). Predicting learning from asynchronous online discussions. Journal of Asynchronous Learning Networks, 8(2), 139-152. BELIEFS AND MOTIVATION IN ASYNCHRONOUS ONLINE LEARNING / 341 Xie, K. (2013). What do the numbers say? The influence of motivation and peer feedback on students’ behavior in online discussions. British Journal of Educational Technology, 44(2), 288-301. Xie, K., DeBacker, T., & Ferguson, C. (2006). Extending the traditional classroom through online discussion: The role of student motivation. Journal of Educational Computing Research, 34(1), 67-89. Xie, K., Durrington, V. A., & Yen, L. L. (2011). Relationship between students’ motivation and their participation in asynchronous online discussions. Journal of Online Learning and Teaching, 7(1), 17-29. Xie, K., & Ke, F. (2011). The role of students’ motivation in peer-moderated asynchronous online discussions. British Journal of Educational Technology, 42(6), 916-930. Xie, K., Miller, N. C., & Allison, J. R. (2013). Toward a social conflict evolution model: Examining the adverse power of conflictual social interaction in online learning. Computers and Education, 63, 404-415. Xie, K., Yu, C., & Bradshaw, A. C. (2014). Impacts of role assignment and participation in asynchronous discussions in college-level online classes. The Internet and Higher Education, 20, 10-19. Zeldin, A., & Pajares, F. (2000). Against the odds: Self-efficacy beliefs of women in mathematical, scientific, and technological careers. American Educational Research Journal, 37(1), 215-246. Zimmerman, B., & Bandura, A. (1994). Impact of self-regulatory influences on writing course attainment. American Educational Research Journal, 31, 845-862. Direct reprint requests to: Dr. Kui Xie, Ph.D. Department of Educational Studies The Ohio State University 310K Ramseyer Hall 29 West Woodruff Avenue Columbus, OH 43210 e-mail: [email protected]