Contemporary Educational Psychology 39 (2014) 134–144
Contents lists available at ScienceDirect
Contemporary Educational Psychology journal homepage: www.elsevier.com/locate/cedpsych
Mediating effects of teacher enthusiasm and peer enthusiasm on students’ interest in the college classroom Taehee Kim a, Diane L. Schallert b,⇑ a b
North Carolina A&T State University, Greensboro, NC, USA University of Texas at Austin, Austin, TX, USA
a r t i c l e
i n f o
Article history: Available online 22 March 2014 Keywords: Interest Teacher enthusiasm Peer influences Affiliative motivation
a b s t r a c t This study explored the mediating effects of students’ perceptions of teacher enthusiasm and of peer enthusiasm on the relation between students’ initial interest and their situational interest at the end of the semester. Students’ motivation for affiliation with their teacher and with peers was also measured to allow for potential associations between these variables and the students’ perceptions of teacher and peer enthusiasm. In path analysis, data from 455 undergraduates showed that perceptions of teacher enthusiasm and of peer enthusiasm had direct associations with two kinds of situational interest, both hold and catch interest. Also, the relations between initial interest and both hold and catch interest were mediated by perceptions of peer enthusiasm but not teacher enthusiasm. Students’ affiliative motivation with the teacher and peers had direct effects on their perceptions of teacher and peer enthusiasm. This study contributes to clarifying the importance of contextual factors as well as students’ own individual variables in understanding the mechanisms by which students’ interest in a course develops and is maintained throughout the semester. Ó 2014 Elsevier Inc. All rights reserved.
0. Introduction Any college course is likely to enroll students who are not interested in the course content and others who are very much intrigued by the course topic, and eager to learn. Another characteristic of college classrooms is that some may have very large enrollment (with over 100, even 500, students), allowing much less interaction between student and instructor and among peers, than other courses. The anonymity that students can sometimes feel in the college classroom can be mitigated or aggravated by whether they are genuinely interested in the material being presented. This interest can either be brought by a student into the classroom, or it can be triggered and enhanced by the social learning environment of the classroom, even to the point that a student can leave a course at the end of the semester with a more developed enduring interest in the topic. Whether it is a K-12 classroom or a college classroom, the importance of interest on learning has been recognized for decades, as attested to by the number of researchers who have studied the role of interest in learning and academic achievement (Alexander,
⇑ Corresponding author. Address: Department of Educational Psychology University of Texas, Austin, TX 78712, USA. Fax: +1 512 471 0288. E-mail addresses:
[email protected] (T. Kim),
[email protected] (D.L. Schallert). http://dx.doi.org/10.1016/j.cedpsych.2014.03.002 0361-476X/Ó 2014 Elsevier Inc. All rights reserved.
Jetton, & Kulikowich, 1995; Alexander & Murphy, 1998; Guthrie, Hoa, Wigfield, Tonks, & Perencevich, 2006; Harp & Mayer, 1997; Renninger & Hidi, 2002). Although there have been several studies that have focused on personal factors that influence the interest students bring into a course, Harackiewicz et al. (2008) concluded a recent, and already much-cited, study by stating that more research was needed on classroom environmental factors associated with students’ situational interest. Even though there have been some studies of the types of tasks and characteristics of activities that could, in the words of Mitchell (1993), ‘‘catch’’ and ‘‘hold’’ students’ interest in the classroom (e.g., Schraw, Flowerday, & Lehman, 2001), there has been relatively less research on social aspects of college learning environments associated with student interest. For instance, we could find no research on how characteristics of one’s teacher and classmates may influence one’s interest throughout a semester, even though contextual support for interest has been acknowledged as an important contributor to interest development (Hidi & Renninger, 2006). Other theorists such as Deci (1992) and Krapp (2002) have pointed to the critical role that interrelationships among persons, activities, and contexts can play in the development of interest by way of interpersonal experiences that could influence the quality of interactions between a person and an object of interest. Additionally, one’s social group and the social norms in one’s environment can influence the degree to which one values an object of interest.
T. Kim, D.L. Schallert / Contemporary Educational Psychology 39 (2014) 134–144
This was exactly the focus of the study we conducted. Compared to the amount of theoretical support for the importance of social and contextual factors on interest, there has been little empirical research on establishing the degree of influence different factors play in predicting interest development, particularly for college students. Nearly two decades ago, Mitchell (1993) mentioned that teachers may not have power over students’ initial interest but can influence throughout instruction the continued interest of the students they are teaching. Thus, we selected teacher enthusiasm as one potential factor that may be associated with influencing student interest in a course because several studies showed that high levels of teacher enthusiasm seemed to exert a positive influence on students’ motivation (Frenzel, Goetz, Ludtke, Pekrun, & Sutton, 2009; Patrick, Hisley, & Kempler, 2000; Wild, Enzle, & Hawkins, 1992) as well as on students’ learning (Brigham, Scruggs, & Mastropieri, 1992; McKinney, Robertson, Gilmore, Ford, & Larkins, 1984). However, the connection between teacher enthusiasm and student interest, particularly the more sophisticated understanding of interest reflected in the recent literature, had not been previously investigated. Also, because the social environment in a course is as much about the other students taking a class as the teacher, we included peer enthusiasm as a factor. Along with teachers’ influence on students’ motivation and learning, there are findings that suggest the role that peers can play on individual students’ motivation. One recent study reported a moderate correlation between peers’ attitude toward learning and students’ motivation to learn (Khamis, Dukmak, & Elhoweris, 2008). Also, Sage and Kinderman (1999) reported a positive association between individual students’ engagement in class and the engagement level of their peers in the classroom, as well as an association between peer support and students’ engagement and discipline problems at school. Finally, because we saw social and contextual factors as interacting inherently with individuals’ own sociohistorical proclivities, we added a measure of affiliative motivation to capture individual differences in students’ sensitivity to teacher and peer messages, in addition to using measures of teacher and peer enthusiasm that relied on the individual students’ perceptions. We chose affiliative motivation because it has been reported as being associated with behavioral mimicry (Lakin & Chartrand, 2003) and described by Sinclair, Lowery, Hardin, and Colangelo (2005) as the ‘‘social tuning of attitude’’ supporting the role of affiliative motivation as an underlying mechanism of convergence of attitudes. These findings pointed to the importance of affiliative motivation in understanding emotional transmission, and the effects of students’ perceptions of teacher enthusiasm and peer enthusiasm. Thus, we saw affiliative motivation as potentially helpful in explaining why some students become more enthusiastic than others despite having the same teacher and the same classmates in the same course. In what follows, we review the literature on each of the key variables of our study. 1. Interest Most commonly seen as a motivational variable, interest has been defined as ‘‘a psychological state of engaging or reengaging with particular content such as objects, events, or ideas over time’’ (Hidi & Renninger, 2006, p. 112). Since the early work of Hidi (1990), educational researchers have generally differentiated between two types of interest: situational and individual interest (e.g., Harackiewicz et al., 2008; Hidi, 2006). Situational interest has been defined as ‘‘environmentally triggered, involve[ing] an affective reaction and focused attention’’ (Hidi, 2006, p. 72), echoing the words of Schiefele (1991) who described situational interest as ‘‘an emotional state brought about by situational stimuli’’ (p. 302). The definitions of situational interest highlight the importance of characteristics of the immediate
135
environment with which learners interact. Using qualitative and quantitative methodologies, Mitchell (1993) modeled situational interest as involving catching and holding interest, claiming that catching interest is related to at least two kinds of immediate stimulation, cognitive and sensory, whereas holding interest refers to empowering students to maintain interest by making the content of learning meaningful. In a recent model of how interest develops, Hidi and Renninger (2006) reiterated the importance of situational interest in the development of long-term individual interest, stimulated by environmental conditions and maintained by such factors as perceived value in the activity. Individual interest has been defined as ‘‘a person’s relatively enduring predisposition to reengage particular content over time as well as to the immediate psychological state when this predisposition has been activated’’ (Hidi & Renninger, 2006, p. 113). Mitchell (1993) identified individual interest as an interest that a person already possesses and brings to a context. As with situational interest, individual interest has been found to have a positive influence on learning processes such as attention, recognition, and recall (Renninger & Wozniak, 1985), and on academic motivation (Scheifele, 2001). The most recent theoretical model of interest is Hidi and Renninger’s (2006) four-phase model of interest development (see Fig. 1). In this model, situational and individual interest intertwine and evolve in consecutive phases labeled triggered situational interest, maintained situational interest, emerging individual interest, and well-developed interest, with each phase differentiated by the degree of influence of three components, affect, value, and cognitive processes. In early phases, affect is expected to play a relatively more important role whereas in later phases, perceived value and knowledge may be more important. Guided by Hidi and Renninger’s model, Linnenbrink-Garcia et al. (2010) studied differences between triggered situational interest, maintained situational interest, and individual interest while validating measures of situational interest in academic domains. Linnenbrink-Garcia et al.’s findings indicated that triggered situational interest is different from maintained situational interest in terms of the types of affective reactions engendered because triggered situational interest includes affective reactions to the way instructors present the material whereas maintained situational interest includes affective reaction to the material itself. Also, these authors reported that maintained situational interest could be further divided into value-related and feeling-related maintained situational interest. Overall, Linnenbrink-Garcia et al.’s findings pointed to the importance of value and affect that could be stimulated by environmental factors in developing learners’ situational interest. Recognizing the importance of environmental factors on students’ interest, Durik and Harackiewicz (2007) explored how conditions in a learning environment could influence learners’ situational interest. In their study, they reported that arousing conditions such as bright colors and pictures meant to stimulate catch interest had positive effects on the task interest of learners with low individual interest whereas those with high individual interest were more influenced by manipulating the personal utility of a task, stimulating their hold interest. Overall, they reported that individuals who already had interest in a domain were more receptive to indicators of the value they may gain from the task. Similarly, Paige (2011) explored how environmental factors could influence students’ interest in reading. He reported a significant association between extrinsic motivation for reading and oral reading proficiency, suggesting that struggling adolescent readers could benefit from a teacher’s and classmates’ modeling of engagement in reading. Our focus for this study of teacher enthusiasm involved exploring the degree of relation we could find between a potentially affect-arousing and valuing variable, students’ perceptions of teacher’s and peers’ enthusiasm, and students’ situational interest.
136
T. Kim, D.L. Schallert / Contemporary Educational Psychology 39 (2014) 134–144
Fig. 1. Hidi and Renninger’s (2006) four-phase model of interest development.
2. Teacher enthusiasm How a teacher influences students’ learning has been a longstanding question in educational research. As one among several cornerstones of teaching, teacher enthusiasm has been recognized as an important attribute of an effective teacher as well as a predictor of student outcomes (Brophy & Good, 1986), and seemed a likely influence on students’ interest development, because of its affective nature. As Meyer and Turner (2006) reported, a positive correlation can be expected between students’ emotions and motivation in the classroom and teachers’ expressions of positive emotions, intrinsic motivation, and emotional support. Teacher enthusiasm has been considered a part of a teacher’s arsenal, a strategy to influence students’ performance by displaying a high energy level and interest in the subject matter (Sass, 1989). Although conceptions of what is meant by teacher enthusiasm continue to need further specification, Kunter, Frenzel, Nagy, Baumert, and Pekrun (2011) most recently addressed this need by capturing the perspectives of both teachers and students. Their results indicated that teacher enthusiasm could be differentiated into two dimensions, enthusiasm for teaching and subject matter enthusiasm. They reported that teaching enthusiasm was more likely to be influenced by students’ motivation and behaviors in the classroom whereas subject matter enthusiasm seemed independent of students’ characteristics and responses. In the past few decades, studies on the effect of teacher enthusiasm have focused on whether level of teacher enthusiasm influences student outcomes by way of encouraging student behaviors such as engagement in classroom tasks (Bettencourt, Gillett, Gall, & Hull, 1983; Brigham et al., 1992; McKinney et al., 1984; Sanders & Gosenpud, 1986). Although most of this research has been conducted in pre-K-12 settings, findings are nevertheless suggestive of the kind of associations one might expect for the college classroom. For example, Frenzel, Goetz, Ludtke, Pekrun, and Sutton (2009) investigated the link between teacher enjoyment and student enjoyment, and reported a positive relation between 7th and 8th grade teachers’ emotions and their students’ emotions. In the study, Frenzel and colleagues used students’ perceptions of their teachers’ enthusiasm, arguing that teacher enthusiasm is an observable variable by which students can recognize their teacher’s enjoyment of classroom teaching. Thus, in testing for such a connection, it is students’ perceptions that matter, we would argue, because students can only be influenced by their teacher’s enjoyment and enthusiasm if they can detect their teacher’s positive affect toward teaching. This is because it is difficult to know a teacher’s actual level of enjoyment if the teacher does not express it in class. Therefore, student-perceived teacher enthusiasm was our main focus in this exploration of how teacher enthusiasm could influence college students’ own affect toward and valuing of a course.
3. Peer influences Most studies on peer influence have been focused on younger students rather than on college-age students, perhaps because young students are thought to be especially susceptible to peer group influence (Brown, Eicher, & Petrie, 1986). In the large classrooms common in higher education, it is often not easy to build close relationships with others. Thus, it seemed important to
investigate how individual students perceived their classmates’ enthusiasm collectively and whether there were associations between such perceptions and individual students’ motivation. Although there is no literature that can directly inform this aspect of our study, the work on peer relationships and classroom climate, with its emphasis on the importance of students’ perceptions of their peers, seemed relevant as indirect support for the link between students’ perceptions of their peers’ enthusiasm for a course and their own interest development. In considering influential factors for student learning, Meyer and Turner (2006) emphasized the importance of a positive climate that teachers and students co-create in the classroom. Along with teachers’ influence on students’ motivation and learning, peers have been shown to be significant contributors to creating a social environment conducive for learning. In a study of the influence of the peer group, Ryan (2001) reported that peers influenced younger adolescent students’ perceptions of their enjoyment of school and their achievement. Although peers co-create norms and values regarding school, each individual student may form different perceptions of these norms and values emerging from their interactions in peer group contexts (Ryan, 2000). Khamis et al. (2008) reported a moderate correlation between peers’ attitude toward learning and students’ own motivation to learn, and Sage and Kinderman (1999) reported a positive relation between individual students’ engagement in class and the engagement level of their classmates. Also, Nelson and DeBacker (2008) showed that a negative peer climate could be especially harmful for students who have performance-approach, performance-avoidance, and socialapproval goals, all achievement motivation goals that relate to students’ concerns about how they look to others. These studies encouraged us to investigate the influence of students’ perceptions of their classmates’ motivation and their relationships with their classmates on their own interest and motivation. 4. Affiliative motivation In general, affiliative motivation has been defined as motivation to affiliate, be connected, and maintain good relationships with others (Hill & Werner, 2006; Huntsinger, Lun, Sinclair, & Clore, 2009), and it has been explored as a way of elucidating why individuals act or think like those with whom they interact. For example, Huntsinger et al. (2009) suggested that affiliative motivation may lead an individual to match in mood and affect the people with whom he or she interacts, as in the social tuning of attitude that Sinclair et al. (2005) investigated. Lakin and Chartrand (2003) reported a positive association between affiliative motivation and behavioral mimicry. In applying the construct to a general educational context, Hill and Werner (2006) studied students’ affiliative motivation in relation to aggressive behavior and school attachment, recognizing the importance of individual differences in explaining students’ adjustment. In their findings, they reported that students with higher affiliative orientation had higher school attachment, and they also found a mediating effect of school attachment on the connection between affiliative orientation and aggressive behaviors. All of these findings support the importance of affiliative motivation in understanding emotional transmission, and may be helpful as well in understanding the effects of students’ perceptions of teacher and peer enthusiasm.
137
T. Kim, D.L. Schallert / Contemporary Educational Psychology 39 (2014) 134–144
Although it has not, as yet, been studied for its effects in actual classroom contexts, affiliative motivation would seem relevant in helping explain underlying mechanisms by which some students become more enthusiastic than others with the same teacher and the same classmates in the same course. Because teacher enthusiasm has been considered an affective component that accompanies intrinsic motivation to teach (Kunter et al., 2008), and because it is expressed as a high energy level and signs of interest in the subject matter one is teaching (Sass, 1989), it may be that students with high affiliative motivation are more likely to recognize and also experience their teacher’s and peers’ enjoyment and positive feelings. Therefore, affiliative motivation may help explain why some students become more enthusiastic about a course than others. 5. The current study The purpose of this study was to explore the kinds of contextual factors associated with college students’ interest in course content. We could find no research that had investigated the relations between students’ perceptions of their teacher and peers’ enthusiasm and their own interest. Also, no previous research had included how students’ own motivation to affiliate with others in the classroom influenced their perceptions of teacher enthusiasm and of peer enthusiasm, or how affiliative motivation was related to students’ situational interest at the end of the semester. It was these gaps in the literature that this study was intended to fill, guided by the following research questions: (a) To what degree are students’ perceptions of their teacher’s enthusiasm and peers’ enthusiasm associated with students’ interest at the end of the semester?; (b) Is initial interest associated with situational interest (catch and hold interest) indirectly via students’ perceptions of their teacher’s and peers’ enthusiasm?; (c) Does students’ motivation for affiliation with the teacher predict their perceptions of teacher enthusiasm, and does students’ motivation for affiliation with peers predict their perceptions of peer enthusiasm?; and (d) Is motivation for affiliation with the teacher associated with catch interest via perceptions of teacher enthusiasm and via perceptions of peer enthusiasm? 6. Method 6.1. Methodological issues To address our research questions, it was important that the data come from several classes of students and that the classes represent different subject matter intended for different majors and for students at different points in their studies. We needed to include different instructors because we needed variance in the construct of teacher enthusiasm. Although we expected variation among students in perceptions of enthusiasm within any one class, we also expected these perceptions to be influenced by the teacher himself or herself. Thus, to ensure variation in the variables, a necessity when using any correlational technique, it was imperative to collect data from as many different classes and instructors as possible. Also, because teachers from one discipline may adopt similar ways of allowing their emotions or attitudes to show while teaching, we needed to represent several disciplines from across campus in the data set. A final methodological issue involves the question of whether the teachers should have been identified by outside criteria as representing a range of enthusiasm by, for example, asking department heads to nominate instructors who were high and low on enthusiasm. Similarly, we might have asked for nominees representative of a range of effectiveness in teaching. Because this study was a first exploration of these constructs, we chose instead simply
to ask for as many volunteer instructors as would agree to participate. On the large campus where the study was conducted, there were many possibilities for recruitment. We began by perusing the course catalogue, looking for courses at different levels (from introductory to more advanced) that were known to enroll substantial numbers of students (as well as smaller classes) in a wide variety of disciplines across all the large colleges on campus (Business, Education, Engineering, Natural Sciences, and Liberal Arts). We then contacted the instructors of particular courses listed in the course schedule. It is perhaps the nature of a volunteer sample that the instructors who agreed to participate had on average course instructor ratings of 4.48 on a 5-point scale as compared to the university-wide mean of 4.2. Note that there was no item on the course instructor survey that directly asked about the instructor’s level of enthusiasm. 6.2. Participants Once IRB approval of the study had been secured, participants were recruited from the students enrolled in 12 different courses representing various disciplines at a large research-intensive university in the southwest of the United States. These students were recruited for the study only after their instructor had given permission for us to collect data in their class. Permission was sought and granted via a mass e-mail message sent a few weeks before the semester was to begin to 65 professors teaching different kinds of courses. The request stated that we wanted to capture students’ initial interest in the course before the instructor had even introduced the syllabus and that we would need to come into the class twice, once at the very beginning and once near the very end of class. Only 12 professors responded after second and third requests, eight who were teaching lower division courses and four upper division courses (see Table 1 for descriptive characteristics of the 12 courses, which were all different courses with no repeated sections of the same course; we differentiated two psychology courses by adding a I and II rather than actual course names to protect the anonymity of participants). We first established with each professor which of their courses would be included in the study, only one per instructor. We then met with each class to explain the study’s purpose as we passed out the IRB-approved consent form. We repeated our study explanation at both Time 1 and Time 2. By the time of the second data collection, many students had dropped or added the course, and thus, pretest data were collected on 746 students, and posttest data were collected on 691 students.
Table 1 Characteristics of the Courses. Courses
Division
n
Class structurea
Psychology I Management Marketing Finance Asian studies History Basic biology Religion Psychology II Ed. Psychology Chemistry Biology
Lower Upper Upper Upper Lower Lower Lower Lower Lower Upper Lower Lower
44 31 19 9 25 106 46 21 47 50 24 33
Lecture-based Mixture of both Mixture of both Lecture-based Mixture of both Lecture-based Mixture of both Mixture of both Lecture-based Mixture of both Mixture of both Mixture of both
Required Yes
No
30 31 19 8 16 100 41 17 38 41 23 33
14 0 0 1 9 3 5 4 9 1 0 0
CISb rating
4.3 4.6 4.0 4.5 4.3 4.6 4.7 4.5 4.3 4.7 4.8 4.5
a Students were asked to indicate their perceptions of the structure of their course, whether it was lecture-based, discussion-based, or a mixture of both. b CIS = course instructor survey. We used the reported mean that each instructor received for the item asking students to respond to the following question: ‘‘Overall, this instructor was very (1) unsatisfactory, (2) unsatisfactory, (3) satisfactory, (4) very good, (5) excellent’’.
138
T. Kim, D.L. Schallert / Contemporary Educational Psychology 39 (2014) 134–144
However, the number of students who completed both pre and post surveys was only 512. Of these, 455 students who had participated in both pre and post surveys remained after data screening for outliers (42 cases) and for incomplete responses (15 cases). In terms of outliers, we used the multivariate outlier approach to identify cases to be removed, an approach in SPSS that notes cases that have extreme scores on more than one variable using the Mahalanobis distance indicator. According to conservative levels of statistical significance, a p value of less than .001 is recommended for a score to be considered an outlier (Kline, 2005), a criterion that was met by 42 cases. In a check of the effect of these outlier cases, we compared the fit indices of the path model with the 42 cases included with the model in which the cases were removed. Fit indices became much worse when the 42 cases were included (CFI = .965, RMSEA = .091 with a 90% confidence interval of 0.060–0.124, SRMR = .066), even though the CFI still met criterion. Therefore, we chose to exclude these 42 outlier cases in our final test of model fit. Note that the CFI (Comparative Fit Index) measures the relative improvement in fit of a researcher’s model compared with a baseline model (Kline, 2005). In general, a CFI greater than around 0.90 suggests reasonably good fit of the researcher’s model (Hu & Bentler, 1999). Because it measures the error of approximation, the RMSEA (Root Mean Square Error of Approximation) is interpreted as ‘‘badness-of-fit,’’ as values that are closer to zero are taken to indicate better fit whereas higher values indicate poor fit (Kline, 2005). The rule of thumb for the RMSEA is that a value of 0.05 or less indicates close fit, values between 0.05 and 0.08 indicate reasonable error, and poor fit is indicated if the RMSEA is 0.10 or more. The SRMR (the Standardized Root Mean Square Residual) measures the overall difference between the observed and predicted correlations. In general, it is considered favorable if the SRMR is less than 0.10. As for the 15 cases that represented missing data, we began by excluding the data of five students who had missed one or more items on scales that had only three items. We then compared the fit of the model with the remaining 10 cases included (N = 465) versus with the 10 cases excluded (N = 455). Results of the path analysis showed slightly worse fit with these 10 cases included (CFI = .971, RMSEA = .071 with 90% confidence interval of 0.039– 0.107, SRMR = .027), even though all fit indices still met criterion. In the end, we decided to use the results with 455 participants in our final model. The number of participants across the 12 classes ranged from 9 to 106. Among the 455 participants, 59.3% (270) were women and 40.7% (185) were men. In terms of their year in school, 15.2% (69) were freshman, 42.4% (193) sophomore, 22.9% (104) junior, and 18.2% (83) senior. Also, 10.1% (46) of the students were taking the course even though it was not required, whereas for 89.5% (407) of the students, the course fulfilled a requirement in their degree plan. In terms of students’ perceptions of the class structure, 241 (53%) students perceived their class to be lecture-based, 19 (4.2%) students considered their class discussion-based, and 195 (42.9%) students considered their class to be a mix of lecture and discussion. 6.3. Procedure Having received permission from 12 instructors, we arranged for as early a visit to the actual classroom as the instructor would allow. For ten of the classes, this meant meeting the students and describing the purpose of the study on the very first day, even before the instructor had explained the course or syllabus. For the other two classes, at the instructors’ request, we conducted the survey on the second day of class. Having distributed consent forms and questionnaires, we allowed students to ask any question they had before starting. If a student agreed to participate, he or
she signed the consent form and then proceeded to the three measures for Time 1: demographic information, initial interest in the course, and background knowledge of course content. All students completed the surveys in fewer than 10 min. For Time 2, 11 of 12 classes allowed us to return for survey administration during the last week of the semester. The instructor of the 12th class requested we administer the survey on the Monday of the penultimate week of the semester. Instructions to the students began by reminding them that the surveys we were distributing were part of the same study in which they had participated at the very beginning of the semester. Students who could not remember participating at Time 1 but who were willing to participate at this time were asked to sign the consent form and complete questionnaires. These students’ data were not included in this report. The second set of surveys included measures of students’ catch interest and hold interest, their perceptions of their teacher’s enthusiasm, their perceptions of their classmates’ enthusiasm, their motivation for affiliation with their teacher, and their motivation for affiliation with their classmates. The second data gathering took at most 20 min. We had asked students to provide their student identification number in order to allow us to match questionnaires from Time 1 to Time 2. 6.4. Measures At Time 1, students filled out three measures: initial interest, background knowledge, and demographic information. At Time 2, they completed six measures: two measures of situational interest (catch interest and hold interest), two measures of perceptions of enthusiasm (one with the teacher as focus and the other with classmates), and two of motivation for affiliation (one with the teacher and the other with classmates). All measures used a 7-point Likert scale, ranging from ‘‘1’’ (strongly disagree) to ‘‘7’’ (strongly agree). Items included some negatively worded statements that were reverse scored so that a high score on a scale indicated a positive rating for that construct. 6.4.1. Interest: initial and situational interest The measures of initial interest (Time 1) and situational interest (Time 2) were adapted from the interest questionnaire used by Harackiewicz et al. (2008). The interest measure at Time 1 included seven items assessing initial interest in the course content (e.g., ‘‘I’ve always been fascinated by this course content’’). For this scale, Harackiewicz et al. (2008) reported Cronbach’s alpha of .90 (a = .90 in the current study). At Time 2, students’ situational interest in the course was measured in two ways, with catch interest focusing on ‘‘students’ affective reactions to course lectures’’ and hold interest focusing on ‘‘students’ feelings about and personal valuing of the course material’’ (Harackiewicz et al., 2008, p. 109). The catch interest scale was composed of five items, and the hold interest scale was composed of nine items. Both measures included negatively worded questions that were reverse scored. Harackiewicz et al. (2008) reported Cronbach’s alpha of .85 for catch interest and .91 for hold interest (a = .85, a = .92, respectively, in the current study). 6.4.2. Background knowledge The measure of background knowledge (Time 1) was adapted from a questionnaire used by Harackiewicz et al. (2008). The measure includes three items (e.g., ‘‘I studied related topics in another class or did readings on my own’’). Harackiewicz et al. (2008) reported a Cronbach alpha of .90 (a = .79 in the current study).
T. Kim, D.L. Schallert / Contemporary Educational Psychology 39 (2014) 134–144
6.4.3. Students’ perceptions of teacher enthusiasm The measure of students’ perceptions of teacher enthusiasm (Time 2) was adapted from a measure of teacher enthusiasm by Frenzel, Goetz, Ludtke, and Pekrun (2009). The measure included four items measuring students’ perceptions of their teacher’s enthusiasm (e.g., ‘‘The teacher of this class teaches with enthusiasm’’). Frenzel et al. (2009) reported Cronbach’s alpha of .85 for the enthusiasm measure (a = .84 for the current study). 6.4.4. Students’ perceptions of peer enthusiasm Because there was no currently available measure of students’ perceptions of peer enthusiasm, we adapted and modified a measure from a study by Patrick (1995) of students’ perceptions of teacher enthusiasm. The questionnaire (Time 2) included four items measuring students’ perceptions of their classmates’ enthusiasm (e.g., ‘‘Overall, the other students in this class really seem to be excited about learning’’). The Cronbach’s alpha we obtained was .84. 6.4.5. Motivation for affiliation with the teacher The measure of students’ motivation for affiliation with their teacher (Time 2) was adapted and modified from a measure of affiliative motivation (Huntsinger, Lun, Sinclair, & Clore, 2009). This measure included three items (e.g., ‘‘I want to get along with the teacher of this class’’). Huntsinger et al. (2009) reported Cronbach’s alpha of .82 for the measure (a = .89 in the current study). 6.4.6. Motivation for affiliation with peers Like the measure of students’ motivation for affiliation with their teacher, the measure of students’ motivation for affiliation with their classmates (Time 2) was adapted and modified from the original measure of affiliative motivation developed by Huntsinger et al. (2009). This measure was composed of three items (e.g., ‘‘I want to cooperate with fellow classmates in this class’’). The Cronbach’s alpha for the measure in the current study was .89. 6.5. Data analysis First, descriptive statistics were calculated to characterize the sample, including information on students’ gender, year in school, class structure, and whether the class was required or not in the student’s program of work. Second, correlations among the eight main variables were also calculated. Third, a path analysis was conducted in order to explore overall direct and indirect relations among variables. Even though the model with all 455 participants as one group showed good fit, the ICC, the intraclass correlation suggested significant differences among classes in these data, therefore necessitating the use of cluster variables in the analysis. Therefore, a path analysis with cluster variables using complex type option in Mplus (Version 6) was used in order to correct the standard errors due to the nested structure of the data. 7. Results 7.1. Preliminary analyses Before conducting the main analysis, we checked for normality of responses to the measures and for multicollinearity, and we calculated descriptive statistics. Normality was checked using skew and kurtosis indices. The sign of the skew index indicates its direction, and a value close to 0 implies that a score distribution is nearly symmetrical around the mean, whereas variables with
139
absolute values of more than 3 indicate extreme skewness (Kline, 2005). At the scale level, skew values were within acceptable levels ranging from 0.24 to 0.99. At the item level, only one item (on the teacher enthusiasm scale) approached a scale value close to unacceptable levels, at 2.39, but it was still within the acceptable range. As for kurtosis, a conservative rule states that values greater than 10 may cause problems in one’s interpretation of results, and values greater than 20 may cause serious problems (Kline, 2005). Again, all subscales were quite low in their kurtosis index, ranging from .05 to .62. The same item on the teacher enthusiasm scale was the only item with kurtosis, at 8.21, approaching criterion. Thus, guided by the literature, we judged all subscales to meet the normality criteria necessary for subsequent analyses. Multicollinearity occurs when variables that are supposed to measure different constructs are apparently measuring the same construct, as shown by intercorrelations greater than .85 (Kline, 2005). To check for multicollinearity, we first examined the correlation matrix with all items. In general, even though items on each scale showed relatively high correlations with each other, there was only one correlation higher than criterion, the correlation between two items on the hold interest scale, which was .89. We then tested collinearity statistics, using as acceptance criteria tolerance values of more than .10 and values on the variance inflation factor (VIF) of less than 10 (Kline, 2005). In our data, there were no items that had tolerance levels of less than .10 or a VIF higher than 10. Also, the tolerance levels for the scales themselves ranged from .43 to .91, with VIF values from 1.09 to 2.33. For a model to be properly identified, the number of observations should be equal to or more than the number of parameters. To calculate the number of observations, one uses the number of variables, in our case 8, and then multiplies this number by the number plus 1 and divides the total by 2: 8 (8 + 1)/2 = 36 (see Kline, 2005). Because the number of parameters estimated with the data was 27, the model degrees of freedom, which is the difference between the number of observations and the number of parameters, equaled 9, indicating that the model was appropriately over-identified. In terms of number of cases, Kline (2005) recommends that the ratio of number of cases to the number of parameters should be more than 5:1, and the preferred ratio is more than 10:1. In this study, the number of cases was 455, so the ratio was more than 16:1. Finally, in order to decide whether data should be treated as clustered, ICC (intraclass correlation) was calculated. Intraclass correlation is the proportion of variance in a variable due to grouping effects, indicating between group variance, q(ICC) = MSb MSw/ MSb + (c 1) MSw). For our data, the ICC was 0.218, indicating that about 22% of the variance could be explained by the grouping variable. Because this was more than 10%, it was important to analyze the data as clustered when running path analyses. Next, we calculated descriptive statistics on the scales (see Table 2). The means of students’ hold interest and catch interest measured at the end of the semester were slightly higher than their initial interest. In terms of perceptions of enthusiasm, the mean for teacher enthusiasm was more than one standard deviation higher than that of peer enthusiasm. Like perceptions of teacher and peer enthusiasm, the mean for motivation for affiliation with the teacher was more than one standard deviation higher than that of motivation for affiliation with peers. Finally, we calculated simple bivariate correlations among the eight measured variables (see Table 2). Results indicated that students’ initial interest was significantly correlated with five variables, hold interest, catch interest, perceptions of peer enthusiasm, motivation for affiliation with peers, and motivation for affiliation with the teacher, but not with background knowledge and perceptions of teacher enthusiasm. Students’ hold interest was significantly correlated with all variables but one, including
140
T. Kim, D.L. Schallert / Contemporary Educational Psychology 39 (2014) 134–144
Table 2 Means (and SDs) of eight measures and bivariate correlations among variables. Mean (SD) 1. 2. 3. 4. 5. 6. 7. 8. *
Initial interest Background knowledge Motivation for affiliation with the teacher Motivation for affiliation with peers Perceptions of teacher enthusiasm Perceptions of peer enthusiasm Hold interest Catch interest
5.12 4.82 6.01 4.89 6.01 4.41 5.27 5.12
(1.12) (1.73) (1.00) (1.45) (.91) (1.91) (1.11) (1.27)
1
2 .01 .22* . 25* .06 .28* .47* .23*
.08 .17* .02 .20* .09 .02
3
4
5
6
7
.60* .54* .35* .42* .35*
.30* .50* .31* .20*
.32* .41* .57*
.40* .46*
.64*
p < .01.
catch interest, motivation for affiliation with peers, motivation for affiliation with the teacher, perceptions of peer enthusiasm, and perceptions of teacher enthusiasm, but not with background knowledge. Like hold interest, students’ catch interest showed significant correlations with all variables except background knowledge. Motivation for affiliation with peers had significant associations with perceptions of peer enthusiasm, motivation for affiliation with the teacher, perceptions of teacher enthusiasm, and background knowledge. Perceptions of peer enthusiasm also had significant correlations with motivation for affiliation with the teacher, perceptions of teacher enthusiasm, and background knowledge. Motivation for affiliation with the teacher had a significant association with all variables except background knowledge. Finally, perceptions of teacher enthusiasm was significantly correlated with six variables but was not associated with initial interest and background knowledge. Interestingly, perceptions of peer enthusiasm and motivation for affiliation with peers were significantly correlated with all eight variables, though the correlation with background knowledge was negative. In sum, it was interesting that motivation for affiliation with peers and motivation for affiliation with the teacher were significantly correlated (r = .60) as were perceptions of teacher enthusiasm and peer enthusiasm (r = .32). Whereas hold interest and catch interest at the end of the semester were significantly correlated (r = .64), there was surprisingly no correlation between initial interest and background knowledge (r = .01).
hold and catch interest. Also, there were mediation effects of perceptions of teacher and peer enthusiasm on the path from motivation for affiliation with teacher and peers to catch interest. However, even though we had predicted an association between initial interest and background knowledge based on the literature, our data in this study did not show a significant relation between these two variables. Next, we discuss the details of direct and indirect effects in the model (see Table 3 for a list of the direct effects). The direct effects of initial interest on hold interest (b = 0.39, p < 0.001) as well as on catch interest (b = 0.14, p < 0.001) were significant. However, there was no direct relation between initial interest and perceptions of teacher enthusiasm (b = 0.07, p = 0.072) even though the direct effects of perceptions of teacher enthusiasm on hold interest (b = 0.32, p < 0.001) and on catch interest (b = 0.56, p < 0.001) were significant. Therefore, the indirect effect from initial interest to hold interest via perceptions of teacher enthusiasm
7.2. Path analysis As the main data analysis technique to address our research questions, we used path analysis using Mplus (version 6) because path analysis is appropriate when investigating direct effects, indirect effects, covariance between observed exogenous variables, and the disturbance correlation between endogenous variables (Kline, 2005). In this study, taking into consideration the clustered nature of the data, we used path analysis with complex type option and cluster variable in Mplus. The hypothesized model with eight observed variables was tested to examine direct relations among initial interest, motivation for affiliation with teacher and peers, perceptions of teacher and peer enthusiasm, and situational interest at the end of the semester along with the mediation effects of perceptions of teacher and peer enthusiasm on the association between initial interest and situational interest and between motivation for affiliation and situational interest (see Fig. 2). Results of the final model (see Fig. 3) indicated good fit with the data, CFI = 0.991, RMSEA = .081 with 90% confidence interval of 0.048–0.116, SRMR = 0.027. As we had hypothesized, initial interest and motivation for affiliation with the teacher and with peers predicted perceptions of teacher and peer enthusiasm. In turn, perceptions of teacher and peer enthusiasm predicted end-of-semester hold as well as catch interest. Also, perceptions of peer enthusiasm showed a mediation effect on the path from initial interest to situational interest, both
Fig. 2. Hypothesized model. INT = initial interest; BK = background knowledge; AFT = motivation for affiliation with the teacher; AFP = motivation for affiliation with peers; ENT = perceived teacher enthusiasm; ENP = perceived peer enthusiasm; HOLD = hold interest; CATCH = catch interest.
Fig. 3. Final path model. INT = initial interest; BK = background knowledge; AFT = motivation for affiliation with the teacher; AFP = motivation for affiliation with peers; ENT = perceived teacher enthusiasm; ENP = perceived peer enthusiasm; HOLD = hold interest; CATCH = catch interest.
141
T. Kim, D.L. Schallert / Contemporary Educational Psychology 39 (2014) 134–144
(b = 0.02, p = 0.07) and the indirect effect from initial interest to catch interest via perceptions of teacher enthusiasm (b = 0.04, p = 0.084) were not supported. By contrast, there was a direct effect of initial interest on perceptions of peer enthusiasm (b = 0.16, p < 0.001). Also, there were direct effects of perceptions of peer enthusiasm on hold interest (b = 0.22, p < 0.001) and on catch interest (b = 0.34, p = 0.001). As hypothesized, the indirect effect from initial interest to hold interest via perceptions of peer enthusiasm (b = 0.04, p = 0.01) and the indirect effect from initial interest to catch interest via perceptions of peer enthusiasm (b = 0.06, p = 0.01) were supported. There was a negative direct effect of motivation for affiliation with peers on catch interest (b = 0.15, p < 0.05) whereas there was no direct relation between motivation for affiliation with the teacher and catch interest (b = 0.00, p = .99). The direct effects of motivation for affiliation with peers on perceptions of peer enthusiasm (b = 0.46, p < 0.001) and of motivation for affiliation with the teacher on perceptions of teacher enthusiasm (b = 0.56, p < 0.001) were significant. An indirect effect from motivation for affiliation with the teacher to catch interest via perceptions of teacher enthusiasm (b = 0.31, p < 0.001) as well as an indirect effect from motivation for affiliation with peers to catch interest via perceptions of peer enthusiasm (b = 0.16, p < 0.001) were supported. 8. Discussion The purpose of this study was to investigate how students’ perceptions of teacher enthusiasm and of peer enthusiasm as contextual variables could help explain the relation between students’ initial interest and their situational interest, as comprised of both hold and catch interest, at the end of the semester. An additional purpose was to explore the association between students’ affiliative motivation for both their teacher and classmates and several key variables such as students’ perceptions of teacher enthusiasm and of peer enthusiasm as well as their end-of-semester catch interest. Overall, the results of this study supported the importance of contextual variables, in the form of teacher and peer enthusiasm, as well as students’ own individual variables, as represented by their initial interest and affiliation motivation in the classroom, in explaining their interest in a course at the end of the semester. In what follows, we begin by discussing some of the limitations that should be kept in mind in interpreting our results before moving to a consideration of the complex intertwined nature of these variables and closing with implications for practice and for future research. 8.1. Constraints on interpretations Even though the results of this study showed promising implications for understanding how student interest, teacher enthusiasm, peer enthusiasm, and affiliative motivation are related, it is important to acknowledge some key aspects that limit
the generalizability of our findings. First, our data were limited to survey responses provided by students about their perceptions of their teacher’s and classmates’ enthusiasm and about their own interest and affiliative motivation. Without corroboration from observation or interview data or from teacher-provided input, we are relying solely on what students reported about their classroom experiences. Nevertheless, the data sources we chose may be quite apt, as students’ own perceptions may be more critical to their motivation in a class than any outside perspective. Another concern is that all professors who participated in this study were generally seen by the students as enthusiastic teachers, even though there was enough variance to yield significant predictions. In addition, the volunteer instructors were generally rated as effective teachers with a mean of 4.48 on a 5-point scale on the course instructor survey results retrieved from public records after the semester was over. Future studies that would include teachers rated much lower on the enthusiasm scale or much less effective might yield interestingly different results. In addition, our measure of peer enthusiasm was one we had constructed simply by adapting a teacher enthusiasm scale, and its psychometric characteristics still need further development. Finally, we only had 12 classes participating in the study, a much lower number than the 20 clusters preferred in analyses with cluster variables, and results need to be interpreted accordingly. Nevertheless, keeping these limitations in mind, we move next to a discussion of the primary findings of the study. 8.2. Connecting contextual and individual variables to students’ situational interest In current formulations of the construct of interest, situational interest is considered critical in interest development because when students develop situational interest, they may come to see value in the object of interest, maintain their interest over time, and eventually develop enduring individual interest (Hidi & Renninger, 2006; Hulleman, Durik, Schweigert, & Harackiewicz, 2008). Using these conceptions of interest as a starting point, we explored how the contextual variables of teacher and peer enthusiasm and the individual variables of initial interest and affiliation motivation would be associated with students’ end-of-semester situational interest. Path analysis results indicated that end-of-semester hold and catch interest (two aspects of situational interest) were predicted directly by initial interest, perceptions of teacher enthusiasm, and perceptions of peer enthusiasm. Other direct effects included the connection between background knowledge and hold interest, between motivation for affiliation with peers and catch interest, between motivation for affiliation with the teacher and perceptions of teacher enthusiasm, and between motivation for affiliation with peers and perceptions of peer enthusiasm. Taking each type of situational interest in turn, we begin with hold interest, described by Harackiewicz et al. (2008) as ‘‘students’ feelings about and personal valuing of the course material’’
Table 3 Direct (D) effects of variables. Predictor variables
INT BK AFT AFP ENT ENP
Predicted variables Perceived teacher enthusiasm (ENT)
Perceived peer enthusiasm (ENP)
Hold interest
Catch interest
No direct effect
D = 0.16
D = 0.39 D = 0.14
D = 0.14
D = 0.56 D = 0.46 D = 0.32 D = 0.21
No direct effect D = .15 D = 0.56 D = 0.34
Note. INT = initial interest; BK = background knowledge; AFT = motivation for affiliation with the teacher; AFP = motivation for affiliation with peers; ENT = perceived teacher enthusiasm; ENP = perceived peer enthusiasm.
142
T. Kim, D.L. Schallert / Contemporary Educational Psychology 39 (2014) 134–144
(p. 109). Previous work had led us to expect that students’ initial interest would naturally be a strong predictor of end-of-semester hold interest, as would background knowledge (e.g., Alexander & Murphy, 1998; Alexander, Sperl, Buehl, & Fives, 2004). That is, if students come into a course with high levels of initial interest, they may already place some value on learning course concepts, and these evaluations of the value of the course may be captured by a measure of hold interest at the end of the semester. Similarly, if students have some background knowledge relevant to the course, it is reasonable to expect that they may value the course content and want to build on their knowledge in the area. Our results supported these predictions from previous research in that both of these individual difference variables measured at the beginning of the semester were significant predictors of hold interest at semester’s end. In this study, the more interesting results were that perceptions of teacher enthusiasm and peer enthusiasm had independent direct effects on students’ hold interest. In fact, perceptions of teacher enthusiasm were almost as strong a direct predictor as initial interest on students’ hold interest. Thus, if we extrapolate from what was measured with the scales of perceptions of teacher and peer enthusiasm, results suggest that expressions of positive emotions toward the content and a sense of positive energy in the classroom from both teacher and peers may convey a sense of value for the course content to students. Teacher enthusiasm especially may contribute to students’ valuing of the course. In terms of direct effects on catch interest, the other form of situational interest measured at the end of the semester, we found that perceptions of teacher enthusiasm was a strong direct predictor, more so than perceptions of peer enthusiasm and initial interest, which also had significant direct connections to catch interest. As operationalized in our study, the concept of catch interest involved affective reactions to course content, in line with the conceptualization of the construct by Harackiewicz et al. (2008). Thus, our results indicated that teacher enthusiasm and peer enthusiasm had stronger direct effects on catch interest than did initial interest, implying that even though students’ initial interest, an individual variable, is important in understanding catch interest, teacher enthusiasm and peer enthusiasm as contextual variables play key roles in the development of students’ catch interest in the classroom. Another individual variable we measured, students’ motivation for affiliation with the teacher, did not have any impact on their catch interest, and their motivation for affiliation with peers even had a negative direct effect on their catch interest, suggesting that if students care highly about having good relationships with classmates, their attention may be diverted from classroom activities, and their catch interest may be negatively affected. In terms of mediating effects, even though perceptions of teacher enthusiasm did not mediate the relation between initial interest and either hold or catch interest, perceptions of peer enthusiasm did so. Such results may suggest that when students have initial interest and in turn, perceive their classmates as enthusiastic, they are more likely to value the course content at the end of the semester, whereas teacher enthusiasm by itself independently affects students’ hold interest as does their initial interest. Also, perceptions of teacher enthusiasm did not have mediator effects on the relation between initial interest and catch interest, but still alone, affected catch interest, even more so than initial interest. Thus, both variables, teacher enthusiasm and initial interest, may have different characteristics that independently influence catch interest. On the other hand, perceptions of peer enthusiasm mediated the relation between initial interest and catch interest. Even though students’ own initial interest had relatively less effect on catch interest compared to contextual variables, if students
perceived their classmates as being enthusiastic, they reported higher level of catch interest at the end of the semester. Also, as mediators, perceptions of teacher enthusiasm and perceptions of peer enthusiasm mediated the relation between affiliative motivation and catch interest. Specifically, even though affiliative motivation itself had no effect or even a negative effect on catch interest, if students wanted to have good relationships with others in the classroom, and if they perceived the teacher and their classmates as enthusiastic in the classroom, then, their catch interest seemed to have been affected by their affiliative motivation indirectly. Because affiliative motivation is indirectly related to situational interest, it may be important that the teacher consider what to do to create a classroom climate for students to become more involved so that their affiliative motivation may increase and so they may perceive their classmates as more enthusiastic. The direct connection between motivation for affiliation with teachers and perceptions of teacher and peer enthusiasm supports the idea that students who are willing to get along with their teachers are more likely to perceive both their teachers and classmates as enthusiastic in the classroom. As predicted, students’ motivation for affiliation with their teacher was strongly related to their perceptions of teacher enthusiasm. The suggestion from this finding is that when students are willing to get along with their instructor, they seem to perceive their instructors as more enthusiastic. Also, students’ motivation for affiliation with peers was strongly connected to perceptions of peer enthusiasm. In line with previous work on affiliative motivation, it is possible that students who have more affiliative motivation may be more sensitive to the feelings of their teacher and classmates, and this may include an influence from their teacher’s and classmates positive expressions of affective reactions to and valuing for course content. In sum, perceptions of teacher enthusiasm and perceptions of peer enthusiasm were strong predictors of students’ hold and catch interest at the end of the semester. Furthermore, teacher enthusiasm and peer enthusiasm seemed to influence students’ catch interest more than their hold interest. Because catch interest emphasizes ‘‘reactions to the course’’ rather than ‘‘value toward the course content’’ it seems reasonable that students’ catch interest would more likely be influenced by others’ enthusiasm in the classroom. Also, even though both contextual variables were significantly important in explaining students’ interest in the course, students’ situational interest was more associated with their teacher’ enthusiasm than with their classmates’ enthusiasm. Also, as the findings of this study showed, even though students’ initial interest was connected to their situational interest at the end of the semester, contextual variables in the form of perceptions of teacher enthusiasm and peer enthusiasm were also significantly associated with situational interest. As Hidi and Renninger’s (2006) four-phase model of interest describes, such increase in situational interest is key to enhancing long-lasting individual interest in the future. Furthermore, the relations among individual student variables were mediated by contextual variables. Thus, we suggest that contextual variables should be recognized for their importance concurrent with individual variables in future research of the construct of interest. 8.3. Implications and suggestions for future research In terms of teachers’ role in students’ interest development, previous work has explored teacher effects on student interest via choice of instructional strategies such as types of tasks (Mitchell, 1993) and types of texts (e.g. Schraw, Flowerday, & Lehman, 2001). By contrast, we investigated in this study the contextual variables of teacher enthusiasm and peer enthusiasm and their
T. Kim, D.L. Schallert / Contemporary Educational Psychology 39 (2014) 134–144
connections to student interest. The finding that teacher enthusiasm was associated with situational interest is noteworthy in that teacher enthusiasm was as powerful as students’ initial interest in predicting students’ end-of-semester hold and catch interest. Thus, one implication of this study is that teacher enthusiasm should be recognized for its potential to contribute to students’ appreciation of the value of a course as well as their experience of positive emotions in the classroom. Even though as a semester goes on, students might come to appreciate a reserved, non-expressive teacher’s interest in her or his teaching, still, this study points to the importance of a teacher’s expressed positive emotions in class in helping students perceive their teacher’s enthusiasm. A worthwhile focus of future research would be to measure directly the connection between teacher displays of enthusiasm and students’ positive valuing of a course. As for peer enthusiasm, the findings of this study may contribute to the literature on peer influence and classroom climate by demonstrating the connection between classmates’ enthusiasm on students’ situational interest. Along with its direct effect on students’ situational interest, peer enthusiasm seems to help explain the relation between initial interest and situational interest. Also, when students seemed more concerned about having good relationships with their classmates, their catch interest seemed negatively impacted but peer enthusiasm mediated this process and was positively associated with students’ catch interest. That perceptions of peer enthusiasm had a direct association with situational interest may suggest that time spent creating a positive classroom environment for students may be well spent. As a final point, we want to acknowledge the need in future research to connect the constructs of enthusiasm, interest, and learning. In this study, we did not tackle the challenging issues associated with measuring learning directly. On the one hand, we would argue that interest is an important outcome in its own right as it may sustain engagement with a topic and investment in effortful actions regardless of immediate knowledge gains. On the other hand, we believe that measuring learning, capturing it in any effective way, is a very difficult task. As Alexander, Schallert, and Reynolds (2009) discussed at length, the construct of learning is complex. It is made particularly complicated when attempting to use a psychometrically equivalent measure across several courses on different subjects. Also, learning implies demonstrating a change in knowledge state, a pretest to post-test difference, something that is rarely available in ‘‘real’’ college classrooms. Nevertheless, if only to establish definitively the connection between interest and learning, both as a process and as an outcome, it would seem important in future research to look for ways to overcome the challenges inherent in measuring learning in situations of high and low interest. In conclusion, contributing to previous work that has recognized the importance of social support and of the environment in interest development (Harackiewicz et al., 2008; Hidi & Renninger, 2006), the findings of this study highlighted the interplay between students’ own academic and social motivation in the classroom and their perceptions of emotional support from their teacher and classmates to help explain students’ interest at the end of the semester.
References Alexander, P., Jetton, T., & Kulikowich, J. (1995). Interrelationship of knowledge, interest, and recall: Assessing a model of domain learning. Journal of Educational Psychology, 87, 559–575. Alexander, P., & Murphy, P. (1998). Profiling the differences in students’ knowledge, interest, and strategic processing. Journal of Educational Psychology, 90, 435–447. Alexander, P. A., Schallert, D. L., & Reynolds, R. E. (2009). What is learning anyway? A topographical perspective considered. Educational Psychologist, 44, 176–192.
143
Alexander, P., Sperl, C. T., Buehl, M. M., & Fives, H. (2004). Modeling domain learning: Profiles from the field of special education. Journal of Educational Psychology, 96, 545–557. Bettencourt, E., Gillett, M., Gall, M., & Hull, R. (1983). Effects of teacher enthusiasm training on student on-task behavior and achievement. American Educational Research Journal, 20, 435–450. Brigham, F., Scruggs, T., & Mastropieri, M. (1992). Teacher enthusiasm in learning disabilities classrooms: Effects on learning and behavior. Learning Disabilities Research & Practice, 7, 68–73. Brophy, J., & Good, T. (1986). Teacher behavior and student achievement. In M. Wittrock (Ed.), Handbook of research on teaching (pp. 328–375). New York: Macmillan. Brown, B. B., Eicher, S. A., & Petrie, S. (1986). The importance of peer group (‘‘crowd’’) affiliation in adolescence. Journal of Adolescence, 9, 73–96. Deci, E. (1992). The relation of interest to the motivation of behavior: A selfdetermination theory perspective. In K. Renninger, S. Hidi, & A. Krapp (Eds.), The role of interest in learning and development (pp. 43–70). Hillsdale, NJ: Erlbaum. Durik, A. M., & Harackiewicz, J. M. (2007). Different strokes for different folks: How individual interest moderates the effects of situational factors on task interest. Journal of Educational Psychology, 99, 597–610. Frenzel, A. C., Goetz, T., Ludtke, O., Pekrun, R., & Sutton, R. E. (2009). Emotional transmission in the classroom: Exploring the relationship between teacher and student enjoyment. Journal of Educational Psychology, 101, 705–716. Guthrie, J. T., Hoa, L. W., Wigfield, A., Tonks, S. M., & Perencevich, K. C. (2006). From spark to fire: Can situational reading interest lead to long-term reading motivation? Reading Research and Instruction, 45(2), 91–117. Harackiewicz, J. M., Durik, A. M., Barron, K. E., Linnenbrink-Garcia, L., & Tauer, J. M. (2008). The role of achievement goals in the development of interest: Reciprocal relations between achievement goals, interest, and performance. Journal of Educational Psychology, 100, 105–122. Harp, S., & Mayer, R. (1997). The role of interest in learning from scientific text and illustrations: On the distinction between emotional interest and cognitive interest. Journal of Educational Psychology, 89(1), 92–102. Hidi, S. (1990). Interest and its contribution as a mental resource for learning. Review of Educational Research, 60, 549–571. Hidi, S. (2006). Interest: A unique motivational variable. Educational Research Review, 1, 69–82. Hidi, S., & Renninger, K. (2006). The four-phase model of interest development. Educational Psychologist, 41, 111–127. Hill, L. G., & Werner, N. E. (2006). Affiliative motivation, school attachment, and aggression in school. Psychology in the Schools, 43, 231–246. Hu, L.-T., & Bentler, P. M. (1999). Cutoff criteria for fit indices in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55. Hulleman, S., Durik, A., Schweigert, S., & Harackiewicz, J. M. (2008). Task values, achievement goals, and interest. Journal of Educational Psychology, 100, 398–416. Huntsinger, J. R., Lun, J., Sinclair, S., & Clore, G. (2009). Contagion without contact: Anticipatory mood matching in response to affiliative motivation. Personality and Social Psychology, 35, 909–922. Khamis, V., Dukmak, S., & Elhoweris, H. (2008). Factors affecting the motivation to learn among United Arab Emirates middle and high school students. Educational Studies, 34, 191–200. Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New York, NY: The Guilford Press. Krapp, A. (2002). An educational-psychological theory of interest and its relation to SDT. In E. Deci & R. Ryan (Eds.), Handbook of self-determination research (pp. 405–427). New York: The University of Rochester Press. Kunter, M., Frenzel, A., Nagy, G., Baumert, J., & Pekrun, R. (2011). Teacher enthusiasm: Dimensionality and context specificity. Contemporary Educational Psychology, 36, 289–301. Kunter, M., Tsai, Y., Klusmann, U., Brunner, M., Krauss, S., & Baumert, J. (2008). Students’ and mathematics teachers’ perceptions of teacher enthusiasm and instruction. Learning and Instruction, 18, 468–482. Lakin, J. L., & Chartrand, T. L. (2003). Using nonconscious behavioral mimicry to create affiliation and rapport. Psychological Science, 14, 334–339. Linnenbrink-Garcia, L., Durik, A. M., Conley, A, M., Barron, K. E., Tauer, J. M, Karabenick, S.A., & Harackiewicz, J. M. (2010). Measuring situational interest in academic domains, Educational and Psychological Measurement, 70, 647–671. McKinney, C., Roberson, W., Gilmore, A., Ford, M., & Larkins, A. (1984). Some effects of three levels of teacher enthusiasm on student achievement and evaluation of teacher effectiveness. Journal of Instructional Psychology, 11, 119–124. Meyer, D. K., & Turner, J. C. (2006). Re-conceptualizing emotion and motivation to learn in classroom contexts. Educational Psychology Review, 18, 377–390. Mitchell, M. (1993). Situational interest: Its multifaceted structure in the secondary school mathematics classroom. Journal of Educational Psychology, 85, 424–436. Nelson, R. M., & DeBacker, T. K. (2008). Achievement motivation in adolescents: The role of peer climate and best friends. The Journal of Experimental Education, 76, 170–189. Paige, D. D. (2011). Engaging struggling adolescent readers through situational interest: A model proposing the relationsips among extrinsic motivation, oralreading proficiency, comprehension, and academic achievement. Reading Psychology, 32, 395–425. Patrick, B. (1995). College students’ intrinsic motivation as a function of instructor enthusiasm. Unpublished Doctoral Dissertation, Department of Psychology, University of Rochester.
144
T. Kim, D.L. Schallert / Contemporary Educational Psychology 39 (2014) 134–144
Patrick, B., Hisley, J., & Kempler, T. (2000). What’s everybody so excited about? The effect of teacher enthusiasm on student intrinsic motivation and vitality. Journal of Experimental Education, 68, 217–237. Renninger, K. A., & Hidi, S. (2002). Student interest and achievement: Developmental issues raised by a case study. In A. Wigfield & J. Eccles (Eds.), Development of achievement motivation (pp. 173–175). San Diego, CA: Academic Press. Renninger, K. A., & Wozniak, R. H. (1985). Effect of interest on attention shift, recognition, and recall in young children. Developmental Psychology, 21, 624–632. Ryan, A. M. (2000). Peer groups as a context for the socialization of adolescents’ motivation, engagement, and achievement in school. Educational Psychologist, 35, 101–111. Ryan, A. M. (2001). The peer group as a context for the development of young adolescent motivation and achievement. Child Development, 72, 1135–1150. Sage, N. A., & Kinderman, T. A. (1999). Peer networks, behavior contingencies, and children’s engagement in the classroom. Journal of Developmental Psychology, 45, 143–171.
Sanders, P., & Gosenpud, J. (1986). Perceived instructor enthusiasm and student achievement. Developments in Business Simulation & Experiential Exercises, 13, 52–55. Sass, E. (1989). Motivation in the college classroom: What students tell us. Teaching of Psychology, 16, 86–88. Schiefele, U. (1991). Interest, learning, and motivation. Educational Psychologist, 26, 299–323. Schiefele, U. (2001). The role of interest in motivation and learning. In J. Collis & S. Messick (Eds.), Intelligence and personality: Bridging the gap in theory and measurement (pp. 163–194). Mahwah, NJ: Erlbaum. Schraw, G., Flowerday, T., & Lehman, S. (2001). Increasing situational interest in the classroom. Educational Psychology Review, 13, 211–224. Sinclair, S., Lowery, B. S., Hardin, C. D., & Colangelo, A. (2005). Social tuning of automatic racial attitudes: The role of affiliative motivation. Journal of Personality and Social Psychology, 89, 583–592. Wild, T. C., Enzle, M. E., & Hawkins, W. L. (1992). Effects of perceived extrinsic versus intrinsic teacher motivation on student reactions to skill acquisition. Personality and Social Psychology Bulletin, 18(2), 245–251.