International Journal of Educational Research 65 (2014) 41–53

Contents lists available at ScienceDirect

International Journal of Educational Research

jo u r nal h o mep age: w ww.els evier .c o m/lo c ate / i jed ur es

Striving to connect and striving to learn: Influences of

relational and mastery goals for teaching on teacher behaviorsand student interest and help seeking

Ruth Butler *, Limor Shibaz

Hebrew University of Jerusalem, Israel

A R T I C L E I N F O

Article history:

Received 31 July 2012

Accepted 9 September 2013

Available online 23 October 2013

Keywords:

Teacher motivation

Achievement goals

Teacher caring

A B S T R A C T

We extended a new achievement goal approach to teacher motivation to predict that

relational strivings to achieve caring relationships with students and mastery strivings to

develop competence have different consequences for teacher behaviors and thus for

students. Participants were 341 teachers in Study 1 and 51 middle school teacher-class

pairs (1281 students) in Study 2. As hypothesized, teacher relational goals predicted

teacher reports of social support for students and teacher mastery goals predicted teacher

reports of cognitively stimulating instruction (CSI). HLM analyses of Study 2 data

confirmed that teacher relational goals predicted student help seeking via the relation

with perceived teacher social support. Teacher mastery goals predicted student interest;

this relation was partially mediated by perceived CSI. Results establish strivings to connect

and to learn as distinct systems of teacher motivation.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Student motivation for learning has been a major focus of research in educational psychology for decades. Interestingly,until recently there was almost no systematic, theory-driven research on teacher motivation. In one attempt to address thislacuna, Butler (2006, 2007) proposed that achievement goal theory can provide a promising conceptual framework becausethe school is an achievement arena not only for students, but also for teachers. She began by extrapolating directly fromresearch on students’ achievement goals to examine whether teacher mastery, ability approach, ability avoidance, and workavoidance goals constitute distinct motivational systems. This strategy can be faulted, however, for failing to consider thatteaching is an inherently interpersonal endeavor. Against this background, Butler (2012) expanded her framework andmeasure to incorporate a novel fifth class of relational goals to achieve close and caring relationships with students. Theoverarching objective of the present research was to extend this line of inquiry by examining whether teachers’ mastery andrelational strivings, the two clearly desirable kinds of achievement goals for teaching studied so far, are associated withdifferent teacher behaviors and thus also with different student outcomes.

1.1. Achievement goals for teaching: from four to five factors

Achievement goal theory views students’ perceptions, strategies, and outcomes as depending importantly on theirconstructions of the goals of schoolwork, and thus on what they want to achieve. Initially, theorists distinguished between

* Corresponding author at: School of Education, Hebrew University of Jerusalem, Jerusalem 91905, Israel. Tel.: +97225881350; fax: +97225881350.

E-mail address: msbutler@mscc.huji.ac.il (R. Butler).

0883-0355/$ – see front matter � 2013 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.ijer.2013.09.006

R. Butler, L. Shibaz / International Journal of Educational Research 65 (2014) 41–5342

mastery, learning, or task goals that orient students to strive to learn and develop competence versus ability, performance, orego goals that orient students to strive to maintain self-esteem by proving superior or masking inferior ability (Ames, 1992;Dweck, 1986; Nicholls, 1989). We use the terms ‘‘mastery’’ and ‘‘ability’’ to refer to these two kinds of achievement goals (seealso Midgley et al., 1998). Subsequently, Elliot and Harackiewicz (1996) proposed that strivings to prove superior ability andstrivings to avoid the demonstration of inferior ability reflect distinct approach versus avoidance forms of ability goals. Inaddition, Nicholls (1989) defined a class of work-avoidance goals whereby students try to get by with little effort. Scores ofstudies have confirmed that achievement goals matter because they are associated with qualitative differences in the waysstudents define and evaluate success, process information, and regulate behavior (for reviews see; Butler, 2000; Kaplan &Maehr, 2007).

Following this lead, Butler (2006) proposed that teachers, like students, presumably strive to succeed, but may differ inthe ways they define success, and thus in their achievement goals for teaching. In support, Butler (2007) showed that Israeliteachers’ responses to a self-report measure of Goal Orientations for Teaching (GOT) loaded on four factors thatcorresponded to previously identified goal orientations for learning and reflected strivings to (a) learn and developprofessional competence (mastery goal orientation), (b) prove superior teaching ability (ability-approach), (c) avoid failureand the demonstration of inferior teaching ability (ability-avoidance), and (d) get through the day with little effort (work-avoidance). This structure has been replicated for samples of pre-service and in-service teachers in Germany (Dickhauser,Butler, & Tonjes, 2007; Retelsdorf, Butler, Streblow, & Schiefele, 2010).

Initial validation studies examined relations between teachers’ goal orientations and their help-related attitudes andbehaviors (Butler, 2007; Dickhauser et al., 2007; see also Nitsche, Oliver Dickhauser, Fasching, & Dresel, 2011). In keepingwith results from studies of relations between achievement goals for schoolwork and help seeking among students (for areview see Butler, 2006), mastery goals for teaching predicted adaptive help seeking, as reflected in teacher reports that theysought needed help and preferred help that could enable them to learn and cope better with difficulties with teaching. Incontrast, ability-avoidance goals predicted perceptions of help seeking as a low ability cue and reluctance to seek help, andwork avoidance predicted preferences to save effort by referring problems to others. Other studies of teachers’ strategies andwellbeing have shown similar patterns, whereby teacher mastery goals were associated with an adaptive pattern ofconstructive problem solving, interest in teaching, and low burnout, ability avoidance with defensive self-handicapping, andwork avoidance with a disengaged pattern of burnout and low interest (Papaioannou & Christodoulidis, 2007; Parker,Martin, Colmar, & Liem, 2012; Retelsdorf et al., 2010).

Evidence is also beginning to accumulate that teachers’ achievement goals are associated with theoretically relevantapproaches to instruction. Achievement goal theorists distinguish between mastery instructional practices, wherebyteachers teach in ways that value and recognize students’ personal effort and progress, and performance instructionalpractices, whereby teachers emphasize the importance of showing high ability and performance relative to other students(e.g., Ames, 1992; Midgley et al., 1998, 2000). As one would expect, the more teachers strove to learn and developprofessional competence, the more they reported using mastery instructional practices (Retelsdorf et al., 2010). In contrast,ability and work avoidance goals predicted teacher reports of performance instructional practices (see also Butler, 2012). In asimilar vein, Retelsdorf and Gunther (2011) showed that different achievement goals oriented teachers to use differentstandards to evaluate student performance. Mastery goals oriented teachers to evaluate students relative to their prioroutcomes, while ability and work avoidance goals oriented teachers to evaluate students relative to one another.

Conceptualizing teacher motivation in terms of previously identified goals for learning is clearly fruitful. Extrapolatingdirectly from theory and research on student motivation does not consider how the roles and motives of teachers andstudents might also differ, however. Perhaps most important, teaching is an interpersonal rather than only a personalendeavor. Achievement goal researchers have begun to study young people’s social or friendship goals (Elliot, Gable, &Mapes, 2006; Ryan & Shim, 2008). Students’ interpersonal goals are viewed as operating alongside their academicachievement goals (Roussel, Elliot, & Feltman, 2011; Wentzel, 2000), but for teachers they are inherent to their professionalrole. Thus, while students’ social relationships do not serve as standards for defining and assessing their achievement aslearners, the quality of the relationships teachers create with students is an important aspect of their competence as teachers(Wentzel, 1997). Indeed, in an influential analysis, Noddings (1992) defined an orientation of care for students as the essenceof good teaching.

Against this background, Butler (2012) proposed that strivings to achieve caring relationships with students comprise adistinct class of relational achievement goals for teaching. In support, she showed that items describing such strivings loadedon a fifth factor, distinct from the four achievement goals for teaching assessed in her original measure. Validating a new goalnecessitates verification that it also ‘‘matters’’ in the sense of uniquely predicting relevant outcomes. Butler’s (2012) mainprediction was that teacher relational goal would predict teachers’ social support for students, as expressed by the degree towhich the teacher cares for students, listens to them, and addresses their personal problems (e.g., Patrick, Ryan, & Kaplan,2007; Trickett & Moos, 1973). As expected, the more teachers aspired to create close and caring relationships with students,the more both teachers and their students reported that the teacher behaved in socially supportive ways (Butler, 2012).Importantly, mastery orientation, the other kind of positive achievement goal for teaching did not predict social support. Thiswas the case for teacher ability and work avoidance goals as well.

Butler (2012) anticipated that teacher mastery, rather than relational goals would predict mastery instructional practices.Surprisingly, although teacher strivings to learn and acquire professional competence were significantly correlated withteacher reports of mastery instruction, when Butler modeled both goals, relational, rather than mastery goals predicted

R. Butler, L. Shibaz / International Journal of Educational Research 65 (2014) 41–53 43

mastery instruction. Moreover, student responses to items from the scales assessing perceived social support and masteryinstruction loaded on a single factor, which was significantly predicted by teacher relational, but not mastery goals. Theseresults provided validation for Butler’s new class of relational goals for teaching, but raised another question – whetherteacher mastery goals have distinct consequences for instruction. The main objective of the present studies was to examinethe predictions that these two positive kinds of achievement goals do indeed orient teachers toward different kinds ofclassroom behaviors, which in turn influence students in different ways.

1.2. Relational and mastery goals for teaching, teachers’ classroom behaviors, and students’ help seeking and interest

The constructs of teacher social support and teacher mastery instruction were developed and validated within differenttraditions that focused on classroom social climate (e.g., Trickett & Moos, 1973) and on classroom goal structure (Ames,1992; Midgley et al., 1998), respectively. However, studies in other countries (Kunter et al., 2008; Patrick, Kaplan, & Ryan,2011) have also shown that student reports of teacher social support and of mastery instruction loaded on a single factor.One possibility is that displaying concern for students’ socio-emotional well-being and teaching in ways that value andpromote students’ academic effort and progress are both experienced by students as indications of the teacher’s supportiveinvolvement in promoting their welfare. If so, it makes sense that Butler (2012) found that items from scales of perceivedsocial support and perceived mastery instruction loaded on a single factor of ‘‘perceived teacher involvement’’ (Skinner &Belmont, 1993) that was significantly predicted by teacher strivings to create relationships of trust and care with theirstudents.

In this case, one can ask what kind of positive instructional approach might both differ from social support and beassociated with a mastery more than relational goal orientation for teaching. Teachers who aspire to deepen theirprofessional knowledge and embrace challenge might be particularly oriented to teach in ways that stimulate students toconsider different possibilities and to think critically and independently about what they are learning. In support, Retelsdorfet al. (2010) showed that teacher mastery goals predicted teacher reports of such cognitively stimulating instruction, or inbrief CSI. These authors did not examine teacher relational goals and social support, or student reports of instruction,however. Against this background, we predicted that both teachers and students will differentiate between teacher socialsupport and CSI. We then hypothesized that teacher relational, but not mastery goals will predict teacher and student reportsof teacher social support, and teacher mastery, but not relational goals will predict CSI. Following prior results (Butler, 2012;Retelsdorf et al., 2010), we did not anticipate that teacher ability and work avoidance goals would significantly predict eithersocial support or CSI.

If teachers’ mastery and relational goals influence teachers’ classroom behaviors, one would expect them to influencetheir students as well. Studies have shown both that asking for help relevant to understanding and overcoming difficulty isan adaptive learning strategy and that students seek such help from some more than from other teachers (for reviews seeButler, 2006; Newman, 2000). In this case, it is important to ask which teachers tend to promote adaptive help seeking. Butlerand Shibaz (2008) reasoned that teachers who themselves strive to develop competence will not only engage in adaptivehelp seeking in their own work (Butler, 2007), but will also encourage their students to turn to them for help andexplanations when they encounter difficulties with their schoolwork. Surprisingly, this was not the case, even though Butlerand Shibaz targeted classes that the teachers liked, and in which they were presumably at their best. One possibilityconsistent with the inherently interpersonal nature not only of teaching, but also of asking others for help is that studentswill turn first and foremost to teachers who care about their students. Even ‘‘positive’’ mastery strivings to learn and acquireprofessional competence reflect personal goals that focus mainly on the self. Indeed, Butler (2012) showed that teacherrelational rather than mastery goals predicted student perceptions that the teacher can be trusted to support and help themin personal matters. We thus predicted that teacher relational, rather than mastery goals would predict students’ academichelp seeking. We also predicted that this direct relation would be mediated by student perceptions of teachers’ socialsupport for students in the class.

Validation of these predictions will provide further support for Butler’s extension of her achievement goal framework,which was motivated in large part by the insight that her initial conceptualization did not consider the interpersonal andinteractional nature of teaching. It will also contribute to research on contextual influences on classroom help seeking. Thereis evidence that students were more likely to seek help from teachers whom they experienced as providing them withpersonal support (Marchand & Skinner, 2007; Newman & Schwager, 1993; Ryan, Patrick, & Shim, 2005). No prior studies,however, have examined the present prediction that students will be more likely to seek academic help from teachers theyperceive as caring for students in general, and not just for them personally.

Given our interest in examining if relational and mastery goals are qualitatively different kinds of motives for teaching,the next challenge was to identify a student outcome that might be uniquely predicted by teacher mastery goal orientation.We focused on interest in the teacher’s lessons for several reasons. First, strivings to learn and acquire competence areassociated with interest in learning among students (e.g., Harackiewicz, Barron, Carter, Lehto, & Elliot, 1997) and withinterest in teaching among teachers (Retelsdorf et al., 2010). Second, there are theoretical grounds for proposing thatcognitively stimulating instruction should arouse and maintain students’ interest (Pintrich, 2004). Third, as we have alreadydiscussed, we anticipate that teacher mastery, rather than relational goals will predict CSI. Thus, we predicted that teachermastery, but not relational goals would predict students’ interest in the teacher’s class, and that this relation would bemediated by students’ perceptions of CSI.

R. Butler, L. Shibaz / International Journal of Educational Research 65 (2014) 41–5344

1.3. The current research

We report two studies. Participants in Study 1 were 341 teachers in Israel. The main objective was to test the predictionsthat teachers’ relational, rather than mastery goals would predict teacher reports of social support for students while teachermastery, rather than relational goals would predict teacher reports of CSI. Data collected in a single session can yield inflatedassociations, so teachers completed measures of instruction several months after reporting on their achievement goals. Thesample of Study 2 comprised 51 teachers and students in one class for each teacher. Following Butler and Shibaz (2008), wetargeted classes that teachers liked teaching. The main objective was to examine the predictions that teacher relational andmastery goals would differentially predict students’ help seeking and interest by influencing students’ perceptions of teachersocial support and CSI, respectively. Another objective of both studies was to verify that teachers and students perceive socialsupport and CSI as distinct dimensions of teacher behavior. Both studies were generated from the same framework andaddress similar questions, so we present and briefly comment on each and then address the results from both in Section 4.

2. Study 1

2.1. Methods

2.1.1. Participants and procedure

The sample comprised 341 full-time teachers (303 females, 28 males, 10 NR) in 30 schools in Israel serving diversestudent populations with a mix of about 60% from middle to high levels of SES and about 40% from lower levels of SES. Thenumber of teachers in each school ranged from 2–37; years of teaching experience ranged from 1 to 40 (M = 16.10, SD = 9.52,Skewness = .42). Teachers were recruited during staff meetings in the first two months of the school year. We offeredteachers the equivalent of $25 for completing two surveys; 80% of teachers consented to participate and completed bothsurveys. During the initial meeting teachers completed the Time 1 survey, provided contact information, and recordedwhether they preferred to receive an electronic or a printed version of the Time 2 survey later in the year. To ensureanonymity, we matched responses to the two surveys using the last four digits of the teacher’s mobile phone number.

2.1.2. Measures

Teachers’ achievement goals were assessed at Time 1 with an expanded version of Butler’s (2012) measure of GoalOrientations for Teaching (GOT). Following the format Nicholls (1989) used to assess students’ goals, the opening stem formost items is ‘‘I would feel that I had a successful day in school if. . .’’. The measure reported by Butler (2012) presents 20items, four per scale, assessing mastery goals (e.g., ‘‘. . .I saw that I was developing as a teacher and teaching more effectivelythan in the past’’), ability-approach goals (e.g., ‘‘. . .my classes did better on an exam than those of other teachers’’), ability-avoidance goals (e.g., ‘‘. . .no-one asked a question that I couldn’t answer’’), and work-avoidance goals (e.g., ‘‘. . .some of myclasses were cancelled’’). One relational goal item follows Nicholls’ format (‘‘. . ..I saw that I was developing closer and betterrelationships with students in my classes’’); the other three are more direct (e.g., ‘‘My main goal as a teacher is to show mystudents that I care about them’’). Teachers responded on five-point scales anchored at 1 (totally disagree) and 5 (totally

agree). Butler (2012) reported adequate reliabilities for all scales, but those for mastery and ability avoidance scales wererelatively low (a = .70 and a = .66). We thus added a fifth mastery item (‘‘A main goal for me in my work is to acquire newprofessional knowledge and skills’’) and a fifth ability avoidance item (‘‘I aspire to avoid teaching things that might show thatI have poor ability’’). A confirmatory factor analysis with full maximum likelihood on responses to this expanded measureshowed that the predicted five-factor model fit the model fairly well, x2 (205, N = 341) = 500.17, p < .001, NFI = .89. TLI = .92,CFI = .93, RMSEA = .05. The scale reliabilities presented in Table 1 were adequate; reliabilities for the mastery scale and theability avoidance scale were higher than those reported by Butler (2012).

In the Time 2 survey teachers rated their approaches to instruction on scales anchored at 1 (totally disagree) and 5 (totally

agree). Social support for students was assessed with five items adapted from two research programs that developed parallelscales for teachers and students (Kunter et al., 2008; Marchand & Skinner, 2007). A sample item is, ‘‘I take care of students ifthey have problems’’. Cognitively stimulating instruction (CSI) was assessed with five items from Retelsdorf et al. (2010). Asample item is ‘‘I ask questions and present problems that don’t have a single correct answer to encourage students to thinkof different possibilities’’. Results from a maximum likelihood factor analysis with promax rotation confirmed that itemsloaded on two factors reflecting social support and CSI, respectively. Scale reliabilities were a = .78 for social support anda = .75 for CSI.

2.2. Results

The correlations presented in Table 1 confirmed that teacher relational, but not mastery goals were significantlycorrelated with teacher reports of social support, As predicted, teacher mastery goals were significantly correlated with CSI,but there was a also a low but still significant correlation between relational goals and CSI. Teacher ability approach, abilityavoidance, and work avoidance goals were not significantly correlated with either social support or CSI. There were nosignificant sex differences in any of the variables.

Table 1

Descriptive statistics and correlations for scales of teacher achievement goals and instruction, Study 1, N = 341.

M SD a 1 2 3 4 5 6 7

Goals

1. Relational 3.80 .82 .86 –

2. Mastery 4.22 .61 .74 .24*** –

3. Ability approach 3.38 .93 .80 .13* .21*** –

4. Ability avoidance 3.00 .81 .71 .21*** .20*** .66*** –

5. Work avoidance 2.55 .90 .75 .09 .02 .44*** .52*** –

Instruction

6. Social support 4.22 .61 .78 .42*** .09 .03 .03 �.06 –

7. CSI 3.83 .63 .75 .14* .25*** .07 .07 .06 .38*** –

Note: In this and subsequent tables, CSI, cognitively stimulating instruction.

* p < .01.

*** p < .001.

Relationa l

Mastery

.25***

.50***

.12

.41***

Socio-emotiona l

suppor t

Cognitive

stimulation

-.03

sroivaheB rehcaeTnoitatneirO laoG rehcaeT

R2 = .24

R2 = .10

Fig. 1. Structural equation model showing associations between teachers’ relational and mastery goals for teaching and teacher reports of their classroom

behaviors, Study 1; all variables are latent. For clarity, errors and indicators are not presented. Dashed lines are not significant. ***p < .001.

R. Butler, L. Shibaz / International Journal of Educational Research 65 (2014) 41–53 45

In the next stage we used Structural Equation Modeling (EQS 6.0; Bentler, 2006) with maximum likelihood estimationand maximum likelihood multiple imputation to model relations between teachers’ mastery and relational goals and theirself-reported classroom behaviors. The SEM model presented in Fig. 1 showed good fit, x2 (146, N = 341) = 281.52, p < .001,NFI = .92. TLI = .97, CFI = .97, RMSEA = .03. Results fully supported the predictions; teacher relational, but not mastery goalssignificantly predicted social support and mastery, but not relational goals significantly predicted CSI.

These results are important for two main reasons. First, they confirmed that teachers distinguished between their socialsupport for students and the degree to which they teach in cognitively stimulating ways. Second, mastery and relationalstrivings were differentially related to teachers’ reports of these positive kinds of classroom behaviors. One cannot discountthe possibility of some confirmation bias, whereby, for example, teachers who aspire to create caring relationships with theirstudents are motivated to see or present themselves as behaving in socially supportive ways. Indeed, researchers haveinterpreted the generally low correlations between teacher and student reports of instruction in terms of teacher bias(Wubbels, Brekelman, & Hooymayers, 1992). In Study 2, we thus examined the predictions that the student reports ofteacher social support and CSI would also load on distinct factors that would be differentially associated with teacherrelational and mastery goals. We also predicted that relational goals for teaching will predict students’ intentions to turn tothe teacher for needed help and explanations and mastery goals will predict students’ interest in the teacher’s lessons viatheir influence on perceived social support and CSI, respectively.

3. Study 2

3.1. Methods

3.1.1. Participants

Participants were 51 teachers and 1280 students (51% female) in 51 classes, one class for each participating teacher, in 10of the middle schools sampled in Study 1. Toward the end of the year, we contacted 90 randomly selected teachers who hadcompleted the expanded GOT measure at the beginning of the year; these teachers were not included in the sample of Study1. Because there are few male teachers in Israel, we targeted only female teachers. We were able to schedule data collectionfor one randomly selected class for each of 70 of these teachers. As part of the Time 2 teacher survey described below

R. Butler, L. Shibaz / International Journal of Educational Research 65 (2014) 41–5346

participants rated how much they liked teaching in the targeted class; we analyzed data for classes that the teacher likedvery much. Classes were in 11 subject areas in the humanities, sciences, and social sciences in Grade 7 (14 classes, 380students), Grade 8 (17 classes, 416 students), or Grade 9 (20 classes, 486 students). Between 10–36 students respondedabout each teacher, with an average of 25 respondents in each class. Years of teaching experience ranged from 1 to 34(M = 15.37, SD = 9.25, Skewness = .47). Data collection was authorized by the Ministry of Education.

3.1.2. Measures

Teachers responded to the expanded goal orientations for teaching measure at the beginning of the year, as described forStudy 1. Toward the end of the year teachers completed the measures of social support and CSI described in Study 1, butparticipants in Study 2 were instructed to report about their behaviors in the target class. Scale reliabilities were generallysimilar to those for the larger sample of Study 1; a = .86 for relational goal orientation, a = .74 for mastery goal orientation,a = .80 for ability approach, a = .70 for ability avoidance, a = .75 for work avoidance, a = .79 for CSI, but a rather low a = .67 forsocial support. On a separate page, teachers rated the degree to which they liked the class on a scale of 1 (dislike very much) to7 (like very much indeed). We chose the 51 classes that teachers rated six or seven.

Students completed measures assessing their perceptions of the target teacher’s behaviors, and their intentions to turn tothe teacher for needed help and interest in the teacher’s class; rating scales ranged from 1 (never) to 5 (always). Perceivedteacher support was assessed with four items analogous to those in the corresponding teacher scale (e.g., ‘‘This teacher reallytries to address students’ problems’’). Perceived CSI was assessed with five items corresponding to those in the teachersurvey (e.g., ‘‘This teacher asks questions and present problems that don’t have a single correct answer to encourage us tothink of different possibilities’’). A maximum likelihood factor analysis with promax rotation yielded the predicted twofactors; all items loaded highest on the designated factor. Scale reliabilities were a = .87 for perceived teacher support anda = .75 for perceived cognitive stimulation.

Help-seeking was assessed with three items from a Hebrew measure of students’ intentions to ask the teacher foradaptive help (Butler, 2008). To control for differences in students’ need for help, items describe situations in which studentsmight encounter difficulty with their schoolwork. Students rate the frequency from 1 (never) and 5 (always) with which theywould ask the teacher for help relevant to promoting understanding and mastery. For example, one item reads, ‘‘Imagine thatthis teacher is going over homework. You see you got a question wrong. Will you ask the teacher to explain the question?’’Although the measure assesses student reports rather than their behaviors, for brevity we use the term help seeking.Students’ interest was assessed with three items (e.g., ‘‘This class is very interesting’’) from Harackiewicz et al. (1997). Amaximum likelihood factor analysis with promax rotation yielded two factors; all items loaded highest on the designatedfactor. Scale reliabilities were a = .78 for help seeking and a = .90 for interest.

3.1.3. Procedure

End-of-year measures for teachers and students were administered by one of four research assistants during a classtaught by the participating teacher; students were instructed to answer all questions with reference to the lessons of thetarget teacher in this particular class. The teacher sat at the back of the class throughout because Israeli adolescents tend tobe quite rowdy when the teacher is absent. The research assistants emphasized that students could feel free to express theirthoughts and feelings because they would be completing surveys anonymously and teachers would not see the results. Topreserve anonymity, teacher and student surveys were matched using the last four digits of the teacher’s mobile phone,which teachers provided when they responded to the initial goal survey.

3.2. Results

3.2.1. Teacher goals as predictors of students’ perceptions of instruction, help seeking, and interest

Descriptive statistics and correlations among teacher goals and teacher reports of instruction are presented in Table 2.Correlations between goals and instruction tended to be lower than those in Study 1, but relational goals were most highlycorrelated with social support (p < .05, one tailed) and mastery goals with CSI (p < .10, one tailed). Correlations betweenrelational goals and CSI and between mastery goals and social support were close to zero. The correlations among studentreports of instruction and their own help seeking and interest presented in Table 3 were all positive, significant, andmoderate. In Table 4, we present the correlations relevant to the predicted relations among teacher goals and studentreports; student reports were aggregated within each class. As predicted, only teacher relational goal orientation wassignificantly correlated with perceived social support and with student help seeking. Teacher mastery goal orientation wassignificantly correlated with student reports of CSI and of their own interest in the teacher’s lessons. In contrast with ourpredictions, teacher relational goal orientation was also significantly correlated with CSI and interest. Teacher abilityapproach, ability avoidance, and work avoidance goals were not significantly correlated with student reports of instructionand engagement.

The zero-order correlations are suggestive, but do not consider the structure of the sample whereby students were nestedwithin classes/teachers. We thus examined the effects of teacher goals with multilevel modeling, using the program HLM-6(Raudenbush, Bryk, Cheong, & Congdon, 2004). First, we examined whether student reports of instruction and their ownengagement differed between classes. Teachers and classes were confounded, but students responded about the targetteacher’s instruction and about their own help seeking and interest in that teacher’s class. Because our hypotheses concerned

Table 3

Descriptive statistics and zero order correlations among student reports of engagement and instruction, N = 1280.

M SD 1 2 3 4

1. Social support 3.75 1.04 –

2. CSI 3.29 .82 .50*** –

3. Help seeking 3.72 1.09 .59*** .43*** –

4. Interest and enjoyment 3.30 1.22 .49*** .44*** .48*** –

*** p < .001.

Table 4

Correlations between teacher reports and aggregated student reports in 51 classes.

Teacher behaviors Student engagement

Social support CSI Help seeking Interest

Teacher reports

Goals

Relational .49*** .32** .33* .27*

Mastery .24 .38** .12 .42**

Ability approach .18 .19 .16 .16

Ability avoidance .12 .18 .02 .08

Work avoidance .12 .11 .09 .20

Instruction

Social support .15 .07 .04 .01

CSI �.05 .09 .01 .07

* p < .05.

** p < .01.

*** p < .001.

Table 2

Descriptive statistics and correlations for scales of teacher achievement goals, Study 2, N = 51 teachers.

M SD 1 2 3 4 5 6 7

Goals

1. Relational 3.72 .74 –

2. Mastery 4.16 .60 .38** –

3. Ability approach 3.56 .82 .27 .22 –

4. Ability avoidance 3.04 .67 .37** .24 .58*** –

5. Work avoidance 2.28 .75 .24 .14 .57*** .34* –

Instruction

6. Social support 3.70 .55 .27 .03 �.08 �.01 �.01 –

7. CSI 3.96 .54 .02 .21 .06 �.08 �.03 .30* –

* p < .05.

** p < .01.

*** p < .01.

R. Butler, L. Shibaz / International Journal of Educational Research 65 (2014) 41–53 47

relations between teacher goals and student reports, for clarity we refer to differences between teachers rather than betweenclasses. Fully unconditional models (equivalent to one-way analyses of variance with teachers as a random factor) confirmedsignificant between-teacher differences in student reports of both social support, x2 (50, N = 1280) = 553.07, p < .001 and CSI,x2 (50, N = 1280) = 311.18, p < .001. The ICCs showed that differences between-teachers accounted for 26% of the variance inperceived social support and 17% of the variance in perceived CSI. Differences between teachers accounted for 16% of thevariance in student interest, x2 (50, N = 1280) = 302.70, p < .001, and for a lower but still adequate 6% of the variance in helpseeking, x2 (50, N = 1280) = 134.84, p < .001.

In the next stage, we examined whether teacher goals contributed to explaining between-teacher variation in eachoutcome. Results from an initial set of Means-as-Outcomes regression analyses confirmed that teachers’ ability approach,ability avoidance, and work avoidance goals did not significantly predict any of the outcomes. These goals were thus notincluded in subsequent analyses. The models predicting outcomes from teacher relational and mastery goals are presentedin Table 5; student gender was included at level 1. There was a significant gender effect only for help seeking, with girlsreporting higher intentions to seek needed help than boys. Comparing between-teacher variance estimates with those fromthe model with gender as the only predictor indicated that adding teacher relational and mastery goal orientations resultedin a significant decrease in model deviance and accounted for a fairly substantial proportion of the between-teacher variancein all models. The coefficients supported most of the predictions. First, teacher relational, but not mastery goals significantlypredicted students’ intentions to ask the teacher for needed academic help. Second, teacher mastery, but not relational goals

Table 5

Multi-level analyses predicting student perceptions of the teacher and own help seeking and interest from teacher mastery and relational goal.

Social support CSI Help seeking Interest

Fixed effects

b t b t b t b t

Level 1

Student gender �.07 1.29 �.04 .61 �.20** 3.27 �.05 .74

Level 2

Relational GO .31 3.30*** .19 2.54* .13 2.32* .08 1.19

Mastery GO .08 .62 .20 2.06* .01 .19 .28 2.75**

Model change

Decreased deviance 14.96*** 15.05*** 6.98* 11.24**

% Between teacher variance 29% 33% 24% 25%

Note: Model change is in comparison with a model including gender as a single predictor. Decreased model variance is chi-square with 2 df.

* p < .05.

** p < .01.

*** p < .001.

Table 6

Final hierarchical linear modeling analysis of student interest.

Model 1 Model 2 Model 3

Fixed effects

b t b t b t

Teacher mastery goals (Level 2) .32 3.27** .32 3.27** .16 1.98

Student perceptions of CSI

Within teacher (Level 1) .49 14.56*** .49 14.56***

Teacher (Level 2) .60 4.63***

Model change

Decreased model deviance 9.82*** 211.82*** 16.32***

%Within teacher variance 16%

%Between teacher variance 22% 33%

Note: Model 1 change represents comparison with a model with no predictors. Teacher goal in Model 1 is relevant to predicting between but not within

teacher variance, adding within teacher perceptions of instruction in Model 2 is relevant to predicting within but not between teacher variance, adding

between teacher perceptions of instruction in Model 3 is relevant to predicting between teacher variance.*p < .05.

** p < .01.

*** p < .001.

R. Butler, L. Shibaz / International Journal of Educational Research 65 (2014) 41–5348

significantly predicted student interest. Third, teacher mastery goals significantly predicted student reports of CSI, but notsocial support. However, in contrast with our prediction and the results for teacher reports in Study 1, teacher relationalgoals significantly predicted student reports not only of social support but also of CSI.

3.2.2. From teacher goals through student perceptions of the teacher to student outcomes

The results presented in Table 3 confirm that student reports of teacher behaviors were significantly correlated with theirreports of help seeking and interest. The next step was to examine whether student perceptions of teacher behaviormediated the effects of teacher goals on student outcomes. Agreement between students in the same class was quite high forboth social support (ICC2 = .90) and CSI (ICC2 = .83), so it was appropriate to aggregate perceptions of instruction withinclasses.

The preceding analyses confirmed that teacher mastery goals predicted interest and perceived CSI, but not perceivedsocial support. Thus, in the final analysis for interest, Model 1 included teacher mastery goals, Model 2 added within-teachervariation in students’ perceptions of CSI, and Model 3 added between-teacher variation in perceived CSI. Within-teacherperceptions were teacher-centered and between-teacher perceptions were grand-mean-centered. Mediation is implied ifadding collective perceptions of the teacher in Model 3 both contributes to predicting student interest and reduces thecoefficient for teacher mastery goals. Student perceptions at level 1 were teacher-centered, so this analysis does not testmediation at the individual student level. The results presented in Table 6 confirmed that individual student perceptions ofCSI significantly predicted student interest and resulted in a substantial increase in the within teachers variance accountedfor and a significant decrease in residual variance. Students who agreed more rather than less than their classmates that theteacher taught in cognitively stimulating ways reported greater interest. Adding collective perceptions of CSI in Model 3resulted in a further significant decrease in model variance and resulted in an increase of 30% in the between-teachervariance accounted for. Moreover, the coefficient for teacher mastery goals was substantially reduced. It was still almost

Table 7

Final hierarchical linear modeling analysis of student help seeking.

Model 1 Model 2 Model 3

Fixed effects

b t b t b t

Student gender (Level 1) �.20 3.26*** �.17 3.01** �.16 2.97**

Teacher relational goal (Level 2) .13 2.21* .13 2.14* �.02 .41

Student perceptions of the teacher within teachers (Level 1)

Social support .37 9.98*** .37 9.97***

CSI .29 7.21*** .29 7.21***

Between teachers (Level 2)

Social support .32 5.05***

CSI .19 1.71

Model change

Decreased model deviance 6.95* 337.95 35.58***

% Within teacher variance 24%

% Between teacher variance 24% 76%

Note: Model 1 change represents comparison with a model with gender as the only predictor.

* p < .05.

** p < .01.

*** p < .001.

R. Butler, L. Shibaz / International Journal of Educational Research 65 (2014) 41–53 49

significant, however (p = .05). Thus, the degree to which students in a class agreed that the teacher taught in cognitivelystimulating ways predicted interest in the teacher’s lessons and these collective perceptions partially mediated the relationbetween teacher mastery goals and student interest.

Regarding help seeking, preceding analyses showed that the direct predictor, relational goals, was significantly associatedwith student perceptions of both social support and CSI. Thus, in the final set of analyses, Model 1 included teacher relationalgoals and student gender, Model 2 added individual student perceptions of teacher social support and CSI, and Model 3added collective perceptions of both kinds of teacher behaviors. Results are summarized in Table 7. As expected, addingindividual students’ perceptions of the teacher resulted in a substantial increase in the within teacher variance accounted forand a significant decrease in residual variance. In keeping with the correlations presented in Table 3, both perceived teachersupport and perceived CSI significantly predicted student help seeking at level 1. Adding collective perceptions of the teacherresulted in a further significant decrease in model variance and a very substantial increase of 71% in the between-teachervariance accounted for. As predicted, only perceived social support significantly predicted help seeking at level 2; thecoefficient for teacher relational goals was now close to zero. Thus, after controlling for individual differences in students’perceptions, the degree to which students in a class perceived the teacher as providing social support for students predictedstudent intentions to turn to the teacher for needed help. Moreover, perceived social support fully mediated the relationbetween teacher relational goals and student help seeking.

4. General discussion

Results from two studies confirmed that teachers’ ‘‘strivings to connect’’ and ‘‘strivings to learn’’ influenced theirapproaches to instruction, their classroom behaviors, and student outcomes in theoretically coherent ways. The firstchallenge was to identify kinds of classroom behaviors that might be differentially associated with teachers’ relational andmastery achievement goals for teaching. As we discussed in earlier sections, students do not seem to differentiate betweenteachers’ social for students, on the one hand, and their use of mastery instructional practices, on the other hand (Butler,2012; Kunter et al., 2008; Patrick et al., 2011). The first novel contribution of the present research was in showing that bothteachers in Study 1 and students in Study 2 perceived the degree to which teachers provide social support, for example byaddressing students’ personal problems, and the degree to which they taught in cognitively stimulating ways, for example bypresenting questions and problems that could have different solutions, as distinct dimensions of teacher behavior. Thus, inboth samples, items assessing these constructs loaded on distinct factors that were only moderately correlated with oneanother.

More important in the present context, teacher strivings to achieve close and caring relationships with studentssignificantly predicted their self-reported social support for students, but not their reports of CSI, while teacher strivings tolearn and develop professional competence predicted their reports of CSI, but not social support. In keeping with priorfindings, teacher ability and work avoidance goals were not significantly associated with either kind of teacher behavior(Butler, 2012; Retelsdorf et al., 2010). With one exception, the results for student reports of instruction in Study 2 were verysimilar. Teacher relational, but not mastery goals significantly predicted between-class variance in student reports of theteacher’s social support for students, teacher mastery goals significantly predicted student reports of the teacher’s CSI, andteacher ability and work avoidance goals did not predict either kind of perceived teacher behavior. However, relational goalsfor teaching also made a significant contribution to predicting perceived CSI. In this case, one might wonder whether teacher

R. Butler, L. Shibaz / International Journal of Educational Research 65 (2014) 41–5350

relational goals would significantly predict not only help-seeking, but also interest. This was not the case, however. Rather, aswe predicted, teachers’ relational and mastery goals were differentially associated with these two kinds of positive studentoutcomes.

4.1. From strivings to connect through social support to student help seeking

The results fully supported our predictions that teachers who value and strive to create close and caring relationshipswith students will behave in ways that support their students’ social and emotional well-being, and by so doing willencourage their students to turn to them for adaptive help in the form of explanations and clarifications when theyencounter difficulty with their schoolwork. Teacher relational, but not mastery goals significantly predicted between-classdifferences in student reports of both teacher social support and their own help seeking intentions. Moreover, the degree towhich students in a particular class agreed that the teacher shows care for their students fully mediated the directassociation between teacher’s relational goals and students’ academic help seeking. Some years ago we were puzzled to findthat teacher mastery goal orientation did not predict students’ reports that they would ask the teacher for needed help andexplanations (Butler & Shibaz, 2008). This was before conceptualization of a novel class of relational goals, however (Butler,2012). The present results support our reasoning that (a) whether students turn to a particular teacher depends crucially onwhether they believe that the teacher cares about their students’ welfare and (b) strivings to create close and caringrelationships with students, rather than strivings to learn and acquire competence orient teachers to such an orientation ofconcern. They thus provide important validation for Butler’s (2012) new construct of relational goal orientation for teaching.

Perhaps because of the dominance of achievement goal theory as a framework for studying student help seeking, researchon teacher influences has focused mainly on the degree to which teachers emphasize the importance of personal effort andlearning or of performance relative to other students (e.g., Ryan, Gheen, & Midgley, 1998). There is some evidence thatstudents who personally experience more social support from the teacher were more likely to seek academic help(Marchand & Skinner, 2007; Newman & Schwager, 1993; Ryan et al., 2005). Variations in the quality of the individualteacher–student relationship mean that the same teacher encouraged some students in the class to ask them for help whilediscouraging others, however (Marchand & Skinner, 2007). As far as we know, ours is the first study to show that the degreeto which teachers showed care and concern for their students in general influenced the average level of intentions to askadaptive help among students in the class. We did not ask or observe how exactly teachers responded to students’ questionsand difficulties, but it seems likely that teachers who provide social support, which includes offering students help with theirpersonal problems, tend also to provide academic help in socially supportive ways that scaffold learning and do notundermine students’ self esteem. If so, they should encourage subsequent help seeking not only by the specific recipient butby other students as well. Extrapolating from recent evidence that perceived teacher support predicted adaptive helpseeking from peers (Ryan & Shim, 2012), one can speculate further that teachers who model empathic concern andconstructive helping may also encourage students to help one another.

Confirming a relation between teacher motivation and student help seeking contributes to the literature on classroomhelp seeking, which has not considered why some teachers are more supportive of students’ help seeking than others. Ourresults also have implications for understanding helper–recipient influences in general. Studies in other contexts haveidentified various motives for helping others, and have shown, for example, that people are more likely to consistently offerhelp when helping is motivated by empathic concern rather than by self-interest (for a review, see Batson, 1998).Surprisingly, however, research in this tradition has not examined how motives for helping might influence help seeking.Our findings suggest that this could be a fruitful direction, not only in classrooms but also in other settings in which helpingand help seeking occur, or not, within the context of ongoing relationships.

4.2. From teachers’ strivings to learn to student interest

The more teachers agreed that their goal was to learn new things and acquire professional knowledge and competencethe more their students reported that they found the teachers’ lessons interesting and enjoyable. Moreover, results from themulti-level analysis showed that mastery, but not relational goals significantly predicted student interest. We also predictedthat teacher mastery goals would influence student interest via students’ perceptions of CSI. Our finding that CSI predictedstudent interest itself makes an important contribution. Thus, Pintrich (2004) argued that there are theoretical grounds forpredicting that teaching methods that encourage students to consider discrepancies and alternatives, to think critically, andto apply what they are learning to other domains will arouse interest, but he also noted the dearth of relevant studies. Ourresults showed further that entering collective student perceptions of the teacher’s CSI substantially reduced the directrelation between teacher mastery goals and student interest. The coefficient for mastery goals was almost significant also inthe final model, however. Thus, it seems that teachers who strove to learn and acquire professional competence evokedstudent interest in other ways as well, and not only by providing cognitive stimulation.

One possibility consistent with Retelsdorf et al.’s (2010) finding that teacher mastery goals predicted interest in teachingis that mastery oriented teachers influence student interest also via processes of social and motivational contagion (Wild &Enzle, 2002). Teachers’ own interest may lead them to experience and convey positive emotions and enthusiasm for teaching(Kunter et al., 2008). Such cues may then lead students both to infer teacher interest and to expect the lesson to beinteresting. In support, Radel, Sarazin, Legrain, and Wild (2010) showed that learners made inferences about instructors’

R. Butler, L. Shibaz / International Journal of Educational Research 65 (2014) 41–53 51

intrinsic versus extrinsic motivation on the basis of both teachers’ instructional style and their positive affect; theseinferences then influenced learners’ intrinsic motivation for the activity. More generally, our results complement some otherevidence that teacher motivation influenced students’ motivation (Roth, Assor, Kanat-Maymon, & Kaplan, 2007). A logicalnext step could be to examine whether teachers’ achievement goals also influence their students’ achievement goals.

4.3. Strengths, limitations, and implications

Our studies had several strengths, some of which were also limitations. Focusing on classes the teachers likedenabled us to directly test our interpretation that Butler and Shibaz (2008) did not show a relation between teachers’achievement goals and student help seeking because they had not yet conceptualized and thus did not assess therelevant, relational goals. It is striking that teachers’ achievement goals mattered even when teachers were presumablyat their best, because sampling only liked classes may have led to some underestimation of between-class differences inthe predicted outcomes and thus of the contribution of teachers’ goals. However, this aspect of our design may havemasked potential influences of other kinds of teachers’ goals. It is curious that associations between less desirableability and work avoidance teacher goals and student reports of positive teacher behaviors and student outcomes,though low and not significant, were positive, rather than negative. On the reasonable assumption that teachers likeclasses that are relatively responsive and easy to teach, an intriguing possibility is that even teachers with relativelyhigh levels of less adaptive kinds of personal goals might teach fairly effectively in classes they like. Already in the earlydays of achievement goal theory, Elliot and Dweck (1988) proposed that ability, or what they termed performance goalsundermine performance and persistence mainly when people encounter difficulty. In this case, teachers concerned toappear competent or to minimize effort might be vulnerable to losing motivation, to teaching poorly, and towithdrawing support from students in more, but not less challenging classes. We could not explore this possibilitybecause few teachers rated the class we entered as one they did not like. In all events, it would be interesting in futureresearch to examine how features of classes and the students in them might influence teacher motivation and relationsbetween teacher goals, teacher behaviors, and student engagement.

The longitudinal design of Study 1 and the multi-level approach of Study 2 inspire confidence that teacher relational andmastery goals did indeed orient teachers toward different approaches and classroom behaviors that in turn influencedstudent engagement. Both studies were correlational, however. In addition, we relied only on survey measures. Suchmeasures provide valuable information about informants’ perceptions of themselves, and in the case of students, of theteacher as well. Combining survey with observational data could provide important information, for example about teachersas helpers and students as help seekers that could shed light on teacher responses to student difficulty that might beuniquely associated with different kinds of goals for teaching. Such a design would also address the fact that teachers andclasses were confounded in Study 2. Thus, although we referred to teacher effects, we cannot discount possible class effectson both teachers and students. Finally, adding a fifth, more direct item to Butler’s (2007) mastery and ability-avoidancescales improved scale reliabilities, but these were still not high. Modifying the GOT to include more items that refer directlyto teachers’ goals for teaching might further improve its psychometric properties (e.g., Nitsche et al., 2011).

Despite their limitations, our studies make several important contributions to the emerging field of teacher achievementmotivation. They confirmed the validity and importance of Butler’s (2012) new class of relational goals for teaching. Theyalso provided the most consistent evidence to date that positive strivings to achieve close and caring relationships withstudents, on the one hand, and to learn and develop professional competence, on the other hand, influence students, and doso in large part by influencing teachers’ classroom behaviors. Interestingly, teachers’ reports of their behaviors in the targetclass were not significantly associated with student reports either of teacher behaviors or of their own help seeking andinterest. The generally poor agreement between teachers and students about instruction has been attributed to self-servingtendencies among teachers to over-report socially desirable teaching practices (e.g., Wubbels et al., 1992). An importantimplication of the present results is even if teachers are not always fully aware of how they behave in the classroom, aspiringto embrace challenge and teach more effectively and aspiring to create caring relationships with students are both suchcentral and enduring aspects of teachers’ identity that they orient teachers to teach and interact with students in ways thatexpress their basic goals and beliefs. Our research also contributes to understanding adaptive help seeking by demonstratingthe influence of helpers’ interpersonal orientation and behaviors on this inherently social coping strategy.

Most generally, highlighting teachers’ positive motivations and influences is important at a time when the fashion is tostereotype teachers as unmotivated and ineffective. A crucial next step is to examine how educational policies and schoolenvironments might support optimal teacher motivation. Even at this point, however, there are grounds for concern thatreform initiatives that define teacher competence mainly in terms of student test scores rather than the quality of studentlearning and of teacher–student relationships (e.g., Day, 2002) will undermine rather than promote both mastery andrelational goals for teaching.

Acknowledgements

This research was supported by Israel Science Foundation Grant 900/09. We thank the participating teachers andstudents.

R. Butler, L. Shibaz / International Journal of Educational Research 65 (2014) 41–5352

References

Ames, C. (1992). Goals, structures, and student motivation. Journal of Educational Psychology, 84, 261–271 http://dx.doi.org/10.1037/0022-0663.84.3.261Batson, C. D. (1998). Altruism and prosocial behavior. In Gilbert, D. T., Fiske, S. T., & Lindzey, G. (Eds.), The handbook of social psychology pp. 282–316). (2New York:

McGraw-Hill.Bentler, P. M. (2006). EQS structural equations program, version 6. Encino, CA: Multivariate Software.Butler, R. (2000). What learners want to know. The role of achievement goals in shaping information-seeking performance and interest. In J. M. Harackiewicz & C.

Sansone (Eds.), Intrinsic and extrinsic motivation: The search for optimal motivation and performance (pp. 161–194). San Diego, CA: Academic Press.Butler, R. (2006). An achievement goal perspective on student help seeking and teacher help giving in the classroom: Theory, research, and educational

implications. In S. Karabenick & R. Newman (Eds.), Help seeking in academic settings: Goals, groups and contexts (pp. 15–44). New York: Erlbaum.Butler, R. (2007). Teachers’ achievement goal orientations and associations with teachers’ help seeking: Examination of a novel approach to teacher motivation.

Journal of Educational Psychology, 99, 241–252 http://dx.doi.org/10.1037/0022-0663.99.2.241Butler, R. (2008). Ego-involving and frame of reference effects of tracking on elementary school students’ motivational orientations and help seeking in math class.

Social Psychology of Education, 11, 5–23 http://dx.doi.org/10.1007/s11218-007-9032-0Butler, R. (2012). Striving to connect: Extending an achievement goal approach to teacher motivation to include relational goals for teaching. Journal of Educational

Psychology, 104, 726–742.Butler, R., & Shibaz, L. (2008). Achievement goals for teaching as predictors of teachers’ support for and inhibition of question asking and help seeking and

students’ help seeking and cheating. Learning and Instruction, 18, 453–467 http://dx.doi.org/10.1016/j.learninstruc.2008.06.004Day, C. (2002). School reform and transitions in teacher professionalism and identity. International Journal of Educational Research, 37, 677–692.Dickhauser, O., Butler, R., & Toenjes, B. (2007). That shows I can’t do it: Goals and attitudes to help seeking among pre-service teachers. Zeitschrift fur

entwicklungspsychologie und paedagogische psychologie, 39, 120–126 http://dx.doi.org/10.1026/0049-8637.39.3.120Dweck, C. S. (1986). Motivational processes affecting learning. American Psychologist, 41, 1040–1048 http://dx.doi.org/10.1037/0003-066X.41.10.1040Elliot, A. J., & Harackiewicz, J. M. (1996). Approach and avoidance achievement goals and intrinsic motivation: A mediational analysis. Journal of Personality and

Social Psychology, 70, 461–475.Elliot, A. J., Gable, S. L., & Mapes, R. R. (2006). Approach and avoidance motivation in the social domain. Personality and Social Psychology Bulletin, 32, 378–391

http://dx.doi.org/10.1177/0146167205282153Elliott, E. S., & Dweck, C. S. (1988). Goals: An approach to motivation and achievement. Journal of Personality and Social Psychology, 54, 5–12 http://dx.doi.org/

10.1027/0022-3512.54.1.5Harackiewicz, J. M., Barron, K. E., Carter, S. M., Lehto, A. T., & Elliot, A. J. (1997). Predictors and consequences of achievement goals in the college classroom:

Maintaining interest and making the grade. Journal of Personality and Social Psychology, 73, 1284–1295 http://dx.doi.org/10.1037/0022-3514.73.6.1284Kaplan, A., & Maehr, M. L. (2007). The contributions and prospects of goal orientation theory. Educational Psychology Review, 19, 141–184 http://dx.doi.org/

10.1007/s10648-006-9012-5Kunter, M., Tsai, Y. M., Klusmann, U., Brunner, M., Krauss, S., & Baumert, J. (2008). Enjoying teaching: Enthusiasm and instructional behaviors of secondary school

mathematics teachers. Learning and Instruction, 18, 468–482 http://dx.doi.org/10.1016/j.learninstruc.2008.06.008Marchand, G., & Skinner, E. A. (2007). Motivational dynamics of children’s academic help seeking and concealment. Journal of Educational Psychology, 99, 65–82

http://dx.doi.org/10.1037/0022-0663.99.1.65Midgley, C., Kaplan, A., Middleton, M., Maehr, M. L., Urdan, T., Anderman, L., et al. (1998). The development and validation of scales assessing students’

achievement goal orientations. Contemporary Educational Psychology, 23, 113–131 http://dx.doi.org/10.1006/ceps.1998.0965Midgley, C., Maehr, M. L., Hruada, L. Z., Anderman, E., Anderman, L., Freeman, K. E., et al. (2000). Patterns of Adaptive Learning Survey (PALS) manual. Ann Arbor:

University of Michigan.Newman, R. S. (2000). Social influences on the development of children’s adaptive help seeking: The role of parents, teachers, and peers. Developmental Review, 20,

350–404 http://dx.doi.org/10.1006/drev.1999.0502Newman, R. S., & Schwager, M. T. (1993). Students’ perceptions of the teacher and classmates in relation to reported help seeking in math class. Elementary School

Journal, 94, 3–17, URL: http://www.jstor.org/stable/1001871.Nicholls, J. G. (1989). The competitive ethos and democratic education. Cambridge: Harvard University Press.Nitsche, S., Oliver Dickhauser, O., Fasching, M. S., & Dresel, M. (2011). Rethinking teachers’ goal orientations: Conceptual and methodological enhancements.

Learning and Instruction, 21, 574–586 http://dx.doi.org/10.1016/j.learninstruc.2010.12.001Noddings, N. (1992). The challenge to care in schools: An alternative approach to education. New York: Teachers College Press.Papaioannou, A., & Christodoulidis, T. (2007). A measure of teachers’ achievement goals. Educational Psychology, 27, 349–361 http://dx.doi.org/10.1080/

01443410601104148Parker, P. D., Martin, A. J., Colmar, S., & Liem, G. A. (2012). Teachers’ workplace well-being: Exploring a process model of goal orientation, coping, behavior,

engagement, and well-being. Teaching and Teacher Education: An International Journal of Research and Studies, 28, 503–513.Patrick, H., Kaplan, A., & Ryan, A. M. (2011). Positive classroom motivational environments: Convergence between mastery goal structure and classroom social

climate. Journal of Educational Psychology, 103, 367–382 http://dx.doi.org/10.1037/a0023311Patrick, 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 http://dx.doi.org/10.1037/0022-0663.99.1.83Pintrich, P. R. (2004). A motivational science perspective on the role of student motivation in learning and teaching contexts. Journal of Educational Psychology, 99,

83–98.Radel, R., Sarrazin, P., Legrain, P., & Wild, T. C. (2010, August). Social contagion of motivation between teacher and student: Analyzing underlying processes. Journal

of Educational Psychology, 102(3), 577–587 http://dx.doi.org/10.1037/a0019051Raudenbush, S. W., Bryk, A. S., Cheong, Y. F., & Congdon, R. T. (2004). HLM 6: Hierarchical linear and nonlinear modeling. Lincolnwood, IL: Scientific Software

International.Retelsdorf, J., Butler, R., Streblow, L., & Schiefele, U. (2010). Teachers’ goal orientations for teaching: Associations with instructional practices and teachers’

engagement. Learning and Instruction, 20, 30–46 http://dx.doi.org/10.1016/j.learninstruc.2009.01.001Retelsdorf, J., & Gunther, C. (2011). Achievement goals for teaching and teachers’ reference norms: Relations with instructional practices. Teaching and Teacher

Education, 27, 1111–1119 http://dx.doi.org/10.1016/j.tate.2011.05.007Roth, G., Assor, A., Kanat-Maymon, Y., & Kaplan, H. (2007). Autonomous motivation for teaching: How self-determined teaching may lead to self-determined

learning. Journal of Educational Psychology, 99, 761–774 http://dx.doi.org/10.1037/0022-0663.99.4.761Roussel, P., Elliot, A. J., & Feltman, R. (2011). The influence of achievement goals and social goals on help seeking from peers in an academic context. Learning and

Instruction, 21, 394–402 http://dx.doi.org/10.1016/j.learninstruc.2010.05.003Ryan, A. M., Gheen, M. H., & Midgley, C. (1998). Why do some students 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, 1–8 http://dx.doi.org/10.1037/0022-0663.90.3.528Ryan, A. M., Patrick, H., & Shim, S. O. (2005). Differential profiles of students identified by their teacher as having avoidant, appropriate, or dependent help-seeking

tendencies in the classroom. Journal of Educational Psychology, 97, 275–285 http://dx.doi.org/10.1037/0022-0663.97.2.275Ryan, A. M., & Shim, S. S. (2008). An exploration of young adolescents’ social achievement goals and social adjustment in middle school. Journal of Educational

Psychology, 100, 672–687 http://dx.doi.org/10.1177/0146167206289345Ryan, A. M., & Shim, S. S. (2012). Changes in help seeking from peers during early adolescence: Associations with changes in achievement and perceptions of

teachers. Journal of Educational Psychology, 104, 1122–1134 http://dx.doi.org/10.1037/a0027696

R. Butler, L. Shibaz / International Journal of Educational Research 65 (2014) 41–53 53

Skinner, E. A., & Belmont, M. J. (1993). Motivation in the classroom: Reciprocal effects of teacher behavior and student engagement across the school year. Journalof Educational Psychology, 85, 571–581 http://dx.doi.org/10.1037/0022-0663.85.4.571

Trickett, E. J., & Moos, R. H. (1973). Social environment of junior high and high school classrooms. Journal of Educational Psychology, 65, 93–102 http://dx.doi.org/10.1037/h0034823

Wentzel, K. R. (1997). Student motivation in middle school: The role of perceived pedagogical caring. Journal of Educational Psychology, 89, 411–419 http://dx.doi.org/10.1037/0022-0663.89.3.411

Wentzel, K. R. (2000). What is it that I’m trying to achieve? Classroom goals from a content perspective. Contemporary Educational Psychology, 25, 105–115 http://dx.doi.org/10.1006/ceps.1999.1021

Wild, T. C., & Enzle, M. E. (2002). Social contagion of motivational orientations. In E. L. Deci & R. M. Ryan (Eds.), Handbook of self-determination research (pp. 141–157). Rochester, NY: University of Rochester Press.

Wubbels, T., Brekelmans, M., & Hooymayers, H. P. (1992). Do teacher ideals distort the self-reports of their interpersonal behavior? Teaching and Teacher Education,8, 47–58 http://dx.doi.org/10.1016/0742-051X(92)90039-6