Psychology
2013. Vol.4, No.8, 629-637
Published Online August 2013 in SciRes (http://www.scirp.org/journal/psych) http://dx.doi.org/10.4236/psych.2013.48090
Copyright © 2013 SciRes. 629
A Multilevel Analysis of Classroom Goal Structures’ Effects on
Intrinsic Motivation and Peer Modeling: Teachers’ Promoting
Interaction as a Classroom Level Mediator
Kazuhiro Ohtani1, Ryo Okada2, Takamichi Ito3, Motoyuki Nakaya4
1Department of Human Sciences, University of Osaka, Osaka, Japan
2Department of Education, University of Kagawa, Takamatsu, Japan
3Department of Education, Kyoto University of Education, Kyoto, Japan
4Departmentof Education and Human Development, Nagoya University, Nagoya, Japan
Email: yamato@hus.osaka-u.ac.jp
Received June 7th, 2013; revised July 8th, 2013; accepted July 17th, 2013
Copyright © 2013 Kazuhiro Ohtani et al. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the
original work is properly cited.
This study investigated how classroom goal structures (mastery and performance goal structures) related
to intrinsic motivation and peer modeling focusing on teachers’ promoting interaction as a classroom
level mediator. Authors tested multilevel mediational models with a sample of 1212 Japanese elementary
and junior high school students from 43 classrooms. Results provided support for classroom level media-
tional process in mastery goal structures. Specifically, mastery goal structures related to promoting inter-
action, and promoting interactions related to both intrinsic motivation and peer modeling. Limitations and
future directions of the study were discussed.
Keywords: Classroom Goal Structures; Promoting Interactions; Intrinsic Motivation; Peer Modeling
Introduction
Students’ learning is influenced by educational environment
factors as well as individual factors. In particular, the classroom
context, in which students spend a great deal of time every day,
can be the most powerful predictor of achievement behaviors
and academic outcomes. Previous motivation research has fo-
cused on classroom goal structures and has produced a number
of findings (Kaplan, Middleton, Urdan, & Midgley, 2002). This
study examined the effects of classroom goal structures in
terms of teachers promoting interaction among students.
Achievement Goal Theory
Recent motivational research has been led by achievement
goal theory (e.g., Ames, 1992; Dweck & Leggett, 1988; Maehr
& Zusho, 2009; Nicholls, 1984). Research on achievement goal
theory has focused on the reasons why students try to succeed
and has examined two types of achievement goals: mastery
goal (also called learning goal or task goal) and performance
goal (also called relative ability goal or ego goal). Mastery
goals are defined in terms of focusing on developing one’s
competence, mastering a new skill, trying to accomplish a chal-
lenging task, and trying to understand learning materials. Suc-
cess is evaluated in terms of self-improvement and internal
norms. By contrast, performance goals are defined in terms of
demonstrating competence and having high ability relative to
others, striving to be better than others, and using social com-
parison standards to judge ability and performance. Success is
derived from doing better than others and surpassing normative
performance standards. In recent years, some researchers have
added an approach-avoidance dimension and proposed a 2 × 2
framework of achievement goals (Elliot, 2005; Elliot &
McGregor, 2001). This conceptualization proposes four types
of achievement goals: 1) mastery-approach goals, which focus
on developing competence through task mastery; 2) mastery-
avoidance goals, which focus on the avoidance of task-based
and intrapersonally based incompetence; 3) performance-ap-
proach goals, which focus on attaining competence relative to
others; and 4) performance-avoidance goals, which focus on
avoiding incompetence relative to others. However, few em-
pirical studies have focused on mastery-avoidance goals.
Many researchers have examined individual-level achieve-
ment goals (i.e., personal goals). Personal goals refer to the spe-
cific goals that individuals strive to attain in achievement con-
texts. Personal achievement goals were found to be associated
with various achievement behaviors and academic outcomes
(Meece, Anderman, & Anderman, 2006; Urdan & Schoenfelder,
2006). Personal mastery goals were found to be associated with
high academic self-efficacy, high levels of effort and persis-
tence, use of learning strategies that enhance conceptual under-
standing, and intrinsic motivation. In contrast, pursuing perfor-
mance-avoidance goals is usually associated with a negative
pattern of motivational behaviors such as lesser persistence,
avoidance of help-seeking, and self-handicapping. The effects
of performance-approach goals are not sufficiently clear. Al-
though pursuing performance-approach goals is associated with
less persistence and use of surface-level learning strategies, stu-
dents with higher levels of performance-approach goals tend to
K. OHTANI ET AL.
achieve higher academic performance (Harackiewicz, Barron,
Pintrich, Elliot, & Thrash, 2002).
Classroom Goal Structures
In addition to personal goals, achievement goals have been
treated as a contextual-level factor of the classroom and school.
Goal structures refer to messages in the learning environment
that make certain goals salient (Ames, 1992). Classroom-
focused research has examined how teachers may create dif-
ferent goal structures in their classrooms by using various in-
structional, evaluation, and grouping strategies (Kaplan et al.,
2002). Ames and Archer (1988) designed and used self-report
measures to assess the salience of mastery and performance
goals in the classroom. In their study, students’ perceptions of
mastery goals were related to the use of learning strategies,
preference challenge, and effort attribution. In contrast, percep-
tions of performance goals were associated with ability attribu-
tion and less favorable attitudes toward the class. Ames (1992)
reported that mastery goals are made salient when value is
placed on the process of learning through emphasis on mean-
ingful learning, self-referenced standards, and opportunities for
self-directed learning.
Researchers inspired by Ames’ work have examined the ef-
fects of goal structures on achievement behaviors and academic
outcomes. Mastery goal structures were found to be associated
with academic self-efficacy (Midgley, Anderman, & Hicks,
1995; Roeser, Midgley, & Urdan, 1996), intrinsic motivation
(Murayama & Elliot, 2009), and self-regulated learning strate-
gies (Wolters, 2004). In contrast, performance goal structures
were related to academic self-consciousness (Roeser et al.,
1996), avoidance of academic help-seeking (Ryan, Gheen, &
Midgley, 1998), and disruptive behavior (Kaplan, Gheen, &
Midgley, 2002). However, some researchers reported no nega-
tive effects of performance classroom goal structures (Midgley
et al., 1995; Ryan & Patrick, 2001). Classroom goal structures
also have longitudinal effects on academic outcomes. Urdan
and Midgley (2003) investigated the relationship between the
change in goal structures from fifth to sixth grades and aca-
demic self-efficacy and academic achievement. Students who
perceived an increase or no change in the mastery goal struc-
tures reported higher academic self-efficacy and attained higher
academic achievement than students who perceived a decrease
in the mastery goal structures.
Many researchers have focused on students’ perceptions of
classroom goal structures. On questionnaires, students reported
their subjective perceptions of the emphasis in the classroom
with respect to the purposes for engaging in academic tasks and
the meanings associated with success and achievement. A ques-
tionnaire measuring students’ perceptions of classroom goal
structures has been developed and validated (Midgley, Maehr,
Hruda, Anderman, Anderman, Gheen, Kaplan, Kumar, Mid-
dleton, Nelson, & Urdan, 2000). Recently, classroom goal
structures tend to be considered as group-level variables. A
growing body of research distinguishes between classroom-
level variance and student-level variance and examines the
effects of classroom goal structures by using multilevel analysis
(Kaplan et al., 2002; Ryan et al., 1998; Turner, Midgley, Meyer,
Gheen, Anderman, & Kang, 2002). Because goal structure in
any classroom is shared by students in the class, it is necessary
to note the difference between variance of goal structures as a
characteristic of the classroom and variance of goal structure as
individual students’ perceptions of the classroom.
Another line of researches has focused on the relationships
between classroom goal structures and teachers’ daily educa-
tional practices. Ames (1992) discussed how teachers’ educa-
tional practices influence whether classrooms have characteris-
tics of mastery goal structures or performance goal structures
and proposed some instructional strategies. Some researchers
with qualitative data revealed that teachers’ discourse and in-
structional strategies had original characteristics in mastery or
performance goal classrooms (e.g., Patrick, 2004; Patrick, An-
derman, Ryan, Edelin, & Midgley, 2001; Turner, Meyer, Midg-
ley, & Patrick, 2003; Turner et al., 2002). For example, Turner
et al. (2002) observed sixth grade mathematics classrooms and
stated that teachers conveyed mastery messages to their stu-
dents through explicit admonitions not to feel inadequate or
ashamed when they did not understand. Ames (1992) stated that
teachers’ instructional strategies led to students’ mastery goals.
Teachers’ Promoting Interaction as a Mediator
We focused on teachers promoting interaction as a teachers’
educational practice that conveys mastery message to the stu-
dents. Promoting interaction is one of the social climate of
classroom that affects students’ motivation and achievement,
and refers to teachers’ encouraging peer interaction among
students, such as sharing ideas during whole-class lessons,
working together in small-group activities, or informal help-
seeking and help-giving during individual seatwork (Ryan &
Patrick, 2001). Ryan and Patrick (2001) reported that teachers
varied in the extent to which they encouraged student interac-
tion, and the encouragement led to students’ academic achieve-
ment and motivation. Patrick, Ryan, and Kaplan (2007) exam-
ined the relationship between fifth-grade students’ perceptions
of promoting interaction (e.g., “My teacher often allows us to
discuss our work with classmates”) and their motivation. In the
study, students who perceived that their teachers promoted
interaction reported higher levels of academic self-efficacy and
more task-related interaction. In addition, Patrick (2004) re-
viewed the relevant researchers and concluded that promoting
task-related interaction is related to students’ liking and interest
of school and subject area (these are the components of intrinsic
motivation).
In early literature of achievement goal theory, cooperative
goal structures had been considered an important factor. Ames
and Ames (1984) identified three types of classroom goal
structures: cooperative, competitive, and individualistic goal
structures. Ames and Ames (1984) defined cooperative goal
structures as situations in which the probability of one student
receiving a reward is enhanced by the presence of capable
others. Cooperative goal structures involve interdependence
among group members, where each member shares responsibil-
ity for the outcome or product. However, three types of goal
structures evolved into distinctions between competitive envi-
ronment (i.e., performance goal structures) and non-competitive
environment (i.e., mastery goal structures), and cooperative
goal structures and individualistic goal structures were inte-
grated into mastery goal structures in the conceptualization
process (Patrick, 2004). Thus, the mastery goal structures seem
to include a nature of cooperative goal structures. That is, it is
thought that teachers use cooperative educational strategies (i.e.,
promoting interaction) in mastery goal classrooms.
Classroom goal structures and teachers’ educational instruc-
Copyright © 2013 SciRes.
630
K. OHTANI ET AL.
tional practices seem to intertwine each other in the actual
classrooms. In this study, we assumed the process in which
classroom goal structures influence students’ motivation and
achievement related behaviors through teachers’ instructional
practices on the basis of Patrick, Kaplan, and Ryan’s (2011)
findings and discussion. Patrick et al. (2011) stated that “when
students evaluate the extent to which their teacher promotes
real understanding and personal improvement (i.e., emphasizes
mastery goal structure), they draw from their perceptions of the
teacher’s messages about interpersonal relationships (e.g., sup-
port, respect, helping one another)” (p. 370). In addition, Pat-
rick et al. (2011) found that mastery goal structures could ac-
count for the variance of social climate dimensions involving
teachers promoting interaction. Although the mastery goal
structures and teachers promoting interaction can be related
reciprocally, we assumed the process in which the mastery
message in the classroom are conveyed through teachers’ edu-
cational practice and treated the promoting interaction as a
mediator of the mastery goal structures.
The Peer Modeling
It is possible that teachers promoting interaction influences
students’ use of self-regulated learning. It is especially likely
that students could develop learning strategies mediated by
their peers (i.e., classmates or friends) in mastery goal class-
rooms. We focused on students’ use of peer modeling as a
self-regulated learning strategy. Peer modeling is the cognitive,
affective, and behavioral change that derives from observing
models (Schunk, 1998). In the classroom setting, children not
only learn from direct interaction but also acquire knowledge
and enhance their motivation by observing peers’ learning be-
haviors. Therefore, peer modeling plays an important role in the
learning process through which children acquire skills, behav-
iors, and beliefs.
Experimental research has revealed that peer modeling has a
significant effect on children’s academic achievement and mo-
tivation (Schunk, 1998; Schunk & Hanson, 1985, 1989). Most
of the findings about peer modeling have emerged from ex-
perimental studies in laboratories, and few studies have paid
attention to the peer modeling that students use as learning
strategies in their classrooms. However, peer modeling could
play an important role in daily classroom learning activities
(Brophy, 2004; Ryan, 2000). This study examined the effects of
mastery goal structures mediated by teachers promoting inter-
action on students’ use of peer modeling in the classroom.
It was possible that peer modeling was positively related to
intrinsic motivation. Research of self-regulated learning has
found that intrinsic motivation was associated with more use of
various self-regulated learning strategies (Meece, Blumenfeld,
& Hoyle, 1988; Pintrich & De Groot, 1990; Yamauchi & Ta-
naka, 1998; Young, 2005). This means that intrinsically moti-
vated students tend to use various self-regulated learning strate-
gies in the learning situations. No research has examined the
relationship between intrinsic motivation and peer modeling.
Given that peer modeling is one of the self-regulated learning
strategies, however, it was expected that intrinsic motivation
was positively related to peer modeling in the classroom, al-
though the relationship was not a main focus of this study.
The Present Study
We examined the effects of classroom goal structures on the
students’ motivation and peer modeling in terms of teachers
promoting interaction. Based on the previous findings (Mura-
yama & Elliot, 2009; Patrick, 2004), this study focused on in-
trinsic motivation as students’ motivation. We predicted that
mastery goal structures would influence intrinsic motivation
and peer modeling through teachers promoting interaction. As a
school subject, we focus on mathematics classes. Turner and
Meyer (2009) suggested that many children tend to experience
difficulties in mathematical learning and consequently lose
their confidence. Some researchers have examined the effects
of goals structures in math classrooms (Kaplan et al., 2002;
Ryan et al., 1998; Turner et al., 2002). Thus, we measured goal
structures, teachers promoting interaction, intrinsic motivation,
and peer modeling just in math classes.
In this research, classroom-level variances and student-level
variances were partitioned, and we tested a multilevel media-
tion model. Because mastery goal structures and promoting
interaction were classroom-level variables, we expected that the
above mediation process would be confirmed, especially at the
classroom level.
Method
Participants and Procedure
The sample consisted of Japanese public elementary (fifth
and sixth graders) and junior high school students (seventh and
eighth graders) in 43 classrooms (17 elementary, 26 junior high)
in 10 schools from Kansai region. The final sample included a
total of 1212 students (617 females, 591 males, 4 unspecified).
Schools were recruited by contacting the board of education
and then school administrators. Schools were selected by the
board of education of the cities so that the diversity of students
was maintained.
Measures
Classroom goal structures, teachers promoting interaction,
peer modeling, and intrinsic motivation were all assessed with a
questionnaire. All used a 6-point scale, ranging from 1 (not at
all true) to 6 (very true). Each variable in this study was spe-
cific to the domain of mathematics.
Classroom goal structures. To assess classroom goal struc-
tures, ten items were created based on the works by Miki and
Yamauchi (2005) and Midgley et al. (2000). These consisted of
five items assessing mastery goal structure (sample item: “In
math class, learning new things is important”) and five items
assessing performance goal structure (sample item: “In math
class, getting a good grade is the important thing for students”).
The reliability and validity of the original scale have been
documented in prior studies (Midgley et al., 2000; Miki & Ya-
mauchi, 2005). To assess internal consistency of the scales,
coefficient omega was calculated using a within-classroom
covariance matrix. The coefficient omegas were .73 for mastery
goal structure and .60 for performance goal structure.
Promoting interaction. To assess teachers promoting interac-
tion, we developed four items based on the Ryan and Patrick’s
(2001) scale. Items were: (1) “In math class, our teacher says
that cooperation in the group is important,” (2) “In math class,
our teacher says that talking to each other in a group is an im-
portant thing,” (3) “In math class, our teacher emphasized that
you need to help friends when you know the answers,” and (4)
“In math class, our teacher says that asking peers for help is
Copyright © 2013 SciRes. 631
K. OHTANI ET AL.
Copyright © 2013 SciRes.
632
important when one doesn’t know the answer.” Acceptable
internal consistency was identified (ω = .77).
Intrinsic motivation. To assess students’ intrinsic motivation
for learning mathematics, we used Tanaka and Yamauchi’s
(2000) scale. Tanaka and Yamauchi’s measure is a Japanese
version of Elliot and Church’s (1997) intrinsic motivation scale.
For the purpose of this study, we selected three items that have
shown high factor loadings on the basis of factor analysis in
Tanaka and Yamauchi (2000). Items were (1) “I think this class
is interesting,” (2) “I enjoy this class very much,” and (3) “I
think this class is fun.” The reliability and validity of the origi-
nal scale have been documented in prior studies (Elliot &
Church, 1997; Tanaka & Yamauchi, 2000). The scale showed
high internal consistency (ω = .93) in this study.
Peer modeling. To assess peer modeling, four items were
created based on Okada, Nakaya, Ito, and Ohtani’s (2010) peer-
modeling scale and the conceptual description of peer modeling
(Schunk, 1998). Items were (1) “I try to study like my friends
with high motivation to learn,” (2) “I try to follow friends’
good performance,” (3) “When I cannot solve the problem, I try
to model my friends who can solve the problem,” and (4)
“When studying, I try to learn like my friends do.” The scale
was internally consistent (ω = .82).
Results
Descriptive Statistics and Correlations of Study
Variables
Descriptive statistics and correlations are presented in Table
1. Correlations were calculated at the classroom level and stu-
dent level. At the classroom level, mastery goal structure was
positively related to promoting interaction (r = .77, p < .001),
intrinsic motivation (r = .86, p < .001), and peer modeling (r
= .55, p < .05). Performance goal structure was not strongly
related to other variables except for mastery goal structure (r =
.50). The same results were obtained at the student level. Mas-
tery goal structure showed positive correlations with promoting
interaction (r = .59, p < .001), intrinsic motivation (r = .28, p
< .001), and peer modeling (r = .45, p < .001). Correlation co-
efficients of performance goal structure were all below .21.
Analysis Plan
This research tested how classroom goal structures related to
intrinsic motivation and peer modeling using hierarchical linear
modeling. As a first step, we conducted a preliminary analysis
to check our variables’ between-classroom variances and intra-
class correlations. Next, we tested a multilevel mediation model
(Zhang, Zyphur, & Preacher, 2009) in which classroom goal
structures indirectly influence outcome variables through pro-
moting interaction. In the multilevel mediation model, vari-
ances of all variables were partitioned into classroom level and
student level. Then, mediations were tested in both levels. Since
our measurement of classroom level variables were students’
perception, our multilevel mediation model equivalent to Zhan
et al.’s (2009) 1-1-1 model. In 1-1-1 model, aggregated indica-
tors of classroom goal structures and promoting interaction in
each classroom were created for testing Level 2 mediational
model. The 1-1-1 model is beneficial because it provides the
information of both student and classroom level processes. The
regression coefficients were reported in unstandardized coeffi-
cients.
A multiple imputation technique was used in all subsequent
analyses. The method was beneficial compared to other tech-
niques such as mean substitution or list-wise deletion. PASW
18.0 Missing Values was used to create five imputed data sets
and integrated results were reported. All subsequent analyses
were performed by HLM 7 with maximum likelihood estima-
tion.
Preliminary Analysis
The analyses tested whether the variables were significantly
varied among classrooms. The classroom goal structures sig-
nificantly varied among classrooms: mastery goal structure, χ2
(42) = 206.48, p < .001; performance goal structure, χ2 (42) =
265.91, p < .001. Intraclass correlations were 12% and 16%
respectively. We also checked reliability of classroom goal
structures in case that the aggregation of student rating cause
biased estimation in the classroom level (Bliese, 2000). The
reliabilities were relatively high, .79 and .84, respectively.
We also examined whether our mediator variable (i.e., pro-
moting interaction) and outcome variables (intrinsic motivation
and peer modeling) varied among classrooms and found the
following significant between-class variances: for promoting
interaction, χ2 (42) = 319.5, p < .001, for intrinsic motivation, χ2
(42) = 131.82, p < .001, and for peer modeling, χ2 (42) = 61.61,
p < .05. Intraclass correlations were 18%, 7%, and 1%, respec-
tively. The reliability of promoting interaction was .86.
Multilevel Mediation Model
Intrinsic motivation. For intrinsic motivation, we tested a
multilevel mediation model (Zhang et al., 2009). As the first
step of the meditational analysis, classroom goal structures
were put in the model as both classroom-level and student-level
predictors for the intercept of intrinsic motivation. Student-level
variables were centered at group mean (Zhang et al., 2009). We
Table 1.
Descriptive statistics and Pearson product-moment correlations for the study variables.
N M SD 1 2 3 4 5 N M SD
1 Mastery goal structure 1172 4.50 0.89 - .50*** .77*** .86*** .55* 43 4.48 0.35
2 Performance goal structure 1146 3.23 0.97 .05 - .11 .25 .27 43 3.26 0.43
3 Promoting interaction 1146 4.04 1.01 .59*** .15*** - .87*** .81*** 43 4.04 0.46
4 Intrinsic motivation 1157 3.32 1.58 .28*** .12*** .20*** - .72** 43 3.33 0.50
5 Peer modeling 1153 3.96 1.10 .45*** .21*** .46*** .36*** - 43 3.96 0.25
Note: Descriptive statistics and correlation coefficients above diagonal represent classroom level, below diagonal represent student level. p < .10, **p < .01, ***p < .001.
K. OHTANI ET AL.
controlled grade (fifth grader = 0, sixth grader = 1 seventh
grader = 2, eighth grader = 3) at classroom level, and gender
(female = 0, male = 1) at student level because prior research
has documented those variable are significant predictors of
academic outcomes (Hyde & Durik, 2005; Lepper, Corpus, &
Iyengar, 2005).
In the classroom level, the results suggested that mastery
goal structure influenced intrinsic motivation (γ = 1.32, p
< .001). In contrast, performance goal structure did not reach
significance, so we omitted performance goal structure from
subsequent analyses at both levels (see Appendixes 1 and 2 for
the final models and results). Next, we predicted that mastery
goal structure was associated with promoting interaction and
found significant positive relationships at classroom level (γ =
1.24, p < .001). Finally, promoting interaction was put in the
model as a mediator to predict the intercept of intrinsic motiva-
tion. Promoting interaction showed a significant relationship to
intrinsic motivation (γ = 0.38, p < .05) at the classroom level.
At this time, the coefficients of mastery goal structure reduced
its value from γ = 1.32, p < .001 to γ = 0.84, p < .01. The results
of Sobel’s test determined that this mediation was significant (z
= 2.37, p < .05).
In the student level of mediational analysis, student level
mastery goal structure also a positive predictor of intrinsic mo-
tivation (γ = 0.50, p < .001), as well as promoting interaction (γ
= 0.65, p < .001). In contrast to findings at the classroom level,
student-level promoting interaction did not show a significant
relationship to intrinsic motivation (γ = 0.09, n.s.). Figure 1
and Table A1 shows the final result.
Peer modeling. For peer modeling, we tested a multilevel
mediation model with the same procedures as were used for
intrinsic motivation. As the first step of the meditational analy-
sis, classroom goal structures were put in the model as both
classroom- and student-level predictors for the intercept of peer
modeling. Grade and gender were controlled.
In the classroom level, the results suggested that mastery
goal structure influenced peer modeling (γ = 0.56, p < .001).
Finally, we tested the mediation and found a significant rela-
tionship between promoting interaction and peer modeling (γ =
0.23, p < .05) at classroom level (see Appendix 1 for the final
models)1. This mediation was significant (z = 2.40, p < .05). At
classroom level, the coefficients of mastery goal structure re-
duced its value from γ = 0.56, p < .001 to γ = 0.27, n.s., and the
value was no longer significant. This means full mediation.
At student level, mastery goal structure was a positive pre-
dictor of peer modeling (γ = 0.50, p < .001). Then promoting
interaction predicted peer modeling (γ = 0.34, p < .001). Stu-
dent level mediation was significant (z = 5.42, p < .001). The
final results of the model are presented in Figure 2 and Table
A2.
Discussion
The purpose of this study was to examine the effects of class-
room goal structures on students’ motivation and achievement
related behaviors and tested the mediation effects of teachers
promoting interaction. Our hypothesis was that mastery goal
structures influenced intrinsic motivation and peer modeling
through teachers promoting interaction. It is noteworthy that we
partitioned between classroom-level variances and student-level
variances and examined the mediation processes. Mastery goal
structures and teachers promoting interaction were classroom
variables by nature. The characteristics of the classroom envi-
ronment and teachers’ educational practice should be shared by
students in the same classroom. However, each student can
perceive the characteristic differently. Thus, it is necessary to
treat the group-level variables (i.e., goal structures and teachers
promoting interaction) after dividing the variance to class-
room-level variance and student-level variance. Hereafter, we
would discuss the mediation process considering this partition-
ing.
At the classroom level, the hypothesized mediation process
for intrinsic motivation was confirmed. That is, mastery goal
structure was related to students’ intrinsic motivation, and the
relationships were partly mediated by teachers promoting in-
teraction. This implies that teachers frequently encourage stu-
dents to interact with their peers in the classrooms that have the
characteristics of mastery goal structures, and the interaction
increases the average level of intrinsic motivation for mathe-
matical learning. The advantages of classrooms with mastery
goal structures are mediated by teachers promoting interaction
among students. In contrast, the mediation process was not
fully confirmed at the student level. Although the perception of
mastery goal structure was associated with the perception of
teachers promoting interaction, it did not predict intrinsic moti-
vation significantly. Mastery goal structure was directly related
to intrinsic motivation. The student level results are consistent
with considerable research on the positive effects of mastery
goal structures (Meece et al., 2006; Midgley et al., 1995; Mu-
rayama & Elliot, 2009; Roeser et al., 1996; Urdan & Midgley,
2003; Urdan & Schoenfelder, 2006; Wolters, 2004). In sum, in
classrooms that emphasize mastering new skills, the average
level intrinsic motivation is high. However, the effects of the
characteristics of mastery goal structures shared by students in
a classroom are mediated by classroom-level variables such as
teachers’ instructional strategies (i.e., promoting interaction).
The hypothesized mediation process at the classroom level
was also confirmed for peer modeling. Mastery goal structure
related to students’ peer modeling and the relationship was me-
diated by teachers promoting interaction. The more classrooms
are mastery oriented, the more students in the classrooms tend
to model peers’ learning behaviors with their teachers’ encour-
agement. At the student level, the mediation process was also
confirmed. Students who perceive more mastery goal structure
than the average of their classmates tend to perceive the em-
phasis of promoting interaction and, then, report higher level of
peer modeling. Peer modeling has been mainly examined in
experimental studies (Schunk, 1998). In contrast, this study sug-
gested that students use peer modeling as a kind of learning
strategy, and the use is influenced by the characteristics of
classrooms.
Performance goal structures did not predict teachers promot-
ing interaction or intrinsic motivation at either the classroom or
student levels, although they did predict peer modeling. The
mediation processes of performance goal structures were not
totally confirmed. Some previous studies reported negative ef-
fects of performance goal structures (Kaplan et al., 2002; Roe-
ser et al., 1996; Ryan et al., 1998), and others reported no nega-
tive effects (Midgley et al., 1995; Ryan & Patrick, 2001). Nega-
tive effects of performance goal structures were not found
1Performance goal structure also reached significance (γ = 0.22, p < .01 for
classroom level, γ = 0.20, p < .001 for student level). However, performance
goal structure did not predict promoting interaction at both levels, so we
omitted performance goal structure from the final model.
Copyright © 2013 SciRes. 633
K. OHTANI ET AL.
Classroom level
z = 2.37
*
Student level
z = 1.61
0.65
***
0.09
0.44
***
(0.50
***
)
1.24
***
0.38
*
0.84
**
(1.32
***
)
Promoting
interactio n
Masterygoal
structure
Intrinsic
motivatio n
Promoting
interactio n
Masterygoal
structure
Intrinsic
motivatio n
Figure 1.
Multilevel mediation model predicting intrinsic motivation. Regression coefficients in parentheses
represent direct effects before adding promoting interaction. *p < .05, **p < .01, ***p < .001.
C lassroom level
z = 2.40
*
Student level
z = 5.42
***
0.37
***
(0. 50
***
)
1.24
***
0.23
*
0.27
(0. 56
***
)
0.65
***
0.34
***
Promoting
interactio n
Masterygoal
structure Peer modeling
Promoting
interactio n
Masterygoal
structure Peer modeling
Figure 2.
Multilevel mediation model predicting peer modeling. Regression coefficients in parentheses rep-
resent direct effects before adding promoting interaction. *p < .05, ***p < .001.
in this study.
This study revealed that teachers promoting interaction me-
diated the effects of classroom mastery goal structures on in-
trinsic motivation and peer modeling. These results are consis-
tent with previous studies (Patrick, 2004; Patrick et al., 2011).
It can be understood that the positive effects of promoting in-
teraction among students for the following two reasons. First,
interaction with peers can give students resources that support
their learning. The more time students spend with peers in their
learning, the more easily they can ask classmates for help when
confronting academic difficulties. Students can also acquire
skills by observing peers. This is reflected in the relationship
between promoting interaction and peer modeling in this study.
Second, students can receive emotional support by interacting
with peers. Some researchers suggest that positive relationships
with friends and peers in the classroom play an important role
in students’ academic motivation (Anderman & Freeman, 2004;
Wentzel, 2005). Promoting peer interaction is an important
teaching strategy that influences students’ learning and motiva-
tion.
The main contribution of this study is to reveal classroom
motivational processes in terms of teachers promoting interac-
Copyright © 2013 SciRes.
634
K. OHTANI ET AL.
tion among students. Previous studies mainly focused on stu-
dents’ personal goals as a mediator of classroom effects (e.g.,
Anderman & Anderman, 1999; Roser et al., 1996; Midgley et
al., 1995). Given that a classroom environment is shared by
students, we should pay attention to classroom variables. In this
study, the mediation process for intrinsic motivation was con-
firmed at the classroom level. Among classroom variables, tea-
chers’ instructional practices can have a significant impact on
students’ learning. Because a classroom is a social place by na-
ture, instructional strategies that support peer relationships and
student interaction are especially important. Therefore, it is ne-
cessary for educators and researchers to examine the role of
teaching strategies of promoting peer interaction.
This study provides some prospections for future motiva-
tional research. In this study, teachers’ promoting interaction
mediated the effects of classroom characteristics (i.e., mastery
goal structures) on the students’ intrinsic motivation and peer
modeling. This findings show that interaction between teachers
and students are critical for students’ achievements related be-
havior and motivation. Thus, future research can gain important
insights regarding achievement processes by focusing on inter-
action between teachers and students with more details. In ad-
dition, this study revealed few differences between classroom
level and student level. It may be profitable to pay attention to
student-student interaction as well as teacher-student interac-
tion.
Limitations
This research has some limitations. First, all data were col-
lected at one point in time. The model examined in this study
assumed a causal order in which mastery goal structures influ-
ence teachers promoting interaction, which in turn affects in-
trinsic motivation and peer modeling. The simultaneous meas-
urement, however, makes it difficult to interpret the findings as
a causal sequence. To examine the motivational processes in
terms of causal sequence in more detail, a longitudinal study
would add valuable information. Second, classroom variables
were measured in terms of students’ ratings. We measured stu-
dents’ perceptions of classroom goal structures and teachers
promoting interaction and partitioned the variance into two
levels (classroom and individual student levels). In addition, we
could measure the variables through teachers’ reports and class-
room observation. Some researchers have examined the effects
of classroom combining students’ and teachers’ self-reported
data with the observational data (Turner et al., 2002; Turner et
al., 2003). Patrick (2004) found that teachers in the classroom
which was rated as high mastery oriented by the students em-
phasized understanding and self-referenced improvement through
classroom observations. It may be that observed data and stu-
dents’ reported data of classroom goal structures and teachers
promoting interaction have different effects on students’ moti-
vation. Thus, multiple measurements including students’ or tea-
chers’ self-report and classroom observation would be useful to
consider the effects of classroom variables.
REFERENCES
Ames, C. (1992). Classroom: Goals, structures, and student motivation.
Journal of Educational Psycholo gy, 84, 261-271.
doi:10.1037/0022-0663.84.3.261
Ames, C., & Ames, R. (1984). Systems of student and teacher motiva-
tion: Toward a qualitative definition. Journal of Educational Psy-
chology, 76, 535-556. doi:10.1037/0022-0663.76.4.535
Ames, C., & Archer, J. (1988). Achievement goals in the classroom:
Students’ learning strategies and motivation processes. Journal of
Educational Psychology, 80, 260-267.
doi:10.1037/0022-0663.80.3.260
Anderman, L. H., & Anderman, E. M. (1999). Social predictors of
changes in students’ achievement goal orientation. Contemporary
Educational Psychology, 25, 21-37. doi:10.1006/ceps.1998.0978
Anderman, L. H., & Freeman, T. M. (2004).Students’ sense of belong-
ing in school. In P. R. Pintrich, & M. L. Maehr (Eds.), Advances in
motivation and achievement: Vol. 13. motivating students, improving
schools: The legacy of carol midgley (pp. 27-63). Greenwich, CT:
JAI Press.
Bliese, P. D. (2000). Within-group agreement, non-independence, and
reliability: Implications for data aggregation and analysis. In K. J.
Klein, & S. W. Kozlowski (Eds.), Multilevel theory, research, and
methods in organizations (pp. 349-381). San Francisco: Jossey-Bass.
Brophy, J. (2004). Motivating students to learn (2nd ed.). Hillsdale, NJ:
Lawrence Erlbaum Associates.
Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to
motivation and personality. Psychological Review, 95, 256-273.
doi:10.1037/0033-295X.95.2.256
Elliot, A. J. (2005). A conceptual history of the achievement goal con-
struct. In A. J. Elliot, & C. S. Dweck (Eds.), Handbook of compe-
tence and motivation (pp. 52-72). New York: Guilford Press.
Elliot, A. J., & Church, M. A. (1997). A hierarchical model of approach
and avoidance achievement motivation. Journal of Personality and
Social Psychology, 72, 218-232. doi:10.1037/0022-3514.72.1.218
Elliot, A. J., & McGregor, H. A. (2001). A 2 × 2 achievement goal
framework. Journal of Personality and Social Psychology, 80, 501-
519. doi:10.1037/0022-3514.80.3.501
Harackiewicz, J. M., Barron, K. E., Pintrich, P. R., Elliot, A. J., &
Thrash, T. M. (2002). Revision of achievement goal theory: Neces-
sary and illuminating. Journal of Educational Psychology, 94, 638-
645. doi:10.1037/0022-0663.94.3.638
Hyde, J. S., & Durik, A. (2005). Gender, competence and motivation.
In A. J. Elliot, & C. S. Dweck (Eds.), Handbook of competence and
motivation (pp. 375-392). New York: Guilford Press.
Kaplan, A., Gheen, M., & Midgley, C. (2002). Classroom goal structure
and student disruptive behaviour. British Journal of Educational Psy-
chology, 72, 191-211. doi:10.1348/000709902158847
Kaplan, A., Middleton, M. J., Urdan, T., & Midgley, C. (2002). Ach-
ievement goals and goal structures. In C. Midgley (Ed.), Goals, goal
structures, and patterns of adaptive learning (pp. 21-53). Mahwah,
NJ: Erlbaum.
Lepper, M. R., Corpus, J. H., & Iyengar, S. S. (2005). Intrinsic and ex-
trinsic motivational orientation in the classroom: Age differences and
academic correlates. Journal of Educational Psychology, 97, 184-
196. doi:10.1037/0022-0663.97.2.184
Maehr, M. L., & Zusho, A. (2009). Achievement goal theory: The past,
present, and future. In K. R. Wentzel, & A. Wigfield (Eds.), Hand-
book of motivation at school (pp. 77-104). New York: Routledge.
Meece, J. L., Anderman, E. M., & Anderman, L. H. (2006). Classroom
goal structure, student motivation, and academic achievement. An-
nual Review of Psychology, 57, 487-503.
doi:10.1146/annurev.psych.56.091103.070258
Meece, J. L., Blumenfeld, P. C., & Hoyle, R. H. (1988). Students’ goal
orientations and cognitive engagement in classroom activities. Jour-
nal of Educational Psycho logy, 80, 514-523.
doi:10.1037/0022-0663.80.4.514
Midgley, C. Maehr, M. L., Hruda, L. A., Anderman, E., Anderman, L.,
Gheen, M., Kaplan, A., Kumar, R., Middleton, M. J., Nelson, J., &
Urdan, T. (2000). Manual for the Patterns of Adaptive Learning
Scale. Ann Arbor, MI: University of Michigan.
Midgley, C., Anderman, E., & Hicks, L. (1995). Differences between
elementary and middle school teachers and students: A goal theory
approach. Journal of Early Adolescence , 15, 90-113.
doi:10.1177/0272431695015001006
Miki, K., & Yamauchi, H. (2005). Perceptions of classroom goal struc-
tures, personal achievement goal orientations, and learning strategies.
Japanese Journal of Psychology, 76, 260-268.
Copyright © 2013 SciRes. 635
K. OHTANI ET AL.
Copyright © 2013 SciRes.
636
doi:10.4992/jjpsy.76.260
Murayama, K., & Elliot, A. J. (2009).The joint influence of personal
achievement goals and classroom goal structures on achievement-
relevant outcomes. Journal of Educational Psychology, 101, 432-
447. doi:10.1037/a0014221
Nicholls, J. G. (1984). Achievement motivation: Conception of ability,
subjective experience, task choice, and performance. Psychological
Review, 91, 328-346. doi:10.1037/0033-295X.91.3.328
Okada, R., Nakaya, M., Ito, T., & Ohtani, K. (2010). Individual differ-
ences in peer-modeling. Poster Presented at the Japanese Associa-
tion of Educational Psycholo gy 52n d Annual Conference, Japan.
Patrick, H. (2004). Re-examining classroom mastery goal structure. In
P. R. Pintrich, & M. L. Maehr (Eds.), Advances in motivation and
achievement: Vol. 13. Motivating students, improving schools: The
legacy of carol midgley (pp. 233-263). Greenwich, CT: JAI Press.
Patrick, H., Anderman, L. H., Ryan, A. M., Edelin, K., & Midgley, C.
(2001). Teachers’ communication of goal orientations in four fifth-
grade classrooms. The Elementa ry Sch ool J ournal, 102, 35-58.
doi:10.1086/499692
Patrick, H., Kaplan, A., & Ryan, A. M. (2011). Positive classroom mo-
tivational environments: Convergence between mastery goal struc-
ture and classroom social climate. Journal of Educational Psychol-
ogy, 103, 367-382. doi:10.1037/a0023311
Patrick, H., Ryan, A. M., & Kaplan, A. (2007). Early adolescents’
perceptions of the classroom social environment, motivational beliefs,
and engagement. Journal of Educational Psychology, 99, 83-98.
doi:10.1037/0022-0663.99.1.83
Pintrich, P. R., & De Groot, E. V. (1990). Motivational and self-regu-
lated learning components of classroom academic performance. Jour-
nal of Educational Psycho logy, 82, 33-40.
doi:10.1037/0022-0663.82.1.33
Roeser, R. W., Midgley, C., & Urdan, T. C. (1996). Perceptions of the
school psychological environment and early adolescents’ psycho-
logical and behavioral functioning in school: The mediating role of
goals and belonging. Journal of Educational Psychology, 88, 408-
432. doi:10.1037/0022-0663.88.3.408
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.
doi:10.1207/S15326985EP3502_4
Ryan, A. M., & Patrick, H. (2001). The classroom social environment
and changes in adolescents’ motivation and engagement during mid-
dle school. American Educational Research Journal, 38, 437-460.
doi:10.3102/00028312038002437
Ryan, A. M., Gheen, M. H., & Midgley, C. (1998). Why do some stu-
dents avoid asking for help? An examination of the interplay among
students’ academic efficacy, teachers’ social-emotional role, and the
classroom goal structure. Journal of Educational Psychology, 90,
528-535. doi:10.1037/0022-0663.90.3.528
Schunk, D. H. (1998). Teaching elementary students to self-regulate
practice of mathematical skills with modeling. In D. H. Schunk, & B.
J. Zimmerman (Eds.), Self-regulated learning, from teaching to self-
reflective practice (pp. 137-159). New York: Guilford Press.
Schunk, D. H., & Hanson, A. R. (1985). Peer models: Influence on child-
ren’s self-efficacy and achievement. Journal of Educational Psy-
chology, 77, 313-322. doi:10.1037/0022-0663.77.3.313
Schunk, D. H., & Hanson, A. R. (1989). Self-modeling and children’s
cognitive skill learning. Journal of Educational Psychology, 81, 155-
163. doi:10.1037/0022-0663.81.2.155
Tanaka, A., & Yamauchi, H. (2000). Causal models of achievement
motive, goal orientation, intrinsic interest, and academic achievement
in classroom. The Japanese Journal of Psycho logy, 71, 317-324.
doi:10.4992/jjpsy.71.317
Turner, J. C., & Meyer, D. K. (2009). Understanding motivation in
mathematics: What is happening in classrooms? In K. R. Wentzel, &
A. Wigfield (Eds.), Handbook of motivation at school (pp. 527-552).
New York: Routledge.
Turner, J. C., Meyer, D. K., Midgley, C., & Patrick, H. (2003). Teacher
discourse and sixth graders’ reported affect and achievement behave-
iors in two high-mastery/high-performance mathematics classrooms.
The Elementary School Journal, 103, 357-382. doi:10.1086/499731
Turner, J. C., Midgley, C., Meyer, D. K., Gheen, M., Anderman, E. M.,
& Kang, Y. (2002). The classroom environment and students’ reports
of avoidance strategies in mathematics: A multimethod study. Jour-
nal of Educational Psycho logy, 94, 88-106.
doi:10.1037/0022-0663.94.1.88
Urdan, T., & Midgley, C. (2003). Changes in the perceived classroom
goal structure and pattern of adaptive learning during early adoles-
cence. Contemporary E du cat ion al Psychology, 28, 524-551.
doi:10.1016/S0361-476X(02)00060-7
Urdan, T., & Schoenfelder, E. (2006). Classroom effects on student
motivation: Goal structures, social relationships, and competence be-
liefs. Journal of School Psychology, 44, 331-349.
doi:10.1016/j.jsp.2006.04.003
Wentzel, K. R. (2005). Peer relationships, motivation, and academic
performance at school. In A. J. Elliot, & C. S. Dweck (Eds.), Hand-
book of competence and motivation (pp. 279-296). New York: Guil-
ford Press.
Wolters, C. A. (2004). Advancing achievement goal theory: Using goal
structures and goal orientations to predict students’ motivation, cog-
nition, and achievement. Journal of Educational Psychology, 96,
236-250. doi:10.1037/0022-0663.96.2.236
Yamauchi, H., & Tanaka, K. (1998). Relations of autonomy, self-ref-
erenced beliefs, and self-regulated learning among Japanese children.
Psychological Reports, 82, 803-816. doi:10.2466/pr0.1998.82.3.803
Young, M. R. (2005). The motivational effects of the classroom envi-
ronment in facilitating self-regulated learning. Journal of Marketing
Education, 27, 25-40. doi:10.1177/0273475304273346
Zhang, Z., Zyphur, M. J., & Preacher, K. J. (2009). Testing multilevel
mediation using hierarchical linear models. Organizational Research
Methods, 12, 695-719. doi:10.1177/1094428108327450
K. OHTANI ET AL.
Appendix 1
Final models of the multilevel meditational models.
Intrinsic Motivatio n
Step 1
Intrinsic motivation = β0j + β1j (gender) + β2j (mastery goal
structure) + rij
β0j = γ00 + γ01 (school grade) + γ02 (mastery goal structure) +
u0j, β1j = γ10, β2j = γ20
Step 2
Promoting interaction = β0j + β1j (gender) + β2j (mastery goal
structure) + rij
β0j = γ00 + γ01 (school grade) + γ02 (mastery goal structure) +
u0j, β1j = γ10, β2j = γ20
Step 3
Intrinsic motivation = β0j + β1j (gender) + β2j (mastery goal
structure) + β3j (promoting interaction) + rij
β0j = γ00 + γ01 (school grade) + γ02 (mastery goal structure) +
γ03 (promoting interaction) + u0j, β1j = γ10, β2j = γ20, β3j = γ30
Peer Modeling
Step 1
Peer modeling = β0j + β1j (gender) + β2j (mastery goal struc-
ture) + rij
β0j = γ00 + γ01 (school grade) + γ02 (mastery goal structure) +
u0j, β1j = γ10, β2j = γ20
Step 2
Promoting interaction = β0j + β1j (gender) + β2j (mastery goal
structure) + rij
β0j = γ00 + γ01 (school grade) + γ02 (mastery goal structure) +
u0j, β1j = γ10, β2j = γ20
Step 3
Peer modeling = β0j + β1j (gender) + β2j (mastery goal struc-
ture) + β3j (promoting interaction) + rij
β0j = γ00 + γ01 (school grade) + γ02 (mastery goal structure) +
γ03 (promoting interaction) + u0j, β1j = γ10, β2j = γ20, β3j = γ30
Appendix 2
Table A1.
The result of multilevel mediation model predicting intrinsic motivation.
Step 1 Step 2 Step 3
To intrinsic motivation To promoting interaction To intrinsic motivation
Classroom level Coefficient SE Coefficient SE Coefficient SE
Intercept (γ00) 2.85*** 0.14 3.87*** 0.14 2.92*** 0.13
Grade (γ01) 0.13* 0.06 0.10 0.06 0.09 0.06
Mastery goal structure (γ02) 1.32*** 0.20 1.24*** 0.20 0.84** 0.04
Promoting interaction (γ03) - - 0.38* 0.15
Random effects of intercept (u0j)
Variance 0.041*** 0.067*** 0.029*
The amount of variance explained in each step 67.98% 52.46% 9.17%
Student level
Gender (γ10) 0.47*** 0.09 0.05 0.04 0.48*** 0.09
Mastery goal structure(γ20) 0.50*** 0.06 0.65*** 0.03 0.44*** 0.06
Promoting interaction(γ30) - - 0.10 0.06
The amount of variance explained in each step 7.73% 34.49% 0.01%
*p < .05, **p < .01, ***p < .001.
Table A2.
The result of multilevel mediation model predicting peer modeling.
Step 1 Step 2 Step 3
To peer modeling To promoting interaction To peer modeling
Classroom level Coefficient SE Coefficient SE Coefficient SE
Intercept (γ00) 3.87*** 0.09 3.87*** 0.14 3.90*** 0.08
Grade (γ01) 0.09* 0.09 0.10 0.06 0.06 0.04
Mastery goal structure (γ02) 0.56*** 0.11 1.24*** 0.20 0.27 0.17
Promoting interaction (γ03) - - 0.23* 0.09
Random effects of intercept (u0j)
Variance 0.010 0.067*** 0.007
The amount of variance explained in each step 46.52% 52.46% 14.21%
Student level
Gender (γ10) 0.14* 0.06 0.05 0.04 0.12* 0.05
Mastery goal structure (γ20) 0.50*** 0.03 0.65*** 0.03 0.37*** 0.05
Promoting interaction (γ30) - - 0.34*** 0.06
The amount of variance explained in each step 20.23% 34.49% 5.30%
p < .10, *p < .05, **p < .01, ***p < .001.
Copyright © 2013 SciRes. 637