Individual and group-level factors for students' emotion management in online collaborative groupwork

Internet and Higher Education 19 (2013) 1–9

Contents lists available at ScienceDirect

Internet and Higher Education

Individual and group-level factors for students' emotion management in online collaborative groupwork Jianzhong Xu a,⁎, Jianxia Du b, Xitao Fan b a b

Department of Leadership and Foundations, Mississippi State University, United States Faculty of Education, University of Macau, China

a r t i c l e

i n f o

Article history: Accepted 4 March 2013 Available online 14 March 2013 Keywords: Self-regulation Emotion regulation Online groupwork Collaborative learning Homework

a b s t r a c t The current study examines empirical models of students' emotion management in online collaborative groupwork. Student- and group-level predictors of emotion management in groupwork were analyzed in a survey of 298 graduate students from 86 online study groups in the Southeast of U.S. Results from the multilevel analyses revealed that most of the variance in emotion management occurred at the student level, with help seeking and learning-oriented reasons being the two significant predictors at the group level. Results further revealed that emotion management in groupwork was positively related to feedback, learning-oriented reasons, arranging the environment, monitoring motivation, and help seeking. In addition, compared with part-time students, full-time students were more likely to take initiative in managing their emotion while doing online groupwork. © 2013 Elsevier Inc. All rights reserved.

1. Introduction Emotion is hardly absent from online learning environments (Artino, 2012; Capdeferro & Romero, 2012; Smith, 2008; Wosnitza & Volet, 2005; Zembylas, 2008; Zembylas, Theodorou, & Pavlakis, 2008). Depending on an individual's goals and the characteristics of online activities (e.g., collaborative learning activities), she may enjoy some activities, feel anxious or confused with some aspects of online assignments, fear of losing her individual voice within the group setting, get upset when some group members do not do their share of work, or become frustrated when she could not get timely feedback from her instructor or group members. Largely due to these challenges, students' efforts to manage or influence their emotion become crucial to their learning in online collaborative learning environments. In other words, students' success in online collaborative learning environments is closely related to their efforts to regulate or manage their emotional states to follow through on online collaborative work. These efforts may include up-regulating positive emotions (e.g., to cheer group members up by telling themselves that they can do it), keeping inhibiting emotional states in check (e.g., anxiety and frustration), or down-regulating unpleasant emotions (e.g., to calm each other down and not to get upset with occasional setbacks).

⁎ Corresponding author at: Department of Leadership and Foundations, Mississippi State University, P.O. Box 6037, Mississippi State, MS 39762, United States. Tel.: +1 662 325 2186; fax: +1 662 325 0975. E-mail address: [email protected] (J. Xu). 1096-7516/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.iheduc.2013.03.001

It is intriguing to note, however, that emotion management in online groupwork is noticeably absent from much contemporary research on online collaborative learning (Jarvenoja & Jarvela, 2005). It is equally intriguing to note that the design of collaborative online collaborative learning activities has received little attention, especially on how to help online students deal with emotional challenges. The lack of inquiry in this area is troubling in light of increasing calls to pay attention to students' emotion in online learning environment in general, and with online collaborative learning environment in particular (Volet, Vauras, & Salonen, 2009; Wosnitza & Volet, 2005; Zembylas, 2008). Consequently, there is a critical need to propose and test models of factors that predict students' emotion management while following through online groupwork. This line of research is important, as emotions have a powerful influence on learning, engagement, and achievement in face-to-face classrooms (Op't Eynde, De Corte, & Verschafel, 2007; Pekrun, Elliot, & Maier, 2009) and in online learning environments (Artino, 2012; Zembylas, 2008). This line of research is particularly important in online collaborative learning environments for three reasons. First, online groupwork (compared with face-to-face groupwork) tends to elicit more negative responses from students (Smith et al., 2011; Tutty & Klein, 2008), as it requires increased time and dependence on others, which is in direct conflict with their expectations toward online courses (Piezon & Ferree, 2008). Second, factors specific to online learning environments may further contribute to negative achievement-related emotions (e.g., anxiety and frustration resulting from technical problems and the social isolation of attending classes online; Artino & Jones, 2012). Finally, online collaborative learning environments often create unique challenges for individual and social regulation (e.g., limited social and

2

J. Xu et al. / Internet and Higher Education 19 (2013) 1–9

emotional cues available, and insufficient human scaffolding; Daniels & Stupnisky, 2012; Smith, 2008; Volet et al., 2009). 2. Theoretical framework One theoretical framework relating to emotion management in groupwork is self-regulated learning (Corno, 1993, 2004; Pintrich, 2004; Sansone & Thoman, 2005; Schunk, 2005). Pintrich (2000, 2004), in his model for self-regulated learning in the classroom, has classified four phases of self-regulation (forethought, monitoring, control, and reflection) and, for each phase, four possible areas for self-regulation (cognition, motivation/affect, behavior, and context). In this model, regulation of affect or emotion is explicitly conceptualized as an important aspect of self-regulation. It involves individuals' attempts to control negative affect and anxiety. Pintrich's model further suggests that emotion regulation may be influenced by individuals' attempts to control their own overt behaviors, including time regulation, study environment regulation, and help seeking. This is in line with others' work that cognitive, motivational, behavioral, and contextual factors may interact to influence self-regulation (Eccles & Wigfield, 2002; Schunk, 2005). Emotion regulation is often discussed under the general heading of volition control (Boekaerts & Corno, 2005; Corno, 1993, 2004; Corno & Kanfer, 1993; Husman, McCann, & Crowson, 2000; Kuhl, 1985, 2000). Volitional control focuses on issues of implementation that occur after goals are set and is characterized by the self-regulation activities of purposive striving, including, for example, organizing one's study environment, budgeting time, and regulating motivation and emotion. In his taxonomy of volitional strategies that an individual may use to facilitate the enactment of an intention, Kuhl (1985) discussed the following three strategies, including environmental, motivation, and emotion control strategies. Environmental control involves structuring one's environment to facilitate motivated behavior (e.g., finding a quiet space or asking others to be quiet). Motivation control involves maintaining or strengthening the motivational base of the current behavior when the intention is weak relative to other possible competing intentions. Emotion control involves keeping inhibiting emotional states in check (e.g., stress and frustration). As an individual who strengthens his or her intention to complete a task is more likely to take initiative in coping with unpleasant emotion, it seems logical to hypothesize that emotion control may be positively related to environmental and motivation control. This hypothesis is in line with recent discussion on how motivation can influence affective experiences (Linnenbrink, 2006; Meyer & Turner, 2006). Many researchers further point to the importance of goals in emotion regulation (Diamond & Aspinwall, 2003; Eisenberg & Spinrad, 2004; Larson & Brown, 2007; Op't Eynde & Turner, 2006). For example, Diamond and Aspinwall (2003) noted that “emotion regulation – at all stages of life – cannot be understood without some consideration of what people were trying to do in the situation that elicited the emotion or in which the emotion was experienced” (p. 137). This view is in line with control-value theory of academic emotions (Pekrun, 2000, 2006), which suggests that achievement emotions can be influenced by value-related beliefs that students bring to the learning situation. Other researchers point to the role of time management in emotion regulation (Op't Eynde & Turner, 2006; Schutz, Hong, Cross, & Osbon, 2006). From the perspective of a dynamic, component system theory of emotions, Op't Eynde and Turner (2006) argued for the inclusion of time dimension in emotion regulation, as goals often take extended periods of time to achieve and as emotions often arise in the process due to externally and internally imposed deadlines. Another line of literature suggests certain individual characteristics that may influence emotion regulation. For example, as individuals mature and their effortful control increases with age (Eisenberg & Morris, 2002), they may increasingly learn to make greater use of emotion regulation strategies (Eisenberg & Spinrad, 2004; John &

Gross, 2004). Meanwhile, girls are found to exhibit more effort to regulate their emotion than boys (McRae, Ochsner, Mauss, Gabrieli, & Gross, 2008; Raffaelli, Crockett, & Shen, 2005). In addition, researchers argue that significant others (e.g., teachers and peers) may play an important role in facilitating students' effort to regulate their emotion (Diamond & Aspinwall, 2003; Larson & Brown, 2007). In summary, a self-regulation learning perspective in general, with volitional control in particular, suggests that emotion regulation may be influenced by a range of variables, including background variables, significant others, values and goals, arranging the environment, managing time, monitoring motivation, and help seeking. Consequently, it is important to incorporate these variables in models of emotion management in groupwork. Furthermore, more recent literature on co-regulation and shared regulation suggests that emotion regulation, at both the individual level and group level, is critical for a successful collaboration (Jarvenoja & Jarvela, 2009). Thus, there is need to incorporate a multilevel perspective to differentiate between group- and student-level effects. 3. Emotional issues in online learning environments It is not until recently, emotion issues in online learning environments have started to receive some attention in the literature (Jarvenoja & Jarvela, 2005; Marchand & Gutierrez, 2012; Smith, 2008; Wosnitza & Volet, 2005; Zembylas, 2008). For example, one study by Wosnitza and Volet (2005) examined secondary school and university students' emotions in social online learning, based on self-reported methods and transcripts of interactions. Data revealed that social emotions played an important role in collaborative learning. For example, one university student in an online course commented: I had a great exchange of ideas with X, it was a very good learning experience. But what makes me really angry is that Y appeared only once to an online chat for two minutes and then disappeared for the rest of the unit which raised the workload for the other members of the group (p. 457). The study further implied that factors such as help seeking and teacher feedback may play an important role to deal with their emotions in online learning environments. For example, one student wrote: Dear Teacher, I have managed to get up the courage to have another look at the Message board. It took me 3 days!! but I eventually managed it. Not only am I unfamiliar with message boards etc., I am struggling with a different computer. . . . so I am allowing myself the excuse of unfamiliarity!! Thanks for your prompt reply and support. I do think it is easy to forget what it was like when we start a new experience, but as educators that is a pretty big lesson to remember when teaching our students to take on new and challenging learning. . . . so this is all ‘grist for the mill’ for me. Thanks again, Anne (p. 459). In another related study, Zembylas (2008) investigated how novice adult learners talked about their emotions in the context of a year-long online course. Data revealed that two major themes were positive and negative emotions related to online learning (e.g., joy, enthusiasm, excitement for the flexibility of online learning; and fear, anxiety, alienation, and the need for connectedness). Data further implied that factors such as encouragement and support from the instructors and peers may help to cope with their feelings of loneliness, stress, and anxiety. For example, one student commented: I deeply appreciate the friendly and emotional relationship that has been developed with my instructors and my classmates. Having this emotional support makes me feel more confident about what I am doing. . . Online communication may not be so bad after all; especially, when you receive ongoing encouragement — via

J. Xu et al. / Internet and Higher Education 19 (2013) 1–9

emails, phone calls, face-to-face meetings. I believe all of these together, including the formation of our study groups, create a feeling of not being alone in this journey. This feeling of ‘being with others’ is very important to me (p. 78). In addition, data revealed that women (compared with men) expressed that they had to deal with intense negative feelings about not being able to balance their professional, family, and social life. For example, one woman wrote: On the one hand, the feeling of isolation invokes intense feelings of stress because of the lack of face-to-face communication. On the other hand, I have so many other responsibilities – I am a mother, a wife, a professional [teacher] and a student – I cannot cope with all of those. I feel guilty for paying no attention to my husband and children. I am beginning to wonder if I have made the right choice to enroll in this program (p. 81). Taken together, these recent studies have started to tap into emotional issues in online learning environments. Although emotion regulation was not the primary focus of these studies, relevant findings from these studies implied that emotion management in groupwork may be influenced by a number of factors, including gender, teacher feedback, peer-oriented reasons, and help seeking. Thus, it would be important to incorporate these variables in the current study. 4. The current study The aim of the present study is to propose and empirically examine multilevel models of students' emotion management in online collaborative groupwork. These models differ with respect to the specific predictor variables they include and the level of these variables. Model 1 includes all student-level variables, whereas Model 2 incorporates additional variables at the group level. Specifically, Model 1 consists of eleven student-level variables related to student characteristics (gender, full-time student status, age, and previous online experiences), feedback, peer-oriented reasons, learning-oriented reasons, arranging the environment, managing time, monitoring motivation, and help seeking. It is hypothesized that online group emotion management is positively related to peer-oriented reasons, learning-oriented reasons, feedback, arranging the environment, managing time, monitoring motivation, and help seeking. These hypotheses are in line with a self-regulated learning perspective that emotion management may be influenced by goal setting, feedback, study environment regulation, time regulation, motivation control, and help seeking (e.g., Diamond & Aspinwall, 2003; Kuhl, 1985; Linnenbrink, 2006; Pintrich, 2004) as well as empirical studies that alluded to several factors (e.g., teacher feedback, peer-oriented reasons, and help seeking; Wosnitza & Volet, 2005; Zembylas, 2008) that may influence emotion management in groupwork. In addition, it is hypothesized that emotion management in groupwork is positively related to age, as individuals may increasingly learn to make greater use of emotion regulation strategies (Eisenberg & Spinrad, 2004; John & Gross, 2004). In line with literature on emotion regulation (McRae et al., 2008; Raffaelli et al., 2005), it is further hypothesized that girls are more likely to expend more effort to manage their emotion. Although no prediction is suggested in the literature regarding other student characteristics such as full-time student status and number of previous taken courses, it would be important to control for these variables in the present study. Model 2 incorporated four group-level variables, including feedback, peer-oriented reasons, learning-oriented reasons, and help seeking. The justification for including these variables is (a) that emotion regulation, at both the individual level and group level, is critical for successful collaboration (Jarvenoja & Jarvela, 2009), (b)

3

that the social and academic contexts (e.g., norm and expectation in online groupwork) may influence affective engagement (Corno & Mandinach, 2004), and (c) that socially shared regulation (Rogat & Linnenbrink-Garcia, 2011), with emotion regulation in online learning contexts in particular (Wosnitza & Volet, 2005), may be influenced by the extent to which group members are respectful of each other, get along as a group, and are supportive of each other. 5. Method 5.1. Participants The participants were 298 graduate students from 86 online study groups in one public university in the Southeast of U.S. They were from the same graduate-level course over several semesters from Fall 2009 to Spring 2012. No noticeable difference was found across semesters on the participants' demographic characteristics (e.g., gender, age, and race/ethnicity). Specifically, of the participants in this sample, 44.0% were male and 56.0% were female. The sample was 46.6% Caucasian, 46.3% African American, 3.4% Latinos, 2.7% Asian American, and 1.0% were students from other racial and ethnic backgrounds. Among them, 76.5% were full-time students. In addition, 67.2% were 30 years or younger, and 32.8% were over 30 years old. The course focused on the design and development of multimedia applications through working with multiple authoring and multimedia tools in project-based learning environments. Its topics included the relationship between human learning and multimedia instructional design, instructional design theories and principles, strategies for multimedia instructional design and development, application of instructional design strategies and models, and evaluation of relevant instructional software. 5.2. Online group activities The course was delivered through myCourses, a distance course management system, designed to aid in the delivery and facilitation of online instruction and learning through a variety of communication tools (e.g., emails, discussion boards, and chat rooms). For example, students may use a discussion board (known also as discussion forum, message board, and online forum) as an online “bulletin board” where they can leave and expect to see responses to messages they have left. Online communication media was set up in several areas (e.g., a group discussion area, a whole-class discussion area, and a student/ instructor discussion area, on discussion boards). The instructor also interacted with group members or the entire class in a chat room in predetermined time periods. For the final group project, group members were asked to work together to develop a complete instructional design portfolio project, which involved the selection of an authentic instructional problem, the presentation of an entire evaluative design, and solution for the instructional problem selected. Due to the complexity, interactivity, and collaboration involved in completing this project, students were asked to attend multiple discussion activities with group members by synchronous or asynchronous communication tools (e.g., related to general discussions, debate discussions, panel discussions, and symposium discussions). 5.3. Measures Students were asked about whether they were full-time students (no = 0, yes = 1). They were also asked for the information about the number of previous online courses they had taken: including none (scored 0), one (scored 1), two (scored 2), three (scored 3), and four or more (scored 4). Several multi-item scales were used in the present study. Table 1 presents the item-level details of these scales, and the reliability information for these scales. Some items were adapted from

4

J. Xu et al. / Internet and Higher Education 19 (2013) 1–9

Table 1 Alpha reliability of multi-item scales. Scales

Items

α (CI)

Feedbacka

Coordinated with the group members. Monitored by the group members. Shared with students in other groups. Monitored by the instructor. Given feedback by the instructor. Participating online groupwork brings you approval from group members. Participating online groupwork gives you opportunities to work with group members. Participating online groupwork gives you opportunities to learn from group members. Participating online groupwork helps you support other group members. Participating online groupwork helps you learn multiple media and technology. Participating online groupwork helps you learn interpersonal skills. Participating online groupwork helps you learn communication skills. Participating online groupwork helps you work more productively. Participating online groupwork helps you get a good grade. Locate the materials I need for my online groupwork. Find a quiet area. Remove things from the table. Make enough space for me to work. Turn off the TV. Physically separate myself from my family members or others. Ask my family members or others to be quiet. Set priority. Plan ahead. Keep track of what remains to be done. Pace ourselves to meet the deadline. Remind myself of the available remaining time. Remind my group of the available remaining time. Tell myself to work more quickly when my group lags behind. Tell group to work more quickly when my group lags behind. Find ways to make online groupwork more interesting. Praise my group members for good effort. Praise my group members for good work. Reassure myself that I can do a group project when I feel it is too hard. Reassure my group members that we are able to do a group project when the group members feel it is too hard. I ask the instructor to clarify concepts I don't understand well. When I don't understand the material in this course, I ask another student in my group for help. When I don't understand the material in this course, I ask another student in this class for help. When I don't understand the material in this course, I ask another student who had previously taken this class for help. I try to identify students in my group whom I can ask for help if necessary. I try to identify students in this class whom I can ask for help if necessary. I try to identify online resources where I can get help if necessary. Tell myself not to be bothered with previous mistakes. Tell my group members not to be bothered with previous mistakes. Tell myself to pay attention to what needs to be done. Tell my group members to pay attention to what needs to be done. Tell myself to calm down. Tell my group members to calm down. Cheer myself up by telling myself that I can do it. Cheer my group members up by telling ourselves that we can do it.

.80 (.77, .84)

Peer-oriented reasonsb

Learning-oriented reasonsb

Arranging the environmentc

Managing timec

Monitoring motivationc

Help seekingd

Emotion managementc

.78 (.74, .82)

.83 (.80, .86)

.83 (.80, .86)

.90 (.88, .91)

.86 (.83, .88)

.82 (.79, .85)

.88 (.86, .90)

Note. The 95% confidence intervals for coefficient alpha were calculated using a method employing the central F distribution (see Fan & Thompson, 2001). a Responses were 1 (none), 2 (some), 3 (about half), 4 (most), and 5 (all). b Responses were 1 (strongly disagree), 2 (disagree), 3 (agree), and 4 (strongly agree). c Responses were 1 (never), 2 (rarely), 3 (sometimes), 4 (often), and 5 (routinely). d Responses ranged from 1 (not at all true of me) to 7 (very true of me).

standard instruments (e.g., Xu, 2008b), whereas others were taken from related literature (e.g., Wolters, 2003). 5.3.1. Feedback The development of this scale was informed by the relevant literature (Walberg, Paschal, & Weinstein, 1985; Xu, 2008a). This scale included five items to assess the extent to which teachers and group members provide feedback (α = .80). It measures how much of the assigned online groupwork is shared, discussed, and checked. 5.3.2. Reasons for doing online groupwork Two subscales assessed reasons (peer-oriented reasons, learningoriented reasons) for doing online groupwork, adapted from the recently validated Homework Purpose Scale (Xu, 2010b, 2010c, 2011). Four items measured “peer-oriented reasons” (α = .78), relating to working with, supporting, and seeking approval from group members. Five items

measured “learning-oriented reasons” for doing online groupwork (α = .83), relating to media, communication skills, and work productivity. Confirmatory factor analysis revealed that reasons for online groupwork comprised these two separate yet related two subscales. 5.3.3. Arranging the environment Arranging the environment refers to students' attempt to structure and manage the learning environment (Xu, 2008b, 2008c). The development of this scale was informed by previous research on self-regulation (e.g., Wolters, 2003; Zimmerman & Martinez-Pons, 1990). This scale included seven items, ranging from finding a quiet area for doing online groupwork to minimizing potential distractions (α = .83). 5.3.4. Managing time Managing time refers to students' attempts to plan, monitor, and regulate time use. Its development is informed by the literature on

J. Xu et al. / Internet and Higher Education 19 (2013) 1–9

time management in traditional classroom settings (Eilam & Aharon, 2003; Macan, Shahani, Dipboy, & Philips, 1990; Pintrich, Smith, Garcia, & McKeachie, 1993) and related studies in homework (Muhlenbruck, Cooper, Nye, & Lindsey, 2000; Xu, 2008b; Xu & Corno, 1998). It included eight items to assess student initiative in budgeting time to meet deadlines, ranging from setting priorities and planning ahead to keeping track of available remaining time (α = .90). 5.3.5. Monitoring motivation Monitoring motivation refers to a student's efforts to enhance his or her motivation as well as that of other group members in order to complete online groupwork that might be boring or difficult. The scale of “monitoring motivation” consisted of five items to assess student initiative in managing motivation (α = .86; Xu, 2008b, 2008c), including self-consequating (Graham, Harris, & Troia, 1998; Xu & Corno, 1998, 2003; Zimmerman & Martinez-Pons, 1990), interest enhancement (Sansone, Wiebe, & Morgan, 1999; Wolters, 2003), and efficacy self-talk (McCann & Garcia, 1999; Wolters, 1998). 5.3.6. Help seeking Informed by related items in the Motivated Strategies for Learning Questionnaire (Duncan & McKeachie, 2005; Pintrich et al., 1993), this scale included seven items to assess student initiative to seek social help while doing online groupwork (α = .82), such as getting help from the instructor and other students in the group. 5.3.7. Emotion management Emotion management refers to strategies that students use to regulate their emotional states while doing online groupwork. The development of this scale was informed by literature on emotional development (e.g., Larson & Brown, 2007), theoretical framework on volitional control (e.g., Corno, 1986, 1993; Kuhl, 1985, 1987), as well as empirical studies in homework (e.g., Xu, 2005, 2008a; Xu & Corno, 1998, 2003). This scale consisted of eight items (α = .88), ranging from down-regulating unpleasant emotions (e.g., “Tell my group members not to be bothered with previous mistakes”) to up-regulating positive emotions (e.g., “Cheer my group members up by telling ourselves that we can do it”). 5.4. Statistical analyses Educational researchers are often confronted with data that have multilevel structures. In the case of the present study, individual students were nested under groups, and their individual characteristics are confounded with those of groups. This clustering effect presents several major statistical issues (e.g., aggregation bias, misestimated standard errors, and heterogeneity of regression). These issues cannot be appropriately handled with traditional approaches of regression and analysis of variance. Multilevel modeling allows for the inclusion of variables at multiple levels and takes into account the non-independence of observations by addressing the variability associated with each level of nesting, e.g., decomposing any observed relationship between variables into separate within-group and between-group components (Raudenbush & Bryk, 2002; Snijders & Bosker, 1999). In the present study, multilevel analyses were conducted using the HLM 6 (Raudenbush, Bryk, Cheong, Congdon, & Toit, 2004). To enhance the interpretability of the resultant regression coefficients, we standardized all continuous variables (M = 0, SD = 1) before performing the multilevel analyses. Thus, the regression weights for all variables (except the dummy-coded variables, including gender, age, and full-time student status) are approximately comparable with the standardized weights that result from multiple-regression procedures (Trautwein, Ludtke, Schnyder, & Niggli, 2006; Xu, 2008a). Model 1 included eleven student-level variables regarding student characteristics (gender, age, full-time student status, and the number

5

of previous online courses taken), feedback, peer-oriented reasons, learning-oriented reasons, arranging the environment, managing time, monitoring motivation, and help seeking. Model 2 included four group-level variables, including feedback, peer-oriented reasons, learning-oriented reasons, and help seeking. Feedback within a group was aggregated at the group level to form an index of students' shared feedback at the group level. Similarly, peer-oriented reasons, learning-oriented reasons, and help seeking within a group were aggregated at the group level to form an index of students' shared view relating to these three constructs. The models we used in this study were random-intercept models. The random part of the intercept was freely estimated to reflect between-group differences in emotion management. As we had no a priori hypotheses concerning between-group differences in the predictive power of the predictor variables, we did not use random-slope model in this study; in other words, the slopes of the model were considered fixed, rather than random. The multilevel models implemented in this study are briefly described below. For the outcome of emotion management Yij, the level 1 model for modeling individuals' emotion management can be described as below (using common notations in multi-level analysis, such as those in Raudenbush & Bryk, 2002): Y ij ¼ β0j þ β1j

      S S S X 1 þ β2j X 2 þ : : : þ βkj X 11 þ r ij

where Yij is the emotion management measure for student i nested S represent eleven student-level (indicatunder group j, and X1S to X11 ed by the superscript S) covariates/predictors. For example, students' emotion management Yij could be related to students' previous number of online courses taken (X1S ), and might be affected by other student-level variables (e.g., student arrangement of environment: X2S ). In our analysis, this representative Level 1 model included all eleven student-level covariates/predictors described previously relevant to emotion management outcome Yij. The term rij represents un-modeled individual residual, and β0j represents the group intercept. In the model above, the model intercept (β0j) represents the group level emotion management. For our research interest, the variability among the groups in the model intercept (i.e., β0j) is potentially affected by group-level predictors (i.e., feedback, peeroriented reasons, learning-oriented reasons, and help seeking), as illustrated in the Level 2 model below:         G G G G β0j ¼ γ00 þ γ 01 Z 1 þ γ 02 Z 2 þ γ03 Z 3 þ γ04 Z 4 þ μ 0j where μ0j is the un-modeled group residual, and group-level variables (i.e., feedback, peer-oriented reasons, learning-oriented reasons, and help seeking) are represented by Z1G, Z2G, Z3G, and Z4G (group level indicated by the superscript G). In this model, the effects of group level predictors on the intercept of Level 1 model are captured by the path coefficients γ01 through γ04. The effects (γ01 through γ04) on the Level 1 model's intercept will, in turn, translate into the effects on individual students' emotion management Yij. Full maximum likelihood was used in our model estimation. To disentangle person level and compositional effects (Raudenbush & Bryk, 2002), feedback, peer-oriented reasons, learning-oriented reasons, and help seeking were centered at the group mean. 6. Results Table 2 presents the descriptive statistics relating to the study variables. It also includes zero-order correlations among independent variables and emotion management. Emotion management was found to correlate significantly with all of the independent variables, except gender, full-time student status, and age.

6

J. Xu et al. / Internet and Higher Education 19 (2013) 1–9

Table 2 Descriptive statistics and Pearson correlations. Variables

M

SD

1

1. Gender (female = 0, male = 1) 2. Full-time student (no = 0, yes = 1) 3. Age (30 and below = 0, above 30 = 1) 4. Number of previous online courses 5. Feedback 6. Peer-oriented reasons 7. Learning-oriented reasons 8. Arranging the environment 9. Managing time 10. Monitoring motivation 11. Help seeking 12. Feedback (group) 13. Peer-oriented reasons (group) 14. Learning-oriented reasons (group) 15. Help seeking (group) 16. Emotion management

.44 .77 .33 2.27 3.25 2.97 2.91 3.53 3.88 3.60 4.91 3.28 2.98 2.91 4.91 3.33

.50 .42 .47 1.55 .90 .53 .60 .77 .76 .80 1.04 .61 .35 .33 .63 .77

– .37a −.35a −.29a −.27a −.13b −.10 −.12b −.14b −.16a −.03 −.21a −.08 −.08 −.06 −.08

2

3

4

5

6

7

8

9

10

11

12

13

14

15

– .51a .45a .35a .47a .54a .14b .66a .42a .42a .11 .50a

– .63a .46a .45a .57a .26a .41a .64a .39a .25a .52a

– .36a .37a .58a .23a .35a .34a .55a .13b .50a

– .59a .50a .28a .21a .31a .23a .17a .50a

– .59a .31a .32a .41a .27a .26a .52a

– .37a .44a .44a .42a .20a .60a

– .10 .23a .15a .58a .44a

– .64a .63a .17a .34a

– .61a .39a .40a

– .24a .37a

– .27a

– .50a −.21a −.23a −.13b −.03 −.05 −.10 −.09 −.09 −.25a −.23a −.12b −.14b .01

– .17a .23a .22a .03 .17a .18a .15b −.01 .26a .28a .08 .03 .08

– .27a .20a .23a .20a .26a .34a −.05 .38a .25a .30a −.02 .14b

Note. N varies from 296 to 298. a p b .01. b p b .05.

The fully unconditional model, as shown below, was conducted to partition the variance in emotion management into between-group and within-group components. Level 1: Yij = β0j + rij, Level 2: β0j = γ00 + μ0j. The results indicated that, with 89.4% of the variance in emotion management at the student level, and 10.6% of the variance at the group level, the nested data has ICC of about 0.11. As is widely discussed in the multilevel modeling literature, this level of ICC makes it necessary to use multi-level analysis for controlling the cluster effects of data nesting (e.g., Von Secker, 2002). The deviance statistics and number of estimated parameters for the unconditional model were 840.78 and 3, respectively. Model 1 in our multilevel analysis included eleven student-level variables regarding student characteristics (gender, full-time student status, age, and the number of previous online courses taken), feedback, peer-oriented reasons, learning-oriented reasons, arranging the environment, managing time, monitoring motivation, and help seeking. The deviance statistics and number of estimated parameters for Model 1 were 637.75 and 14, respectively. The likelihood ratio test comparing the unconditional model to Model 1 indicated that Model 1 was a significantly better fit to the data than the unconditional model (χ2(df = 11) = 203.03; p b .001). Model 1 explained 48.9% of the student-level variance in emotion management, and explained 38.0% of the group-level variance (see Table 3). Model 2 included four group-level variables (feedback, peer-oriented reasons, learning-oriented reasons, and help seeking). The deviance statistics and number of estimated parameters for Model 2 were 598.85 and 18, respectively. The likelihood ratio test comparing Model 2 to Model 1 indicated that Model 2 was a significantly better fit to the data than Model 1 (χ2(df = 4) = 38.90, p b .001). Model 2 accounted for an additional .9% of the variance in emotion management at the student level and an additional 60.6% of the variance at the group level. Overall, the final model (Model 2) explained 49.8% of the variance in emotion management at the student level, 98.6% of the variance at the group level, and 55.0% of the total variance. In educational research practice, these would generally be considered as representing very large effect sizes (Fan & Konold, 2010). As shown in Table 3, six student-level variables were found to have statistically significant effects on emotion management. Emotion management was positively related to help seeking (b = .27, p b .01), feedback (b = .24, p b .01), arranging the environment (b = .15, p b .01), monitoring motivation (b = .15, p b .05), and learning-oriented reasons (b = .14, p b .05). In

addition, compared with part-time students, full-time students were more likely to take initiative in managing their emotion while doing online groupwork (b = .33, p b .01). At the group level, emotion management was positively related to learning-oriented reasons (b = .22, p b .05) and help seeking (b = .19, p b .01). On the other hand, the positive effect of feedback and peer-oriented reasons aggregated at the group level did not reach statistical significance.

7. Discussion The present study examined models of students' emotion management in online collaborative groupwork. Results from the multilevel analyses revealed that most of the variance in emotion management occurred at the student level. As most of the variance in emotion management in groupwork occurred at the student level rather than at the Table 3 Emotion management: results from hierarchical linear modeling. Model predictor

Student level Gender (female = 0, male = 1) Full-time student (no = 0, yes = 1) Age (30 and below = 0, over 30 = 1) Number of previous online courses Feedback Peer-oriented reasons Learning-oriented reasons Arranging the environment Managing time Monitoring motivation Help seeking Group level Feedback Peer-oriented reasons Learning-oriented reasons Help seeking 2 R individual level R2 group level R2 total Deviance statistics Number of estimated parameters

Model 1

Model 2

b

SE

b

SE

.03 .21 .07 .00 .19a −.01 .10 .17a .11 .30a .22a

.09 .11 .08 .05 .05 .08 .07 .05 .07 .06 .06

.02 .33a .05 −.03 .24a .02 .14b .15a .09 .15b .27a

.08 .10 .08 .05 .05 .08 .06 .05 .06 .06 .05

.489 .380 .477 637.75 14

.15 .17 .22b .19a .498 .986 .550 598.85 18

.09 .11 .11 .07

Note. N = 298 from 86 groups. b = unstandardized regression coefficient. SE = standard error of b. R2 = amount of explained variance. a p b .01. b p b .05.

J. Xu et al. / Internet and Higher Education 19 (2013) 1–9

group level, emotion management in groupwork was largely a function of individual experiences. Results further revealed that six student-level variables accounted for the variance in emotion management, including full-time student status, feedback, learning-oriented reasons, arranging the environment, monitoring motivation, and help seeking. At the group level, learning-oriented reasons and help seeking were the two significant predictors for between-group variance. With respect to student-level variables, how do we explain the finding that full-time students (compared with part-time students) were more likely to manage their emotion while doing online groupwork? One possible explanation is that full-time students would have committed more to their online learning in terms of the time spent on learning activities than those students with jobs who had less time due to work constraints (Lim & Kim, 2003). Thus, it is likely that full-time students would have committed more to manage their emotions in the groupwork process. Meanwhile, how do we interpret the finding that there was no gender difference emotion management in groupwork? This lack of gender difference is not in line with previous work that girls, compared with boys, exhibit more effort to regulate their emotion (e.g., McRae et al., 2008; Raffaelli et al., 2005; Xu, 2010a). However, none of the above studies were conducted in the context of online learning environments. Previous research suggested that, in the environment of individual online learning (i.e., not collaborative groupwork online learning), female learners were more likely than male learners to be overwhelmed by intense negative emotions resulting from not being able to balance their professional, family, and social life (Zembylas, 2008). The current findings suggest that, in online collaborative groupwork learning environment, this situation could be alleviated, and gender differences in emotion regulation appear to be less pronounced. In addition, the finding that emotion management in groupwork was not related to age was not in line with previous research in this area (Eisenberg & Spinrad, 2004; John & Gross, 2004). One possible explanation is that the age range among graduate students in the present study may be relatively small (with younger students being 30 years or below, and older students ranging from 31 to 60) as compared with participants in previous studies. For example, Gross, Carstensen, Tsai, Skorpen, and Hsu (1997) found that older participants were more likely to control their emotion than younger participants, with younger participants ranging in age from 19 to 56 and older participants ranging in age from 58 to 96. The finding that feedback was positively related to emotion management is in line with the literature on emotion regulation that teachers and peers may play a vital role in supporting students' effort to regulate their emotion in general (Diamond & Aspinwall, 2003; Larson & Brown, 2007), and with homework emotion management in particular (Xu, 2010a). It is also in line with some studies that implied that feedback from teachers and peers (e.g., prompt reply, encouragement, and support) may help students to cope with their emotion (e.g., loneliness, stress, frustration, and anxiety) in online learning environments (Wosnitza & Volet, 2005; Zembylas, 2008). Consequently, it would be beneficial to promote feedback among the instructor and group members in the online groupwork process (e.g., providing ongoing support, acknowledgment, and encouragement). The results from the present study provided empirical support to theoretical claims regarding the importance of goals and intentions in emotion regulation (Eisenberg & Spinrad, 2004; Op't Eynde & Turner, 2006) in online collaborative learning environments, by revealing that emotion management in groupwork was positively related to learning-oriented reasons, but not peer-oriented reasons. It is interesting to note that these results are not in line with the findings from a previous study on secondary homework emotion management (Xu, 2010a), in which homework emotion management was positively related to peer-oriented reasons, but not learning-oriented reasons. One possible explanations is that, compared with secondary school

7

students, adult online learners with higher scores in learning-oriented reasons are more likely to take initiative to manage their emotion, due to school level difference (secondary school students versus graduate students), the difference in the nature of learning tasks (individual face-to-face homework versus online collaborative groupwork), or both. The importance of learning-oriented reasons on emotion management in groupwork for graduate students is further substantiated by the findings that emotion management in groupwork is positively related to learning-oriented reasons at the group level (but unrelated to peer-oriented reasons at the group level). Therefore, online instructors need to pay more attention to how to make online group learning activities more purposeful, meaningful, and relevant. This may include developing high-quality online group learning activities, with particular emphasis on purposes, formats, and types of collaborative activities that will meaningfully engage online students. The findings that arranging the environment, monitoring motivation, and help seeking are related to emotion management in groupwork are consistent with self-regulated learning perspective that emotion management may be influenced by study environment regulation, motivation control, and help seeking (e.g., Diamond & Aspinwall, 2003; Linnenbrink, 2006; Pintrich, 2004). These findings are also in line with the findings from a previous study with secondary school students that homework emotion management was positively related to arranging the homework environment and monitoring homework motivation (Xu, 2010a). They are further in line with empirical findings that implied help seeking may positively influence emotion management in groupwork (e.g., Wosnitza & Volet, 2005). In addition, the present study takes an important step forward, by showing that help seeking in a given group has a positive effect on emotion management in groupwork above and beyond the positive effect of help seeking at the student level. Consequently, online instructors need to encourage their students to take more initiatives (arranging the environment, monitoring motivation, and help seeking) to better manage their emotion in the collaborative learning process. For example, they may want to promote a culture of help seeking, encouraging students to learn how to ask for assistance from multiple sources (e.g., the instructor, peers, and friends) through multiple channels (e.g., email, web chat, and video conferencing) when they confront challenging tasks and perceive the need for help. It is intriguing to note that both self-regulated learning perspective (e.g., Pintrich, 2004) and the previous study with secondary school homework (Xu, 2010a) state that time regulation was positively related to emotion management. However, this was not the case in the present study in the sense that managing time was not related to emotion management in groupwork. One possible explanation for the lack of the association is that the influence of managing time may be mediated by help seeking. To test this hypothesis, we conducted additional analyses by excluding help seeking from Model 1. Indeed, managing time was found to be positively related to emotion management in groupwork, suggesting that managing time may positively influence online emotion management indirectly through help seeking. 8. Limitations and implications for research Unlike face-to-face learning environments, a significant challenge for studying emotions in online learning environments is that neither the emotions nor the process of emotion arousal are readily available for public scrutiny. Thus, the issue of emotion disclosure is paramount in social online learning. However, the potential influences on the disclosure process have been given little attention in research on social online learning (Wosnitza & Volet, 2005). To address this gap in previous research on emotion regulation in online collaborative learning environments, our study has taken an important step forward by examining models of factors that predict emotion management in groupwork through the use of multilevel analyses. However, our study has some limitations that should be acknowledged as well.

8

J. Xu et al. / Internet and Higher Education 19 (2013) 1–9

First, it was based on self-reported data and therefore may be subject to social desirability bias (Duncan & McKeachie, 2005; Wentzel & Wigfield, 2007). In addition, it was based on a cross-sectional survey. Although much care was taken to control for possible confounding variables (informed by research and theorizing on emotion regulation), other predictor variables might have had an effect on emotion management in groupwork had they been included. As this study is the first study that we are aware of to link emotion management in groupwork to a broad spectrum of variables at the student level and group level, further research is needed in other settings (e.g., with a sample of undergraduate students). It would be beneficial to conduct longitudinal studies to better examine how emotion management in groupwork is influenced by a range of variables such as those examined in the present study and possible linkages among these variables (e.g., relationships among managing time, help seeking, and emotion management). Future research would benefit from incorporating multiple data sources (e.g., a diary study, think-aloud protocol measures, and trace logs in online environments) to better understand the antecedents and ongoing dynamic processes of emotion management in groupwork (e.g., how online group members regulate their emotion as it occurs). In addition, it would be informative to conduct qualitative studies to better understand the nature of emotion management in groupwork. For example, it would be particularly revealing to pay attention to students' perspectives about how issues of emotion and emotion management unfold in the online groupwork process, how these issues are influenced by online course design (e.g., nature and structuring of learning activities) as well as other factors in their lives, and what universities can do to help them better manage their emotions in the online collaborative learning environments. Although there are multiple barriers to random assignments in applied settings (Shadish, Cook, & Campbell, 2002), controlled experiments are needed to better address the issue of causation (Pekrun, 2006). For example, there is a need to test the causal hypotheses more directly by experimentally manipulating certain variables (e.g., feedback or help seeking) and by testing the effects of such manipulations on emotion management in groupwork and subsequent desirable learning outcomes. Finally, it would be particularly beneficial to use multiple methodological approaches (as noted above) in cross-cultural settings, as achievement-related emotional experiences (e.g., anxiety) and regulation may be influenced by cultural values (e.g., levels of individualism/collectivism and tolerance of ambiguity; Frenzel, Thrash, Pekrun, & Goetz, 2007; Op't Eynde & Turner, 2006; Pekrun, 2006; Tapanes, Smith, & White, 2009; Volet et al., 2009). References Artino, A. R. (2012). Emotions in online learning environments: Introduction to the special issue. Internet and Higher Education, 15, 137–140. Artino, A. R., & Jones, K. D. (2012). Exploring the complex relations between achievement emotions and self-regulated learning behaviors in online learning. Internet and Higher Education, 15, 170–175. Boekaerts, M., & Corno, L. (2005). Self-regulation in the classroom: A perspective on assessment and intervention. Applied Psychology, 54, 199–231. Capdeferro, N., & Romero, M. (2012). Are online learners frustrated with collaborative learning experiences? International Review of Research in Open and Distance Learning, 13(2), 26–44. Corno, L. (1986). The metacognitive control aspects of self-regulated learning. Contemporary Educational Psychology, 11, 333–346. Corno, L. (1993). The best-laid plans: Modern conceptions of volition and educational research. Educational Researcher, 22(2), 14–22. Corno, L. (2004). Introduction. In work habits and work styles: Volition in education [Special issue]. Teachers College Record, 106, 1669–1694. Corno, L., & Kanfer, R. (1993). The role of volition in learning and performance. In L. Darling-Hammond (Ed.), Review of research in education, vol. 19. (pp. 301–341) Washington, DC: American Educational Research Association. Corno, L., & Mandinach, E. B. (2004). What we have learned about student engagement in the past twenty years. In D. M. McInerney, & S. V. Etten (Eds.), Big theories revisited: Vol 4. Research on sociocultural influences on motivation and learning (pp. 299–328). Greenwich, CT: Information Age. Daniels, L. M., & Stupnisky, R. H. (2012). Not that different in theory: Discussing the control-value theory of emotions in online learning environments. Internet and Higher Education, 15, 222–226.

Diamond, L. M., & Aspinwall, L. G. (2003). Emotion regulation across the life span: An integrative perspective emphasizing self-regulation, positive affect, and dyadic processes. Motivation and Emotion, 27, 125–156. Duncan, T. G., & McKeachie, W. J. (2005). The making of the Motivated Strategies for Learning Questionnaire. Educational Psychologist, 40, 117–128. Eccles, J. S., & Wigfield, A. (2002). Motivational beliefs, values, and goals. Annual Review of Psychology, 53, 109–132. Eilam, B., & Aharon, I. (2003). Students' planning in the process of self-regulated learning. Contemporary Educational Psychology, 28, 304–334. Eisenberg, N., & Morris, A. S. (2002). Children's emotion-related regulation. Advances in Child Development and Behavior, 30, 189–229. Eisenberg, N., & Spinrad, T. L. (2004). Emotion-related regulation: Sharpening the definition. Child Development, 75, 334–339. Fan, X., & Konold, T. R. (2010). Statistical significance versus effect size. In P. Peterson, E. Baker, & B. McGaw (Eds.), International Encyclopedia of Education, 7. (pp. 444–450)Oxford: Elsevier. Fan, X., & Thompson, B. (2001). Confidence intervals about score reliability coefficients, please: An EPM guidelines editorial. Educational and Psychological Measurement, 61, 517–531. Frenzel, A. C., Thrash, T. M., Pekrun, R., & Goetz, T. (2007). Achievement emotions in Germany and China: A cross-cultural validation of the Academic Emotions Questionnaire-Mathematics. Journal of Cross-Cultural Psychology, 38, 302–309. Graham, S., Harris, K., & Troia, G. (1998). Writing and self-regulation: Cases from the self-regulated strategy development model. In D. Schunk, & B. Zimmerman (Eds.), Self-regulated learning: From teaching to self-reflective practice (pp. 20–41). New York: Guilford. Gross, J. J., Carstensen, L. L., Tsai, J., Skorpen, C. G., & Hsu, A. Y. C. (1997). Emotion and aging: Experience, expression, and control. Psychology and Aging, 12, 590–599. Husman, J., McCann, E., & Crowson, H. M. (2000). Volitional strategies and future time perspective: Embracing the complexity of dynamic interactions. International Journal of Educational Research, 33, 777–799. Jarvenoja, H., & Jarvela, S. (2005). How students describe the sources of their emotional and motivational experiences during the learning process: A qualitative approach. Learning and Instruction, 15, 465–480. Jarvenoja, H., & Jarvela, S. (2009). Emotion control in collaborative learning situations: Do students regulate emotions evoked by social challenges? British Journal of Educational Psychology, 79, 463–481. John, O. P., & Gross, J. J. (2004). Healthy and unhealthy emotion regulation: Personality processes, individual differences, and lifespan development. Journal of Personality, 72, 1301–1334. Kuhl, J. (1985). Volitional mediators of cognition-behavior consistency: Self-regulatory processes and action versus state orientation. In J. Kuhl, & J. Beckmann (Eds.), Action control: From cognition to behavior (pp. 101–128). New York, NY: Springer-Verlag. Kuhl, J. (1987). Action control: The maintenance of motivational states. In F. Halisch, & J. Kuhl (Eds.), Motivation, intention and volition (pp. 279–307). Berlin, Germany: Springer-Verlag. Kuhl, J. (2000). The volitional basis of personality systems interaction theory: Applications in learning and treatment contexts. In L. Corno (Ed.), Conceptions of volition: Theoretical investigations and studies of practice. International Journal of Educational Research, 33. (pp. 665–704). Larson, R. W., & Brown, J. R. (2007). Emotional development in adolescence: What can be learned from a high school theater program? Child Development, 78, 1083–1099. Lim, D. H., & Kim, H. (2003). Motivation and learner characteristics affecting online learning and learning application. Journal of Educational Technology Systems, 31, 423–439. Linnenbrink, E. A. (2006). Emotion research in education: Theoretical and methodological perspectives on the integration of affect, motivation, and cognition. Educational Psychology Review, 18, 307–314. Macan, T. H., Shahani, C., Dipboy, R. L., & Philips, A. P. (1990). College students' time management: Correlations with academic performance and stress. Journal of Educational Psychology, 82, 760–768. Marchand, G. C., & Gutierrez, A. P. (2012). The role of emotion in the learning process: Comparisons between online and face-to-face learning settings. Internet and Higher Education, 15, 150–160. McCann, E., & Garcia, T. (1999). Maintaining motivation and regulating emotion: Measuring individual differences in academic volitional strategies. Learning and Individual Differences, 3, 250–279. McRae, K., Ochsner, K. N., Mauss, I. B., Gabrieli, J. J. D., & Gross, J. J. (2008). Gender differences in emotion regulation: An fMRI study of cognitive reappraisal. Group Process and Intergroup Relations, 11, 143–162. Meyer, D. K., & Turner, J. C. (2006). Re-conceptualizing emotion and motivation to learn in classroom contexts. Educational Psychology Review, 18, 377–390. Muhlenbruck, L., Cooper, H., Nye, B., & Lindsey, J. J. (2000). Homework and achievement: Explaining the different strengths of relation at the elementary and secondary school levels. Social Psychology of Education, 3, 295–317. Op't Eynde, P., De Corte, E., & Verschafel, L. (2007). Students' emotions: A key component of self-regulated learning? In P. Schutz, & R. Pekrun (Eds.), Emotions in Education (pp. 185–204). New York: Elsevier. Op't Eynde, P., & Turner, J. E. (2006). Focusing on the complexity of emotion issues in academic learning: A dynamical component systems approach. Educational Psychology Review, 18, 361–376. Pekrun, R. (2000). A social cognitive, control-value theory of achievement emotions. In J. Heckhausen (Ed.), Motivational psychology of human development (pp. 143–163). Oxford, England: Elsevier. Pekrun, R. (2006). The control-value theory of achievement emotions: Assumptions, corollaries, and implications for educational research and practice. Educational Psychology Review, 18, 315–341.

J. Xu et al. / Internet and Higher Education 19 (2013) 1–9 Pekrun, R., Elliot, A. J., & Maier, M. A. (2009). Achievement goals and achievement emotions: Testing a model of their joint relations with academic performance. Journal of Educational Psychology, 101, 115–135. Piezon, S. L., & Ferree, W. D. (2008). Perceptions of social loafing in online learning groups: A study of public university and U.S. naval war college students. International Review of Research in Open and Distance Learning, 9(2), 1–17. Pintrich, P. R. (2000). The role of goal orientation in self-regulated learning. In M. Boekaerts, P. Pintrich, & M. Zeidner (Eds.), Handbook of self-regulation (pp. 451–502). San Diego, CA: Academic Press. Pintrich, P. R. (2004). A conceptual framework for assessing motivation and self-regulated learning in college students. Educational Psychology Review, 16, 385–407. Pintrich, P. R., Smith, D. A. F., Garcia, T., & McKeachie, W. J. (1993). Reliability and predictive validity of the Motivated Strategies For Learning Questionnaire (MSLQ). Educational and Psychological Measurement, 53, 801–813. Raffaelli, M., Crockett, L. J., & Shen, Y. (2005). Developmental stability and change in self-regulation from childhood to adolescence. Journal of Genetic Psychology, 166, 54–75. Raudenbush, S., & Bryk, A. (2002). Hierarchical linear models: Applications and data analysis (2nd ed.). Thousand Oaks, CA: Sage. Raudenbush, S., Bryk, A., Cheong, Y. F., Congdon, R., & Toit, M. (2004). HLM 6: Hierarchical linear and nonlinear modeling. Lincolnwood, IL: Scientific Software International. Rogat, T. K., & Linnenbrink-Garcia, L. (2011). Socially shared regulation in collaborative groups: An analysis of the interplay between quality of social regulation and group processes. Cognition and Instruction, 29, 375–415. Sansone, C., & Thoman, D. B. (2005). Does what we feel affect what we learn? Some answers and new questions. Learning and Instruction, 15, 507–515. Sansone, C., Wiebe, D., & Morgan, C. (1999). Self-regulating interest: The moderating role of hardiness and conscientiousness. Journal of Personality, 67, 701–733. Schunk, D. H. (2005). Self-regulated learning: The educational legacy of Paul R. Pintrich. Educational Psychologist, 40, 85–94. Schutz, P. A., Hong, J. Y., Cross, D. I., & Osbon, J. N. (2006). Reflections on investigating emotion in educational activity settings. Educational Psychology Review, 18, 343–360. Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal influence. Boston, MA: Houghton Mifflin. Smith, R. O. (2008). Adult learning and the emotional self in virtual online contexts. New Directions for Adult and Continuing Education, 120, 35–43. Smith, G. G., Sorensen, C., Gump, A., Heindel, A. J., Caris, M., & Martinez, C. D. (2011). Overcoming student resistance to group work: Online versus face-to-face. Internet and Higher Education, 14, 121–128. Snijders, T. A. B., & Bosker, R. J. (1999). Multilevel analysis: An introduction to basic and advanced multilevel modeling. London: Sage. Tapanes, M. A., Smith, G. G., & White, J. A. (2009). Cultural diversity in online learning: A study of the perceived effects of dissonance in levels of individualism/collectivism and tolerance of ambiguity. Internet and Higher Education, 12, 26–34. Trautwein, U., Ludtke, O., Schnyder, I., & Niggli, A. (2006). Predicting homework effort: Support for a domain-specific, multilevel homework model. Journal of Educational Psychology, 98, 438–456.

9

Tutty, J. L., & Klein, J. D. (2008). Computer-mediated instruction: A comparison of online and face-to-face collaboration. Educational Technology Research and Development, 56, 101–124. Volet, S., Vauras, M., & Salonen, P. (2009). Self- and social regulation in learning contexts: An integrative perspective. Educational Psychologist, 44, 215–226. Von Secker, C. (2002). Effects of inquiry-based teacher practices on science excellence and equity. The Journal of Educational Research, 95, 151–160. Walberg, H. J., Paschal, R. A., & Weinstein, T. (1985). Homework's powerful effects on learning. Educational Leadership, 42, 76–79. Wentzel, K. R., & Wigfield, A. (2007). Motivational interventions that work: Themes and remaining issues. Educational Psychologist, 42, 261–271. Wolters, C. (1998). Self-regulated learning and college students' regulation of motivation. Journal of Educational Psychology, 90, 224–235. Wolters, C. (2003). Regulation of motivation: Evaluating an underemphasized aspect of self-regulated learning. Educational Psychologist, 38, 189–204. Wosnitza, M., & Volet, S. (2005). Origin, direction and impact of emotions in social online learning. Learning and Instruction, 15, 449–464. Xu, J. (2005). Homework emotion management reported by high school students. School Community Journal, 15(2), 21–36. Xu, J. (2008a). Models of secondary school students' interest in homework: A multilevel analysis. American Educational Research Journal, 45, 1180–1205. Xu, J. (2008b). Validation of scores on the homework management scale for high school students. Educational and Psychological Measurement, 68, 304–324. Xu, J. (2008c). Validation of scores on the homework management scale for middle school students. The Elementary School Journal, 109, 82–95. Xu, J. (2010a). Homework emotion management at the secondary school level: Antecedents and homework completion. Teachers College Record, 113, 529–560. Xu, J. (2010b). Homework purpose scale for high school students: A validation study. Educational and Psychological Measurement, 70, 459–476. Xu, J. (2010c). Homework purposes reported by secondary school students: A multilevel analysis. The Journal of Educational Research, 103, 171–182. Xu, J. (2011). Homework purpose scale for middle school students: A validation study. Middle Grades Research Journal, 6(1), 1–14. Xu, J., & Corno, L. (1998). Case studies of families doing third-grade homework. Teachers College Record, 100, 402–436. Xu, J., & Corno, L. (2003). Family help and homework management reported by middle school students. The Elementary School Journal, 103, 503–518. Zembylas, M. (2008). Adult learners' emotion in online learning. Distance Education, 29(1), 71–87. Zembylas, M., Theodorou, M., & Pavlakis, A. (2008). The role of emotions in the experiences of online learning: Challenges and opportunities. Educational Media International, 45, 107–117. Zimmerman, B. J., & Martinez-Pons, M. (1990). Student differences in self-regulated learning: Relating grade, sex, and giftedness to self-efficacy and strategy use. Journal of Educational Psychology, 82, 51–59.