at SciVerse ScienceDirect

Teaching and Teacher Education 28 (2012) 936e947

Contents lists available

Teaching and Teacher Education

journal homepage: www.elsevier .com/locate/ tate

Teachers’ conceptions of excellent teaching and its relationships to self-reportedteaching practices

Junjun Chen a,*, Gavin T.L. Brown b, John A.C. Hattie c, Pam Millward b

aDepartment of Education Policy and Leadership, The Hong Kong Institute of Education, 10 Lo Ping Road, Tai Po, NT, Hong Kong SARbUniversity of Auckland, New ZealandcUniversity of Melbourne, Australia

h i g h l i g h t s

< Survey validated new Teachers’ Conceptions of Excellent Teaching inventory.< Chinese middle school teachers identified 5 factors for excellent teaching.< Student-oriented ideas of excellence were endorsed more than examination success.< Excellence factors predicted similar importance of factors of teaching practices.< The professionalism factor predicted both examination and student-engagement practices.

a r t i c l e i n f o

Article history:Received 5 September 2011Received in revised form22 April 2012Accepted 24 April 2012

Keywords:Teacher thinkingExcellent teachingChinese middle schoolStructural equation modeling

* Corresponding author. Tel.: þ64 7 8384466x6149E-mail address: jchen@waikato.ac.nz (J. Chen).

0742-051X/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.tate.2012.04.006

a b s t r a c t

This study surveyed Chinese middle school (n ¼ 951) teachers’ conceptions of excellent teaching andexamined the relationship of those conceptions to their self-reported teaching practices. Responses wereanalyzed using confirmatory factor analysis and structural equation modeling. These teachers identifiedone examination-oriented dimension and four interactive, pedagogical dimensions of excellent teachingand four dimensions of teaching practice. The structural model indicated a high consistency rate (83%)between teachers’ conceptions of excellent teaching and the corresponding self-reported practices.Implications for teaching standards, teacher professional development, and examination system arediscussed.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

China is going through massive education reforms in order toprepare more students for success in the 21st century (Feng,2006). A key to the success of the reforms is the quality ofteachers and their teaching methods (Zhang & Collis, 1995).Chinese teachers tend to rely heavily on traditional forms ofteaching highly focused on school examinations, hence, thesuccess of the reforms depends in part on changing how teacherscarry out their pedagogical role (Gao & Watkins, 2002). Researchin China has shown that unless teachers realize the need fora new approach to teaching and challenge their traditional viewsof the nature of learning and teaching, real change is unlikely to

; fax: þ64 7 8384712.

All rights reserved.

occur (Gao, 1992; Gao et al., 1989). As Liu (1995) claimed:“Changing the conceptions of teaching and schooling is now thekey to improving students’ quality” (p. 9). Thus changing howteachers conceive of teaching appears to be urgent in the currentChinese educational context. However, first it is importantto understand teachers’ current conceptions of teachingexcellence.

Excellent teaching is considered the basis for enhancing studentlearning, achieving school effectiveness, improving teacherevaluation, and designing and improving teacher educationprograms (Betoret & Artiga, 2004; Skamp & Mueller, 2001). Itwould be useful to understand how teachers conceive of excellentteaching and how their conceptions relate to teaching practices asa way of informing education reforms in China. It may be thata clearer understanding of how excellent teaching relates toteaching practices will contribute to improved teaching andstudent learning.

J. Chen et al. / Teaching and Teacher Education 28 (2012) 936e947 937

2. Literature review

2.1. International perspectives of excellent teaching

Researchers have been investigating the very best teaching fordecades. A variety of terms have been used: for example, ‘good’(Watkins & Zhang, 2006), ‘effective’ (Witcher, Onwuegbuzie, &Minor, 2001), ‘highly accomplished‘ (National Board forProfessional Teaching Standards (NBPTS), 1987), ‘excellent’ (Kane,Sandretto, & Heath, 2004), and ‘qualified’ (Darling-Hammond &Youngs, 2002). Despite the variation in terminology, all of thestudies seem to be describing a similar set of attributes concerningthe very best teaching. Hence, the term ‘excellent’ is used in thisstudy to describe the very best teaching.

Much of the significant research into teaching excellence hasfocused on identifying the characteristics of excellent teachers andexcellent teaching and developing models of an excellent teacherand excellent teaching. A substantial body of research aroundconceptions of teaching has also been developed. An influentialframework for understanding teachers’ conceptions of teachingseems to be Kember’s (1997) three major approaches. The ‘teacher-centered’ orientation involved imparting information and trans-mitting structured knowledge. The ‘student-centered’ orientationincluded facilitating understanding and encouraging conceptualchange. The ‘student-teacher interaction’ conception acted asa bridging conception between the two contrasting orientations.Most of the identified conceptions of excellent teaching can belocated on this teacher-student continuum.

However, it might be argued that there are alternative ways ofconceiving teaching. For example, Brown, Lake, and Matters (2009)suggested that teacher- and student-orientations may be parallelemphases rather than opposites and further suggested, based on ananalysis of the teaching conception frameworks in Pratt (1992) andFenstermacher and Soltis (1998), that there may also be an indi-vidualistic versus collectivist approach to understanding teachingconceptions.

The point of teaching quality or excellence is that excellentteachers work more on the nature of tasks and activities so as toengage their students in conceptual understating, analyticalthinking, and reasoning during instruction rather than engage ineither teacher-centered or student-centered teaching practices(Boston & Smith, 2009; Rittle-Johnson, Siegler, & Alibali, 2001).Indeed, Kember and Wong (2000) found that Hong Kong universitystudents weremore interested in the quality of teaching, rather thanwhether it was teacher- or student-oriented. Nonetheless, theteacherestudent continuum approach to interpreting conceptions ofteaching has been productive and reasonably robust (Kember, 1997)and, thus,wasused as a framework for interpreting data in this study.

Beishuizen, Hof, Van Putten, Bouwmeester, and Asscher (2001)examined students’ and teachers’ conceptions of excellentteachers in the Netherlands at the primary and high school level byasking participants to write about the characteristics of an excellentteacher. The results indicated that excellent primary teachers wereconceived as acting as competent instructors focusing on trans-mitting knowledge and skills. Excellent teachers at high schoolwere conceived as establishing the studenteteacher relationship.This study focused more on teacher subject knowledge, pedagog-ical knowledge, and interactions between the teacher and students,which were embedded in the teacher-centered orientation and theteacherestudent interaction. However, this study overlookedhuman dimensions (e.g., charisma, passion compassion, egalitari-anism, and a sense of humor) that have been frequently advocatedin theWestern literature. This human dimension “gives all teachers,whether in the classroom, the sports arena, or the home, theirpower as effective influencers” (Kottler & Zehm, 2000, p. 2).

Kottler and Zehm (2000) reported a number of characteristics ofexcellent teachers, at the primary and high school level in theUnited States, which included sound subject knowledge, propermethods of instruction, and other related skills. In addition, theyclaimed that teachers weremore responsible for presenting subjectknowledge and assisting their students in achieving high scoresrather than teaching as “away of life” (p. 20). More recently, Kottler,Zehm, and Kottler (2005), added this humanistic perspective as oneof the essential characteristics of excellent teachers. Their work,therefore, identified three categories of excellent teaching: that is,the teacher-centered conception (sound subject knowledge), andthe two more student-centered conceptions (i.e., proper pedagog-ical strategies and skills and way of being a person).

Another study from the United States (Witcher et al., 2001)examined 219 pre-service teachers’ perceptions of excellentteachers at the tertiary level and identified the three categories:that is, a teacher-centered category (classroom and behaviormanagement and knowledge about subject), a teacherestudentinteraction category (teaching methodology and enthusiasm forteaching) and a student-centered category (love for students andbeing supportive of them). In a follow-up investigation, Minor,Onwuegbuzie, Witcher, and James (2002) found a fourth categorytermed as professionalism. These findings were confirmed byOnwuegbuzie et al. (2007).

Rather than just identifying characteristics of excellent teachers,the management consulting company Hay McBer (2000) identifieda model of teaching excellence across all school levels in the UnitedKingdom. They collected multiple data sources (i.e., interviews,questionnaires, observations, and focus-group discussions) anddeveloped a model consisting of three interactive categories(i.e., professional characteristics, teaching skills, and classroomclimate) that accounted for over 30 percent of the variance instudent progress. The model identified three conceptual categoriesof teaching excellence, which they expected to be predictors ofstudent progress.

To sum up, the Western teaching excellence literature hasidentified the following main categories of excellent teachers:personal and professional characteristics, sound subject andpedagogical knowledge, classroom climate and management,studenteteacher relationships, professionalism, and the effects ofthese categories on student progress. Most categories identified inthese studies can be located on Kember’s (1997) teacher-centeredand student-centered continuum.

Research into conceptions of excellent teaching is still at thevery beginning stage in China. Watkins and Zhang (2006) reviewedseveral related studies, which were conducted across all schoollevels in Hong Kong or Mainland China. They concluded that anexcellent Chinese teacher had a deep level of knowledge, focusedon knowledge delivery, helped students to dowell in examinations,cared about students’ personal problems, had close relationshipswith students, was a good moral guide, and promoted positiveattitudes toward society (see also Cortazzi & Jin, 1996; Kelly &Watkins, 2002; Watkins & Sachs, 2004; Zhang, 2007). Chen(2007) conducted a small qualitative survey about excellentteaching among 22 Chinese middle school teachers and identifiedfour major conceptions of excellent teaching. The teacher-centeredorientation comprised planning and preparing structured lessons;the more student-centered conceptions included connectingschool knowledge to other areas, caring for students, and guidingstudents’ all-round development.

When these conceptions were compared with findingsfrom Western literature, some similar categories were identified.These included sound subject and pedagogical knowledge,studenteteacher relationships, a commitment to teaching and anunderstanding of students’ development. However, Chinese

J. Chen et al. / Teaching and Teacher Education 28 (2012) 936e947938

conceptions of excellent teaching excluded the notion of profes-sionalism and personal and professional characteristics. Westernconceptions of excellent teaching on the other hand did not includebeing a moral guide or caring about a student’s personal life. Insummary, the studies on teaching excellence provided a theoreticalframework to inform this study and to guide the construction of aninstrument aimed at collecting teacher thinking about excellentteaching.

2.2. Teaching practices

Researchers have classified teaching practices using differentdimensions. Each of these dimensions could exemplify Kember’s(1997) continuum ranging from more student-centered perspec-tives to more teacher-centered perspectives. The teacher-centeredperspective is associated with transmission-oriented practices,while the student-centered approach involves more progressivepractices (Brown, Lake, et al., 2009; Onwuegbuzie, Witch, Filer, &Downing, 2000). The progressive student-centered practicefocused on human development, interacting with the world ofpeople and materials, and building humanist values (Nager &Shapiro, 2000). Progressive practices linked school learning tostudents’ lives outside the school context. Thus, decisions con-cerning practices implemented by progressive teachers were basedon students’ personal experiences and were more likely to developpractical abilities and skills (Zhao, 2007). On the other hand,transmissive teachers adopt a set of teacher-centered practices thatemphasize dispensing knowledge to students, using the lecture asa primary teaching method, and assessments that focus onmemorization of facts and details (Brown, Lake, et al., 2009).

Similarly, teaching practices in China could be placed on theteacher-centered and student-centered continuum (Kember, 1997).The Chinese model of teaching is characterized by the transmissionof knowledge principally through an imitative, repetitive, andmemorizing process (Hughes & Yuan, 2005; Paine, 1992; Tang &Absalom, 1998). The role of a teacher is to deposit knowledge intostudents (Zhang & Collis, 1995). Teaching methods are largelyexpository and teacher-directed activities include drilling forexternally-mandated, high-stakes or consequence examinations(Watkins & Biggs, 2001). The teaching process is teacher-centeredand text-based. Both the teacher and the textbook are regardedas authoritative sources of knowledge (Cortazzi & Jin, 1996). Thisdescription indicates that Chinese teaching practices are veryteacher-centered, but it does not exclude the possibility of student-centered practices. This review of the literature enabled the authorsto develop an instrument relating to statements of teaching prac-tices in this study.

2.3. The relationship of teaching conceptions to practices

Kember (1997, p. 207) concluded that “the methods of teachingadopted, the learning tasks set, the assessment demands made andtheworkload specified are strongly influenced by the orientation toteaching”. Kember and Kwan (2000) concluded that teachers whoconceived of excellent teaching as transmitting knowledge weremore likely to use content-centered approaches to teaching. On theother hand, teachers who conceived of excellent teaching as facil-itative in nature tended to use learning-centered approaches.Similar patterns have been reported elsewhere (Gow & Kember,1993; Ho, Watkins, & Kelly, 2001; Samuelowicz, 1994).

Kane, Sandretto, and Heath (2002), however, commented thatKember and Kwan (2000) had missed an opportunity to examinethe extent to which teachers’ conceptions of excellent teachingwere related to their claimed teaching practices. In addition, theKember and Kwan study did not reveal how these conceptions

were inter-related or cross-related. Apart from Hay McBer’s (2000)model for excellent teaching, it seemed that no other projects haddeveloped the categories of excellent teaching into an entire model.Even HayMcBer’s model did not explore the relationships betweenthe categories. The studies in teaching practices are in a similarsituation.

2.4. Theoretical framework e the theory of planned behavior

The theory of planned behavior (Ajzen, 2005) providesa powerful framework for positioning research into conceptionsand practices.What people believe, the amount of control they haveor perceive they have, societal norms, and people’s intentionsinteract to shape the behaviors and practices people carry out.Generally speaking, the more favorable the attitudes and subjectivenorms with respect to a behavior, and the greater the perceivedbehavioral control, the more likely it is that people will perform thebehavior in accordance with their intentions (Ajzen, 2005). Inaddition, it seems that the stronger conceptions are, the more likelythey are to influence the corresponding behavior, while weakconceptions have little impact on behavior (Ajzen, 2002). It is notedthat subordinates (e.g., teachers) are especially influenced by thenormative views of their superiors (e.g., principals) to whom theyare accountable (Lerner & Tetlock, 1999). Not every belief-actionprocess, however, moves in this consistent chain. In someinstances, people’s actions are not consistentwith their conceptions(Ajzen, 2002, 2005). Ajzen (2005) suggested reutilization ofbehavior as a possible explanation for any inconsistencies.

Therefore, it would be worthwhile to explore not only whatteachers’ conceptions of excellent teaching are but also how theyrelate to their self-reported teaching practices. The researchinstruments employed in this study would be helpful in makingexplicit the participants’ conceptions of excellent teaching andpractices.

2.5. The contemporary Chinese educational context

In order to understand teaching excellence in Chinese middleschools, it is necessary to understand more about the context ofmiddle schools in China. Note that middle school in China includesthree grades (i.e., Years 7, 8, and 9). Currently, the Chinese Ministryof Education is the country’s highest policy body. Since the late1990s, Chinese basic education has adopted a new system charac-terized by “central leadership, local responsibility, and manage-ment at various levels” (Wang, 2003, p. 3). The central governmenthas assigned the responsibility of popularizing compulsoryeducation to provincial and municipal governments. Thus, Chineseeducation is still greatly centralized, but has begun to be decen-tralized over the past two decades (Cheng & DeLany, 1999).Authorities at local or higher level run schools and these schoolshave consultative powers in terms of school management. TheCompulsory Education Law in 1986 stipulated what was to be themanagement system of compulsory education. In most cases,a 6e3e3 system has been established. Basic education in Chinaincludes six years of primary school, three years of middle school,and three years of high school education. The first nine years (i.e.,primary and middle school) of education are compulsory and thecentral government pays for tuition. Students who are six years oldare eligible to start their primary education.

In China, primary school students are able to move to themiddleschool without sitting for a selective examination. At the end ofmiddle school, however, students have to sit for the high schoolentrance examination to gain promotion to high school since highschool education is not included in compulsory education in China.High schools enroll students based on students’ examination scores

1 Values calculated at http://www.raosoft.com/samplesize.html.

J. Chen et al. / Teaching and Teacher Education 28 (2012) 936e947 939

in the high school entrance examination. The local educationalbureau and the school board, however, make decisions aboutschool entry (Liu, 2005). Each high school has its own entrancescore and the students who reach the school’s entrance score areable to attend the school by paying the required tuition fee. A highschool also has the right to enroll the students whose scores arelower than the school entrance score, but within a certain scorerange by charging a higher tuition fee and/or an additional fee. Thescore and fee ranges are dependent on the policy of the individualschool. In general, the lower a student’s score is, the higher the feethat will be charged. This situation pushes teachers, students,school leaders, and parents to put a high value on the results thatstudents achieve in Chinese middle school examinations.

2.6. Research questions and hypotheses

The immediate goal of this studywas to identify the conceptionsof middle school teachers in one province of China concerning thenature of excellent teaching and their self-reported teachingpractices, and to explore the relationship of those conceptions totheir teaching practices. The long-term goal of the study was toinform the development of excellent teaching standards for use inteacher education and professional development in Chinese middleschools. The two research questions in this study were proposed:

1) What are the conceptions of excellent teaching and self-reported teaching practices of Chinese middle school teachers?

2) How are Chinese middle school teachers’ conceptions ofexcellent teaching related to their teaching practice?

Based on the theory of planned behavior (Ajzen, 2005) and therelated previous empirical findings regarding the relationshipbetween teachers’ conceptions of teaching and their practices, itwas hypothesized firstly that teachers’ conceptions of excellentteaching would have statistically significant relationships to self-reported teaching practices. Specifically, it was expected thatthere would be alignment between more teacher-centered andstudent-centered conceptions of excellent teaching with moreteacher-centered and student-centered teaching practices.

3. Method

This study was based on two pilot studies from the same samplepopulation (Chen, 2010). One was a qualitative study that inde-pendently explored how Chinese middle school teachers, students,and principals conceived of excellent teaching. The findings wereused to construct the questionnaires of excellent teaching andteaching practice in Chinese middle schools. The other study wasa quantitative study to explore teachers’ conceptions of excellentteaching using trial versions of the self-administered question-naires with a relatively small sample of middle school teachers. Thepresent study used the revised questionnaires to investigate thestrength of agreement toward various conceptions of excellentteaching and teaching practice among a large sample of teachers.The data were examined independently in two models and thenexamined concurrently in a structural model.

3.1. Sample

A sample of 2200 Chinese middle school teachers, from 29middle schools in nine cities from one province in China, wasapproached and 951 valid questionnaires were returned givinga response rate of 43.2%. While population demographic charac-teristics for China middle school teachers are not available, thecurrent sample of 951 relative to the population of just over

5,000,000 middle school teachers in China produces a margin oferror, based on sampling theory,1 of only 3.18%, meaning that greatconfidence can be placed in the sample means. The majority of theteachers (62.0%) were female. Most (81.1%) of the teachers helda Bachelor qualification, 13.9% held a Diploma qualification, andonly 5.0% held a Master qualification. Most teachers were relativelyyoung (i.e., 25.3% under 33 years old and 40.1% aged between 33and 40 years), with only 34.6% over 40 years old. Years of teachingexperience were consistent with age (i.e., 29.9% < 8 years, 40.0%8e15 years, and 30.1% > 15 years). Just over a quarter of theteachers held a junior teacher certificate, 40.3% of the teachers heldan intermediate teacher certificate, and about one-third helda senior teacher certificate. There were 42.1% of the teachers taughtsocial science and 57.9% of them taught science. Approximatelyequal percentages of the teachers taught at Years 7 (32.5%), 8(31.5%), and 9 (33.5%) in the middle school years in China, and only2.5% taught across three years.

3.2. Instruments

Two instruments were developed for this study based on thereviewed literature and the two aforementioned pilot studies(Chen, 2010). The Teachers’ Conceptions of Excellent Teaching(TCET) comprised 58 statements about the nature of excellentteaching in Chinese middle schools. Teachers were asked to indi-cate how strongly they agreed that each statement located in thefirst column was the part of excellent teaching in Chinese middleschools. They were asked to tick one box using a six-point, posi-tively-packed agreement response rating scale in the secondcolumn. This response scale included two negative and four posi-tive agreement responses with identical scores (e.g., stronglydisagree ¼ 1, mostly disagree ¼ 2, slightly agree ¼ 3, moderatelyagree ¼ 4, mostly agree ¼ 5, and strongly agree ¼ 6). Positively-packed rating scales are known to generate discrimination incontexts of social desirability (Brown, 2004; Lam & Klockars, 1982).

The Teaching Practice (TP) questionnaire had 48 statementsabout the importance of teachers’ own teaching practices (e.g.,linking new knowledge to prior knowledge; criticizing studentswhen they made mistakes or misbehave) used in Chinese middleschools. Teachers were asked to indicate how important theyconsidered each practice on the list to be. The importance scale hadtwo negative and three positive responses (e.g., not important ¼ 1,of little importance ¼ 2, of some importance ¼ 3, of considerableimportance ¼ 4, and of critical importance ¼ 5).

3.3. Administration

Three qualified research assistants, who were trained in Chinaby the first author, conducted the survey. Since they had helpedcollect data for the two previous pilot studies, they clearly knew theprocess of data administration and collection. In the first step, theresearch assistants contacted 50 middle schools in nine cities fromone province in China. The research assistants visited 29 schoolsthat had agreed to cooperate. They briefed each principal witha written Participant Information Sheet, and asked for permissionto conduct the research within the school. Once each principalsigned a consent form, they were asked to call for teacher volun-teers within their school. Teacher Participant Information Sheetsand questionnaires were then distributed to teacher volunteers bythe principals. Teachers were asked to return them directly to theresearch assistants within four weeks using the addressed stampedenvelopes supplied with the questionnaires.

J. Chen et al. / Teaching and Teacher Education 28 (2012) 936e947940

3.4. Analysis

The two inventories (i.e., excellent teaching and teaching prac-tice) were analyzed separately. First, variables and cases with morethan 5% missing values were removed and values for missingresponses were calculated using the expectation maximization(EM) missing values procedure (Dempster, Laird, & Rubin, 1977).Since 106 variables had less than 5% missing values, all items in thequestionnaire were kept for analysis. However, the 100 cases withmore than 5% missing values were removed. The EM procedure isa two-stage process of estimating the value of missing data andmodeling the parameters (i.e., means, standard deviations, andcorrelations) assuming the missing data are as estimated. The datadistribution of the variables was checked before analysis. Skewnessvalues ranged from .09 to 1.95, and kurtosis ranged from .77 to 1.39.Thus, it was concluded that the variables had a sufficiently normaldistribution for further analysis (Byrne, 2010). Note validity andreliability of the instrument had been checked in the two earlierpilot studies.

The two models, TCET and TP, were generated using exploratoryfactor analysis with maximum likelihood estimation and obliquerotation (Costello & Osborne, 2005). Items were removed that hadloadings smaller than .30 on their intended conceptual factors, orwhich had cross loadings greater than .30 on other factors, or whichdid not match logically and theoretically with other items in thesame factors. Confirmatory factor analysis (CFA) (or rathera restrictive factor analysis) was utilized to test these two models(Anderson & Gerbing, 1988) using Amos 18 (Arbuckle, 2009). Itemswere removed in CFA that still had low loadings (<.30) on theirconceptual factors or which had strong modification indices toother factors. At this stage, the linear structural relations for the twomeasurement models were identified and estimated (Marsh &Hocevar, 1985). Note that estimates of reliability (a) for the TCETmodel factors were within the range of .74e.89, with an averagevalue of .82, and for the TP model alpha fell in the range of.72e.88, with an average value of .81. These values indicated thatthe items were sufficiently consistent within each factor so as topermit meaningful further analysis.

Structural equationmodeling (SEM) is a goodmethod for testinghypotheses about linear relations among latent factors (Wegener &Fabrigar, 2000). SEM was employed to identify the statisticallysignificant paths between the two different structures (i.e.,teachers’ conceptions of excellent teaching and teaching practices)(Hoyle, 1995; Klem, 2000; Thompson, 2000). In structural models,paths between latent factors are standardized beta regressionweights (b), meaning that a path value of 1.0 indicates that thedependent variable increases by one standard deviation for everystandard deviation increase in the predictor variable. When testingthe hypothesized regression paths in this study, three criteria wereapplied. The first of these was that the introduction or eliminationof regression paths improved the quality of fit. The second was thatthe introduced regression paths did not cause negative error vari-ances. The last one was that introduced regression paths hadstatistically significant values.

In CFA and SEM, fit indices indicate how well a model fits thedata andwhether there is sufficient evidence that themodel shouldor should not be rejected. It is known that most fit indices aresensitive to different conditions (e.g., sample size, modelcomplexity, or model misspecification). The c2 statistic is extremelysensitive to large sample size (Byrne, 2010; Kline, 2005;Vandenberg & Lance, 2000). Some goodness-of-fit indices (e.g.,Tucker-Lewis Index and Comparative Fit Index) are overly sensitiveto complex models, that is, those with more than three factors orwith hierarchical structure (Cheung & Rensvold, 2002). The mostpopular index, the root mean square error of approximation

(RMSEA), is sensitive to complex models and specification error(Fan & Sivo, 2007). In contrast, the standardized root mean squareresidual (SRMR) and gamma hat have been shown to be relativelyresistant to the impact of large samples, complex models, andmodel misspecification (Fan & Sivo, 2007). Nonetheless, it is rec-ommended that multiple fit indices are reported when assessingmodel fit (Fan & Sivo, 2005; Hu & Bentler, 1999). In this study, thefollowing values for the various fit indices are used to test a modelfit: the c2 package including c2, df, c2/df, SRMR, RMSEAwith 90% CI,CFI, and gamma hat. For the c2 package, perfect fit occurs when c2

is roughly equal to its df and good fit is inferred when the ratio of c2

to df has p > .05. When RMSEA and SRMR are �.05 fit is good andwhen �.08, it is acceptable (Browne & Cudeck, 1989; Byrne, 2010;Hu & Bentler, 1999). When the 90% CI for RMSEA falls in the rangefrom .050 to .080, fit is acceptable (Kenny, 2008). For goodness-of-fit indicators, when CFI and gamma hat are �.95, fit is good andwhen they are >.90, fit is acceptable (Byrne, 2010; Hoyle, 1995).

4. Results

4.1. TCET model

Initial analysis identified five factors of excellent teaching:Examination-oriented, Developing Lifelong Learners, StudentFocused, Being Responsible For Engaging Students In Learning, andBeing A Professional Learner. However, it was clear from inspectionof factor inter-correlations that the examination-oriented factorwas relatively independent of the four other factors that werehighly inter-correlated. Hence, a hierarchical model with twodimensions and five first-order factors was tested. This modelcorrelated the examination factor with a second-order factor underwhich the four other factors were aggregated. This model, based on26 items, of teacher conceptions of excellent teaching had good fit(c2 ¼ 1154.19, df ¼ 294, c2/df ¼ 3.93, p ¼ .05; RMSEA ¼ .055, 90%CI ¼ .052 w .059; SRMR ¼ .055; CFI ¼ .91; gamma hat ¼ .94) (seeFig. 1). The interactive, pedagogical dimension of excellent teachinghad only a weak inter-correlation (r ¼ .25) with the examinationdimension of excellent teaching.

Table 1 shows the descriptive statistics for the five factors.Teachers gave the strongest endorsement to Excellence As Devel-oping Lifelong Learners and the lowest mean to Excellence AsExamination-oriented.

These five factors and two dimensions indicated the multi-faceted nature of Chinese middle school teachers’ conceptions ofexcellent teaching. Clearly, the pedagogical interactive dimensionsvery much accounted for teachers’ understanding of excellentteaching, whereas, the examination orientation had only a weakrelationship to teachers’ conceptions of excellence. The model ofexcellent teaching seems largely consistent with Western andConfucian ideas of caring for and motivating students to be all thatthe they can be while excellent teachers fulfill professionalresponsibility by maintaining their own teaching excellence. Incontrast, despite the strong contextual emphasis on examinationsin Chinese schooling, teachers only weakly conceived that exami-nation preparation and drilling was part of excellence.

It should be noted that the factor loadings for the four peda-gogical factors of excellence are very high and there may be someredundancy in using as many as four factors. Therefore, futureresearch should develop items that increase the independence ofthese four factors.

Teacher characteristics were examined as a possible source ofvariance in teachers’ conceptions of excellent teaching. The rulesused to classify these demographic characteristics aimed to ensurethat the categories were statistically and practically logical inChinese middle schools. Age was broken into three categories

Fig. 1. Chinese middle school teachers’ conceptions of excellent teaching model. Error terms removed for simplicity.

J. Chen et al. / Teaching and Teacher Education 28 (2012) 936e947 941

(i.e., <33 years, 33e40 years, and >40 years). Years of work expe-rience were collapsed into three categories (i.e., <8 years, 8e15years, and >15 years). The three teacher certificate categorieswere below junior or junior, intermediate, senior or above. Teacherqualification was also grouped into three categories (i.e., College,Bachelor, and Master). The year levels teachers taught werecollapsed into four categories (i.e., Year 7, Year 8, Year 9, and mixedyears). The subjects teachers taught were divided into two cate-gories: social science (e.g., arts, Chinese, English, geography,history, music, PE, and politics) and science (e.g., chemistry,computer information and technology, mathematics, and physics).

Multivariate analysis of variance (MANOVA) with main effectsfound that all seven teacher characteristics had statistically signif-icant mean differences in their levels of agreement with theseconceptions of excellent teaching (sex: F5,930 ¼ 5.75, p < .001; age:F101,794 ¼ 2.46, p < .05; teacher certificate: F101,806 ¼ 2.30, p < .05;qualification: F101,730 ¼ 2.25, p < .05; subject: F5,919 ¼ 2.80, p < .05;year: F152,521 ¼ 2.12, p < .05; work experience: F101,822 ¼ 2.04,p < .05).

However, only one effect size was clearly of medium size;teachers with College qualifications (M ¼ 5.37) were more likely tosee excellence in terms of developing their students as lifelonglearners than their colleagues holding Bachelor or Master qualifi-cations (M ¼ 5.10) (d ¼ .36). Otherwise, all differences in meanendorsement were either trivial or small. School type led toa statistically significant mean difference for all five factors with an

Table 1Descriptive statistics for TCET factors.

Scale M SD Alpha

Excellence As Examination-oriented 3.98 1.10 .77Excellence As Developing Lifelong Learners 5.26 .69 .80Excellence As Student Focused 5.21 .68 .85Excellence As Being Responsible For

Engaging Students In Learning4.85 .80 .70

Excellence As Being A Professional Learner 5.16 .72 .70

average effect size of d ¼ .21. Teachers in urban schools tended toagree more with all conceptions than teachers in rural schools.MANOVA found that school size had statistically significant meandifferences in their levels of agreement with Examination-oriented(d ¼ .34) and Being Responsible For Engaging Students In Learning(d¼ .20). The teachers in schools larger than 1600 students held thehighest level of agreement for all factors compared to teachers whoworked in small or medium schools with an average effect size ofd ¼ .15. The consistency of the mean level of endorsement for thesefive factors across school and teacher characteristics suggests thatthe TCET taps into stable priorities in how Chinese middle schoolteachers conceive of excellent teaching.

4.2. TP model

A four factor, inter-correlated model of self-reported teachingpractices had good fit (c2 ¼ 1047.85, df ¼ 291, c2/df ¼ 3.60, p ¼ .06;RMSEA ¼ .052, 90% CI ¼ .049 w .056; SRMR ¼ .049; CFI ¼ .90; andgamma hat ¼ .94) (see Fig. 2). This model involved four majorfactors, although one of which had two first-order factors (i.e.,Being Strictly Teacher-Oriented). The four factors were BeingStrictly Teacher-oriented, Using Novelty And Variety Of Methods,Keeping Extramural Connections, and Encouraging StudentInvolvement. The two first-order factors for Being Strictly Teacher-oriented were Teacher Is In Charge and Doing Examination Prac-tices. Unlike the four factors of TCET, the inter-correlations betweenthe four major practices were moderate indicating that the prac-tices were relatively independent of each other.

Teachers endorsed as most important the factor EncourageStudent Involvement (M ¼ 4.13) and gave the least importance(albeit still positive) to Teacher Is In Charge (M¼ 3.65) (see Table 2).

This model showed that middle school teachers in Chinareported thinking that highly student-oriented teaching practices,as well as strongly teacher-oriented practices focused aroundpreparing students for examinations, were related. The muchstronger inter-correlation between examination-oriented andinteractive, student-oriented teaching practices showed the most

Fig. 2. Teaching practice model in Chinese middle school.

J. Chen et al. / Teaching and Teacher Education 28 (2012) 936e947942

salient difference in the thinking of Chinesemiddle school teachers.While examinations were clearly not part of their conception ofexcellent teaching, they were seen as a highly consistent teachingpractice.

As with the conceptions of excellent teaching, MANOVA withmain effects found that six teacher characteristics excluding workexperience had statistically significant mean differences in theirviews of importance of teaching practices (sex: F5,930 ¼ 4.56,p < .01; age: F101,794 ¼ 2.44, p < .01; teacher certificate:F101,806¼ 3.00, p< .01; qualification: F101,730¼ 1.88, p< .05; subject:F5,919 ¼ 3.51, p < .05; year: F152,521 ¼ 2.00, p < .05). All effect sizeswere small and only one effect size was d ¼ .30 related to teacherswith a junior or intermediate teacher certificate being more likelyto think that keeping extramural connections was important thantheir colleagues with a senior teacher certificate. There wasa statistically significant mean difference for school type and sizeon the same factor e Strict Teacher-Oriented practice. Teacherswho worked in rural schools and those working in schools withbetween 1300 and 1600 students gave lower importance for thisteaching practice than the teachers in urban (d ¼ .26), as well aslarger and smaller schools (d ¼ .13). Otherwise, mean scores onlyhad small to trivial differences according to school or teachercharacteristics. The consistency of the mean level of endorsement

Table 2Descriptive statistics for TP scales.

Scale M SD Alpha

Being Strictly Teacher-oriented 3.70 .69 .80Using Novelty And Variety of Methods 3.76 .72 .72Keeping Extramural Connections 3.65 .83 .79Encouraging Student Involvement 4.13 .55 .81

for these four factors suggests that the TP taps into stable prioritiesin how Chinese middle school teachers conceive of the importanceof various teaching practices.

4.3. TCET and TP structural model

In keepingwith the Ajzenmodel of beliefs acting as predictors ofbehavior, we examined the predictive value of the five conceptionsof excellent teaching on self-reported teaching practices. To answerResearch Question 2, we regressed the TCET factors on the TPfactors and removed statistically non-significant paths. The struc-tural model revealed that Chinese middle school teacher concep-tions of excellent teaching were meaningful predictors of theconceptually similar teaching practices (c2 ¼ 4477.09, df¼ 1257, c2/df¼ 3.56, p¼ .06; RMSEA¼ .052, 90% CI¼ .050w .054; SRMR¼ .07;CFI ¼ .83; gamma hat ¼ .90) (see Fig. 3).

In general, the two major dimensions of excellent teachingconceptions were conceptually aligned to their correspondingteaching practices. The teacher-centered, examination-orientedconception of excellent teaching significantly predicted the BeingStrictly Teacher-oriented teaching practices. Likewise, the student-centered Excellence As Being Responsible For Engaging Students InLearning conception significantly predicted the teaching practice ofKeeping Extramural Connections with families. Two more of thestudent-centered conceptions of excellent teaching (i.e., ExcellenceAs Developing Lifelong Learners and Excellence As StudentFocused) each predicted one of themore student-centered teachingpractices (i.e., Encouraging Student Involvement and Using NoveltyAnd Variety Of Methods). It appeared that the first pair shared anelement of preparation for life beyond school, while the second pairboth had to do with high-quality classroom environments.However, the purportedly student-centered conception, Excellence

Fig. 3. The structural model of Chinese middle school teachers’ conceptions of excellent teaching and teaching practices. ET ¼ Excellent Teaching; TP ¼ Teaching Practice; *p < .05;**p < .01. Correlations removed for simplicity.

J. Chen et al. / Teaching and Teacher Education 28 (2012) 936e947 943

As Being A Professional Learner, significantly predicted both theBeing Strictly Teacher-oriented teaching practice and student-centered practice of Encouraging Student Involvement. However,the path to the latter, conceptually similar, factor was almost twiceas strong as to the teacher-oriented practice factor.

These results show that statistically significant, strong regres-sion paths, which explained a substantial proportion of variance(see SMC values), existed between teachers’ conceptions of excel-lent teaching and their conceptions of the importance of four majortypes of teaching practices and that these paths were generallyaligned between teacher-centered and student-centered orienta-tions. However, the conception that excellent teaching required theteacher to be a professional learner, contrary to the distinctionmade between teacher- and student-centered approaches,contributed to both teacher- and student-centered teaching prac-tices. Thus, it would appear that teacher perceptions of excellentteaching around sustaining their own excellence serves both thegoal of exercising control and providing preparation for examina-tions, as well as encouraging student involvement in learning.Nonetheless, professionalism exists to serve student involvementmuch more (b ¼ .60) than it does to explain teacher control forexamination practices (b ¼ .37). Thus, inclusion of teacherconceptions of excellent teaching does aid in understanding howteachers perceived the importance of teaching practice.

5. Discussion

This study answered two research questions about the structureand relationship of teacher conceptions of excellent teaching andthe importance of various teaching practices. The survey of Chinesemiddle school teachers established further the usefulness of theteacher-centered vs. student-centered continuum (Kember, 1997)

in understanding teachers’ thinking. The teacher-centeredconception of excellence was the most weakly endorsed and wasvery much focused on examinations. While this conception hadonly a weak correlation with the dominant, student-centeredconception of excellent teaching, it had a strong relationship tostrictly teacher-oriented practices. In contrast, the teachers agreedstrongly with the student-centered conceptions of excellencewhich were aligned with student-centered teaching practices.

The one exception to this alignment was the dual relationship ofthe excellent teacher as professional learner to both strict exami-nation practices and student-centered practices. It is noted thatthese teachers were predominantly student-centered in theimportance they gave to various practices and the path fromprofessionalism to student-centered practices was much strongerthan to teacher-centered practices. The student-centered emphasisis somewhat surprising given the traditional emphasis on exami-nations in China. Nonetheless, in general, this study shows thatChinese middle school teachers’ conceptions of excellent teachingare consistent with those results previously reported in bothChinese and Western research literature.

However, unlike the extreme duality of teaching practiceproposed by some researchers (Chiu, 2006; Maxwell, McWilliam,Hemmeter, Ault, & Schuster, 2001; Onwuegbuzie et al., 2000), theteacher conceptions of excellence in this study were rather morepluralistic; that is, excellence included, to some extent, multipleteaching conceptions ranging from the examination-oriented,teacher-centered perspective to the learning-oriented student-centered perspective. The bias, however, was definitely toward thestudent-centered views, unlike Cuban’s (2007) findings, that“teacher classroom practices” in the Unites States “hugged themiddle of the continuum” (p. 6). The reasons for the strongerstudent-centered orientation in conceiving of teaching excellence

J. Chen et al. / Teaching and Teacher Education 28 (2012) 936e947944

and important teaching practices seem to lie in Confucianapproaches to teaching (e.g., LaoTzu’s aphorism ‘yi ri wei shi, zhongsheng wei fu’; a teacher for a day equals a father for life) and in theearly adolescent stage of human development in which theteachers were working (i.e., Chinese teachers consider that middleschool students are not yet ready for the responsibility of highschool and the pressures of performance for examinations).

It can be seen that teachers’ conception of Excellence As Being AProfessional Learner was related to both teacher-oriented andstudent-oriented teaching practices. Some contextual factors maycontribute to this plurality (Beach, 1994; Brown, Lake, & Matters,2011; Fang, 1996; Lim & Chai, 2008; Sapon-Shevin, 1991;Windschitl, 2002). The three most likely factors are expounded.Firstly, the assessment systemmatters since examinations for entryto high school play a large part of the third year of middle school(i.e., Year 9). Secondly, cultural consequences may influenceteachers’ conceptions (Brown, Lake, et al., 2011) since the students’high school entrance examination results are used to demonstratethe quality of school performance and the higher these are themore highly regarded a middle school is, the more readily schoolswill be able to attract excellent students, recruit excellent teachers,and obtain more funding and support from government. Thirdly,educational bureaus and school leadership may have high expec-tations for student performance in the high school entranceexamination, and therefore, set stringent requirements on teachingpractices.

Hence, teachers are required to participate in professionaltraining programs focused on acquiring the necessary knowledgeto help students succeed in the high-stakes high school entranceexaminations. Under this pressure, improving student achievementbecomes a practice related to the professional status of the teacher,as much as it is a systemic requirement. This, however, as thissample of teachers has indicated, is only one element of excellentteaching. There are many student-oriented values and practicesthat characterize middle school teacher thinking and attitudes inChina. Nonetheless, the role of examinations as an aspect ofexcellent teaching is a complex phenomenon.

5.1. Understanding the Chinese aspects of middle school teacherconceptions

Under the influence of the Confucian culture, Chinese peopleshave an extremely high regard for education. Traditionally, anindividual’s merit, worth, and value were ascribed throughacademic performance (China Civilisation Centre, 2007) and highexamination results reflect the moral quality of an individual. Highresults depend on a serious commitment to loving learning andperfecting oneself morally (Li, 2009; Tsui &Wong, 2009). The recipefor success, then, involves hard work by the students themselves,combined with having devoted teachers and supportive parents(Feng, 1994; Smith, 1992).

One surprising major finding was that the conception of Excel-lence As Examination-oriented held the lowest mean score andwasleast correlated with other conceptions. Under the influence ofConfucian culture, examinations play a crucial role in the Chinesemiddle schools since enrollment in high schools depends heavilyon students’ examination scores (Liu, 2004, 2005). Gao andWatkins (2002) also found that Chinese science teachers seemedto regard improving student examination performance as the mostimportant indicator of excellent teaching. Thus, Excellence AsExamination-oriented was expected to have had a higher meanscore and be highly correlated with other conceptions. This,however, was not the case.

A number of explanation are possible. Firstly, with the recent,increasing openness of China to the West (e.g., teachers traveling

and/or studying outside the country) it may be that teachers’thinking has been influenced by western educational awarenessand ideology (Ding, 2001). Secondly, the Basic Education Curric-ulum Reform issued by the Ministry of Education in China in 2001,which has the objective of changing pedagogical approaches fromexamination-oriented education to quality-oriented education,criticizes indoctrination, and requires teachers to actively interactwith students, cultivate a spirit of innovation, and individuality ofstudents, may have had an impact (Dello-Iacovo, 2009; Ministry ofEducation (MOE) of the People’s Republic of China, 2001). Thirdly, itseems only rational that excellent teachers, in an ecology thatjudges school quality by student examination results as it does inChina, should want their professionalism to support students’doing well in examinations.

Indeed, it is ecologically rational (Rieskamp & Reimer, 2007) thata professional, excellent teacher should engage in examination-related practices given the importance examinations have hadhistorically and culturally for Chinese student life-chances (ChinaCivilisation Centre, 2007). Hence, it may be that teachers inChinese middle schools are aware that examinations should notplay such a significant role in modern China, whilst still acceptingtheir place in Chinese middle school education.

Being Strictly Teacher-oriented in the TP model, however,accounts for a relatively important place in Chinese middle schoolteaching practices and stands in contrast to the three student-oriented practices identified in the same model. One probablereason for the relative prominence of these practices is the highschool entrance examination that may push teachers, students,school leaders, and parents to put a high value on the examinationin Chinese middle schools. The teachers are being strict aroundexaminations in order to help students to achieve high outcomes.This is consistent with other studies of how Chinese teacherconceive assessment purposes which showed that the student andschool accountability roles of assessment were highly correlatedwith improvement of teaching and learning (Brown, Hui, Yu, &Kennedy, 2011; Brown, Kennedy, Fok, Chan, & Yu, 2009).

5.2. Contrasting Chinese middle school teachers’ conceptions toWestern perspectives

Compared to the Western literature on teaching excellence,these Chinese teachers did not conceive of excellence in terms ofworking collaboratively with school boards and communities orunderstanding state and local educational objectives. Further,Chinese teachers seemed to have little awareness of curriculumdevelopment, whilst the NBPTS model (2009) and other Westernliterature (e.g., Lee, 2006; McGee, 2006) stress this conception. Thisresult, however, is consistent with the context in China in whichcurriculum development is more a top-down set of regulations andteachers do not have opportunities to be involved in such devel-opment. Chinese teachers teach subject knowledge to studentsusing the assigned textbooks and teacher reference books. Thetextbooks provide the curricula that teachers follow and theteacher reference books provide the goals and objectives thatstudents are required to achieve.

These Chinese teachers endorsed and gave much importance tobeing involved in activities outside of the classroom, such asextracurricular activities, after-school activities, home calls or visits,and meetings with parents. The Western literature on teachingexcellence rarely mentions these practices or approaches eventhough a strong relationship between students and teachers isadvocated as one of characteristics of excellent teaching (Reilly,Lilly, Bramwell, & Kronish, 2011). These Chinese teachers alsoexpected to be holistic role models for students, which is absentfrom the Western literature.

J. Chen et al. / Teaching and Teacher Education 28 (2012) 936e947 945

5.3. Implications

The findings about excellent teaching reported here maycontribute to the development of advanced teaching standardsthat are absent in China but have been employed in the West fordecades. The TCET model was established through a large sampleof teachers. Thus, this model could provide policymakers andresearchers with a baseline for designing teaching standards formiddle schools and advanced teaching standards for teacherpromotion and evaluation systems. The current evaluation andpromotion standards in Chinese middle schools include two majorsections (i.e., moral and political requirements and academicrequirements), with the former accounting for more than half ofthe total number of pages in the document. Academic require-ments do not address details about teaching practice. It is expectedthat this study could provide a reference point for establishinga sound basis for developing excellent teaching in Chinese middleschools.

The current study may make a contribution to professionaldevelopment and teaching improvement. Since teachers’ concep-tions of excellent teaching, teaching practices, and their relation-ship have been identified, teachers may be able to reflect anddiscover their own conceptions of excellent teaching that havebeen shaping their daily teaching. Through a critical self-evaluation, teachers may try to dispense with their ineffectiveroutines and be willing to take up the challenges entailed in therenewal endeavor (Pang, 1999) to achieve professional preparationimprovement (Clark, 1988) and later teaching excellence (Hofer,1994). On the other hand, researchers and policymakers coulddesign and implement effective professional developmentprograms which could result in changes in teacher conceptions,and, in turn have, an impact on their actual teaching practices.Further, exploring teachers’ conceptions of excellent teaching andtheir relation to teaching practices is essential in order to facilitatethe changes (Sakofs, Armstrong, Proudman, Howard, & Clark,1995).Thus, the findings could inform both current teachers’ professionaldevelopment and future teachers’ aspirations, which, in turn, couldlead to an improvement in teaching.

This study also alerts us to the importance of examinations as anindependent aspect of excellence and as part of being a professionalrather than being for student development in Chinese middleschools. Chinese policymakers and the school boards shouldconsider why this has appeared and played an important role forexcellent teaching in Chinese middle school. The National Outlinefor Mid and Long-Term Education Reform and Development2010e2020 (MOE of the People’s Republic of China, 2010) hasstressed only that the university entrance examination should bechanged in the next ten years, but has ignored the high schoolentrance examination which has an influential impact on teachingin basic education including middle schools. Policymakers mayconsider setting an urgent agenda for changing the high schoolentrance examination in such a policy as well, and then researchersand the school boards may set down specific strategies to imple-ment these in teachers’ professional development.

A limitation to this study was the relatively low response rate ofonly 43.2%. This may be a consequence of having the questionnairesreturned by mail, which is not a robustly reliable mechanism in allparts of China. Alternative collection methods (e.g., face-to-facecollection or the creation of a drop-box within participatingschools) should be considered in future studies. The relatively lowresponse rate may mean that the results reflect the views ofteachers who perceive their own practices as ‘excellent’. Nonethe-less, as an initial study of middle school teachers in China, sucha bias, should it exist, is effective in identifying how such teachersconceive of excellence.

It is recommended for further research to use multiple datasources in the form of classroom observations, teacher interviews,and focus-group discussions as this would make the data morevalid given that self-reported practices do not necessarily reflectteachers’ actual practices. It would be valuable to have teacherfocus-group discussions to explore the reasons for the key findings,the exceptions, and the surprising findings. Furthermore, if studentachievement was included in the conceptions of excellent teachingand teaching practice, the classic triangle, conception-practice-achievement, could be established. If the relationships betweenthese three aspects or between any two of them could be estab-lished, such a study would be most valuable. This could then lead toa series of professional development programs, hopefully resultingin the improvements to teaching practice, and, thence, to studentlearning gains.

References

Ajzen, I. (2002). Residual effects of past on later behavior: habituation and reasonedaction perspectives. Personality and Social Psychology Review, 6(2), 107e122.doi:10.1207/S15327957PSPR0602_02.

Ajzen, I. (2005). Attitudes, personality, and behavior (2nd ed.). Milton-Keynes, UK:Open University Press/McGraw-Hill.

Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice:a review and recommended two-step approach. Psychological Bulletin, 103(3),411e423.

Arbuckle, J. L. (2009). AMOS (Version 18.0.0) [computer program]. Crawfordville, FL:Amos Development Corporation.

Beach, S. (1994). Teacher’s theories and classroom practice. Reading Psychology,15(3), 189e196.

Beishuizen, J. J., Hof, E., Van Putten, C. M., Bouwmeester, S., & Asscher, J. J. (2001).Students’ and teachers’ cognitions about good teachers. British Journal ofEducational Psychology, 71, 185e201. doi:10.1348/000709901158451.

Betoret, F. D., & Artiga, A. G. (2004). Trainee teachers’ conceptions of teaching andlearning, classroom layout and exam design. Educational Studies, 30(4),355e372. doi:10.1080/0305569042000310309.

Boston, M. D., & Smith, M. S. (2009). Transforming secondary mathematicsteaching: increasing the cognitive demands of instructional tasks used inteachers’ classrooms. Journal for Research in Mathematics Education, 40(2),119e156.

Brown, G. T. L. (2004). Measuring attitude with positively packed self-reportratings: comparison of agreement and frequency scales. Psychological Reports,94, 1015e1024. doi:10.2466/pr0.94.3.1015-1024.

Brown, G. T. L., Hui, S. K. F., Yu, W. M., & Kennedy, K. J. (2011). Teachers’ conceptionsof assessment in Chinese contexts: a tripartite model of accountability,improvement, and irrelevance. International Journal of Educational Research,50(5), 307e320. doi:10.1016/j.ijer.2011.10.003.

Brown, G. T. L., Kennedy, K. J., Fork, P. K., Chan, J. K. S., & Yu, W. M. (2009).Assessment for student improvement: understanding Hong Kong teachers’conceptions and practices of assessment. Assessment in Education: Principles,Policy & Practice, 16(3), 347e363.

Brown, G. T. L., Lake, R., & Matters, G. (2009). Assessment policy and practiceeffects on New Zealand and Queensland teachers’ conceptions of teaching.Journal of Education for Teaching, 35(1), 61e75. doi:10.1080/02607470802587152.

Brown, G. T. L., Lake, R., & Matters, G. (2011). Queensland teachers’ conceptions ofassessment: the impact of policy priorities on teacher attitudes. Teaching andTeacher Education, 27(1), 210e220. doi:10.1016/j.tate.2010.08.003.

Browne, M. W., & Cudeck, R. (1989). Single sample cross-validation indices forcovariance structures. Multivariate Behavioral Research, 24, 445e455.doi:10.1207/s15327906mbr2404_4.

Byrne, B. M. (2010). Structural equation modeling with AMOS: Basic concepts, appli-cations, and programming (2nd ed.). New York, NY: Routledge.

Chen, J. (2007). Teachers’ conceptions of excellent teaching in middle schools in thenorth of China. Asia Pacific Education Review, 8(2), 288e297. doi:10.1007/BF03029263.

Chen, J. (2010). Teachers’ conceptions of excellent teaching in Chinese middle schools(Doctoral thesis, The University of Auckland, New Zealand), Retrieved from. http://hdl.handle.net/2292/5864.

Cheng, H., & DeLany, B. (1999). Quality education and social stratification: theparadox of private schooling in China. Comparative Education, 1(2).

Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indices fortesting measurement invariance. Structural Equation Modeling, 9(2), 233e255.doi:10.1207/S15328007SEM0902_5.

China Civilisation Centre. (2007). China: Five thousand years of history and civiliza-tion. Hong Kong: City University of Hong Kong Press.

Chiu, Y. I. (2006). Cross-cultural comparison of teacher beliefs about discipline andbehavior management, teaching practices, and students (Unpublished doctoraldissertation). University of Virginia, Virginia.

J. Chen et al. / Teaching and Teacher Education 28 (2012) 936e947946

Clark, C. M. (1988). Asking the right questions about teacher preparation: contri-butions of research on teacher thinking. Educational Researcher, 17(2), 5e12.

Cortazzi, M., & Jin, L. (1996). English teaching and learning in China. LanguageTeaching, 29(2), 61e80. doi:10.1017/S0261444800008351.

Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory factor analysis:four recommendations for getting the most from your analysis. PracticalAssessment Research and Evaluation, 10(7), Retrieved from. http://pareonline.net/pdf/v10n7.pdf.

Cuban, L. (2007). Hugging the middle: teaching in an era of testing and account-ability 1980e2005. Educational Policy Analysis Archives, 15(1), 1e29, Retrievedfrom. http://hub.mspnet.org/index.cfm/13877.

Darling-Hammond, L., & Youngs, P. (2002). Defining “highly qualified teachers”:what does “scientifically-based research” actually tell us? EducationalResearcher, 31(13).

Dello-Iacovo, B. (2009). Curriculum reform and quality education in China: anoverview. International Journal of Educational Development, 29, 241e249.

Dempster, A. P., Laird, N. M., & Rubin, D. B. (1977). Maximum likelihood estimationfrom incomplete data via the EM algorithm. Journal of the Royal StatisticalSociety, 39, 1e38. doi:10.2307/2984875.

Ding, G. (2001). Nationalization and internationalization: two turning points inChina’s Education in the twentieth century. In G. Peterson, R. Hayhoe, & Y. Lu(Eds.), Education, culture, and identity in twentieth-century China (pp. 161e186).Ann Arbor, MI: The University of Michigan Press.

Fan, X., & Sivo, S. A. (2005). Sensitivity of fit indices to misspecified structural ormeasurement model components: rationale of two-index strategy revisited.Structural Equation Modeling, 12(3), 343e367. doi:10.1207/s15328007sem1203_1.

Fan, X., & Sivo, S. A. (2007). Sensitivity of fit indices to model misspecification andmodel types. Multivariate Behavioral Research, 42(3), 509e529. doi:10.1080/00273170701382864.

Fang, Z. (1996). A review of research on teacher beliefs and practices. EducationalResearch, 38, 47e65.

Feng, J. (1994). AsianeAmerican children: What teachers should know? Urbana, IL:ERIC Clearinghouse on Elementary and Early Childhood Education.

Feng, D. (2006). China’s recent curriculum reform: progress and problems. Planningand Changing, 27(12), 131e144.

Fenstermacher, G. D., & Soltis, J. F. (1998). Approaches to teaching (3rd ed.). NewYork: Teachers College Press.

Gao, L. (1992). The generative style of teaching: Rationale and practice. Paper pre-sented at the meeting of the ICTPE, Dortmund, Germany.

Gao, L., Adcock, H., Carr, M., Hume, A., Nicholson, D., Silvester, J., et al. (1989). Anaccount of action research investigating teacher change. Research in ScienceEducation, 19(1), 112e122. doi:10.1007/BF02356851.

Gao, L., & Watkins, D. A. (2002). Conceptions of teaching held by school scienceteachers in P.R. China: identification and cross-cultural comparisons. ScienceEducation, 24(1), 61e79. doi:10.1080/09500690110066926.

Gow, L., & Kember, D. (1993). Conceptions of teaching and their relationship tostudent learning. British Journal of Educational Psychology, 63, 20e33.

Hay McBer. (2000). Research into teacher effectiveness: A model of teacher effective-ness. Report to the Department for Education and Employment. Retrieved from.https://www.education.gov.uk/publications/RSG/Teachersandschoolworkforce/Page11/RR216.

Ho, A., Watkins, D., & Kelly, M. (2001). The conceptual change approach toimproving teaching and learning: an evaluation of a Hong Kong staff devel-opment programme. Higher Education, 42, 143e169.

Hofer, B. K. (1994, August). Epistemological beliefs and first year college students:Motivation and cognition in different instructional contexts. Paper presented atthe Annual Meeting of the American Psychological Association, 102nd, LosAngeles.

Hoyle, R. H. (1995). The structural equation modeling approach: basic concepts andfundamental issues. In R. H. Hoyle (Ed.), Structural equation modeling: Concepts,issues, and applications (pp. 1e15). Thousand Oaks, CA: Sage.

Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indices in covariance structureanalysis: conventional criteria versus new alternatives. Structural EquationModeling, 6, 1e55. doi:10.1080/10705519909540118.

Hughes, I., & Yuan, L. (2005). The status of action research in the People’s Republicof China. Action Research, 3(4), 383e402. doi:10.1177/1476750305058488.

Kane, R. G., Sandretto, S., & Heath, C. (2002). Telling half the story: a critical reviewof the research into tertiary teachers’ beliefs. Review of Educational Research,72(2), 177e228, Retrieved from. http://www.jstor.org/stable/3516032.

Kane, R. G., Sandretto, S., & Heath, C. (2004). An investigation into excellent tertiaryteaching: Emphasizing reflective practice. Higher Education, 47(3), 283e310.

Kelly, E., & Watkins, D. A. (2002). A comparison of the goals and approaches toteaching of expatriate and Chinese lecturers at universities in Hong Kong. Paperpresented at the Hong Kong branch of the Higher Education Research andDevelopment Society of Australasia, City University of Hong Kong, Hong Kong.

Kember, D. (1997). A reconceptualization of the research into university academics’conceptions of teaching. Learning and Instruction, 7(3), 255e275. doi:10.1016/S0959-4752(96)00028-X.

Kember, D., & Kwan, K. (2000). Lecturers’ approaches to teaching and their rela-tionship to conceptions of good teaching. Instructional Science, 28, 469e490.doi:10.1023/A:1026569608656.

Kember, D., & Wong, A. (2000). Implications for evaluation from a study of students’perceptions of good and poor teaching. Higher Education, 40, 69e97.

Kenny, D. A. (2008). Measuring model fit. Retrieved from. http://davidakenny.net/cm/fit.htm.

Klem, L. (2000). Structural equation modeling. In L. G. Grimm, & P. R. Yarnold (Eds.),Reading and understanding more multivariate statistics (pp. 227e260). Wash-ington, DC: APA.

Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.).New York: The Guilford Press.

Kottler, J. A., & Zehm, S. J. (2000). On being a teacher (2nd ed.). Thousand Oaks, CA:Corwin.

Kottler, J. A., Zehm, S. J., & Kottler, E. (2005). On being a teacher: The humandimension. Thousand Oaks, CA: Corwin Press.

Lam, T. C. M., & Klockars, A. J. (1982). The influence of Labels and Positions in RatingScales. Journal of Educational Measurement, 19, 312e322.

Lee, T. T. D. (2006). Elementary school educators’ perceptions of needed competencies,beyond those taught in a teacher education program, to be effective elementaryschool teachers in Taiwan (Unpublished doctoral dissertation). Idaho StateUniversity, Idaho.

Lerner, J. S., & Tetlock, P. E. (1999). Accounting for the effects of accountability.Psychological Bulletin, 125(2), 255e275.

Li, J. (2009). Learning to self-perfect: Chinese beliefs about learning. In C. K. K. Chan,& N. Rao (Eds.), Revisiting the Chinese learner: Changing contexts, changingeducation (pp. 35e69). Hong Kong: Springer/The University of Hong Kong,Comparative Education Research Centre.

Lim, C. P., & Chai, C. S. (2008). Teachers’ pedagogical beliefs and their planning andconduct of computer-mediated classroom lessons. British Journal of EducationalTechnology, 39(5), 807e828. doi:10.1111/j.14678535.2007.00774.x.

Liu, B. (1995). Some ideas about quality education. People’s Education Beijing, 7, 8e12.Liu, H. (2004). Imperial examination system and study of imperial examination.

Guiyang, China: Guizhou Educational Press.Liu, H. (2005). The influence of imperial examination on the educational assessment in

contemporary China. Paper presented at the meeting of the Future of Educa-tional Assessment: East meets West. Kuala Lumpur, Malaysia.

McGee, P. J. (2006). Qualities of effective high school teachers in grades 9e12:Perceptions of high school teachers and administrators from the State of RhodeIsland (Unpublished doctoral dissertation). Johnson and Wales University,Providence, Rhode Island.

Marsh, H. W., & Hocevar, D. (1985). Application of confirmatory factor analysis tothe study of self-concept: first- and higher order factor models and theirinvariance across groups. Psychological Bulletin, 97(3), 562e582. doi:10.1037/0033-2909.97.3.562.

Maxwell, K. L., McWilliam, R. A., Hemmeter, M. L., Ault, M. J., & Schuster, J. W.(2001). Predictors of developmentally appreciate classroom practices inkindergarten through third grade. Early Childhood Research Quarterly, 16,431e452. doi:10.1016/S0885-2006(01)00118-1.

Ministry of Education of the People’s Republic of China. (2001). Guidelines for basiceducation curriculum reform (pilot). Retrieved from. http://www.moe.edu.cn/edoas/website18/32/info732.htm.

Ministry of Education of the People’s Republic of China. (2010). National Outline forMid and long-term education reform and development 2010-2020. Retrieved from.http://www.gov.cn/jrzg/2010-07/29/content_1667143.htm.

Minor, L. C., Onwuegbuzie, A. J., Witcher, A. E., & James, T. L. (2002). Pre-serviceteachers’ educational beliefs and their perceptions of characteristics of effectiveteachers. Journal of Educational Research, 96(2), 116e127. doi:10.1080/00220670209598798.

Nager, N., & Shapiro, E. K. (2000). Revisiting progress pedagogy. New York: StateUniversity of New York Press.

National Board for Professional Teaching Standards (NBPTS). (1987). General infor-mation about national board certification. Retrieved from. http://www.nbpts.org/standards/nbcert.cfm.

National Board for Professional Teaching Standards (NBPTS). (2009). What teachersshould know and be able to do: National Board for Professional Teaching Standards.Retrieved from. http://www.nbpts.org/UserFiles/File/what_teachers.pdf.

Onwuegbuzie, A. J., Witch, A. E., Collins, K. M. T., Filer, J. D., Wiedmaier, C. D., &Moore, C. W. (2007). Students’ perceptions of characteristics of effective collegeteachers: a validity study of a teaching evaluation form using a mixed-methodsanalysis. American Educational Research Journal, 44(1), 113e160. doi:10.3102/0002831206298169.

Onwuegbuzie, A. J., Witch, A. E., Filer, J., & Downing, J. (2000). Factors associated withteachers’ beliefs on discipline. Paper presented at the Annual Meeting of the Mid-South Educational Research Association, Lexington, KY.

Paine, L. W. (1992). Teaching and modernization in contemporary China. InR. Hayhoe (Ed.), Education and modernization: The Chinese experience (pp.183e209). Oxford: Pergamon.

Pang, W. (1999). The interface between theory and practice: The role of teachereducators and teachers in a school-based teacher development initiative. Retrievedfrom. http://www.aare.edu.au/99pap/won99474.htm.

Pratt, D. D. (1992). Conceptions of teaching. Adult Education Quarterly, 42(4),203e220.

Reilly, R. C., Lilly, F., Bramwell, G., & Kronish, N. (2011). A synthesis of researchconcerning creative teachers in a Canadian context. Teaching and TeacherEducation, 27(3), 533e542. doi:10.1016/j.tate.2010.10.007.

Rieskamp, J., & Reimer, T. (2007). Ecological rationality. In R. F. Baumeister, &K. D. Vohs (Eds.), Encyclopedia of social psychology (pp. 273e275). ThousandOaks, CA: Sage.

Rittle-Johnson, B., Siegler, R. S., & Alibali, M. W. (2001). Developing conceptualunderstanding and procedural skill in mathematics: an iterative process. Jour-nal of Educational Psychology, 93, 346e362.

J. Chen et al. / Teaching and Teacher Education 28 (2012) 936e947 947

Sakofs, M., Armstrong, G., Proudman, S., Howard, J., & Clark, T. (1995). Developinga teacher development model: a work in progress. The Journal of ExperientialEducation, 18(3), 128e132.

Samuelowicz, K. (1994). Teaching conceptions and teaching practice: a case ofassessment. In R. Ballantyne, & C. Bruce (Eds.), Phenomenography: Philosophyand practice (pp. 343e353). Brisbane, Aus: Queensland University of Tech-nology, Centre for Applied Environmental and Social Education Research.

Sapon-Shevin, M. (1991). As we teach, we change our students and ourselves.Teaching Education, 4(1), 63e67.

Skamp, K., & Mueller, A. (2001). Student teachers’ conceptions about effectiveprimary science teaching: a longitudinal study. International Journal of ScienceEducation, 23(4), 331e351. doi:10.1080/09500690119248.

Smith, D. C. (1992). The Chinese family in transition: Implications for education andsociety in modern Taiwan. Paper presented at the Comparative EducationAssociation and World Bank Seminar, Annapolis, Maryland.

Tang, D. G., & Absalom, D. (1998). Teaching across cultures: considerations forWestern EFL teachers in China. Hong Kong Journal of Applied Linguistics, 3(2),117e132.

Thompson, B. (2000). Ten commandments of structural equation modeling. InL. G. Grimm, & P. R. Yarnold (Eds.), Reading and understanding more multivariatestatistics (pp. 261e283). Washington, DC: APA.

Tsui, A. B. M., & Wong, J. L. N. (2009). In search of a third space: teacher develop-ment in mainland China. In C. K. K. Chan, & N. Rao (Eds.), Revisiting the Chineselearner: Changing contexts, changing education (pp. 281e311). Hong Kong:Springer/The University of Hong Kong, Comparative Education Research Centre.

Vandenberg, R. J., & Lance, C. E. (2000). A review and synthesis of the measurementinvariance literature: suggestions, practices, and recommendations for organi-zational research. Organizational Research Methods, 3, 4e70. doi:10.1177/109442810031002.

Wang, D. (2003). China’s rural compulsory education: Current situation, problems andpolicy alternatives. Retrieved from. http://iple.cass.cn/file/36.pdf.

Watkins, D. A., & Biggs, J. B. (2001). Teaching the Chinese learner: Psychological andpedagogical perspectives. The University of Hong Kong, Hong Kong:Comparative Education Research Center & Australian Council of EducationalResearch.

Watkins, D. A., & Sachs, J. (2004). Comparing concepts across cultures: The applicationof dual scaling to analyze perceptions of the ‘good teacher’ (Unpublished doctoraldissertation). The University of Hong Kong, Hong Kong.

Watkins, D. A., & Zhang, Q. (2006). The good teacher: a cross-cultural perspective. InD. McInerney, M. Dowson, & S. van Etten (Eds.), Effective schools (pp. 185e204).Greenwich: Connecticut Information Age Publishing.

Wegener, D. T., & Fabrigar, L. R. (2000). Analysis and design for nonexperimentaldata: addressing causal and noncausal hypotheses. In H. T. Reis, & C. M. Judd(Eds.), Handbook of research methods in social and personality psychology (pp.412e450). Cambridge, UK: Cambridge University Press.

Windschitl, M. (2002). Framing constructivism in practice as the negotiationof dilemmas: an analysis of the conceptual, pedagogical, cultural, andpolitical challenges facing teachers. Review of Educational Research, 72(2),131e175.

Witcher, A. E., Onwuegbuzie, A. J., & Minor, L. C. (2001). Characteristics of effectiveteachers: perceptions of preservice teachers. Research in the Schools, 8(2),45e57.

Zhang, Q. (2007). Conceptions of a good English Language Teacher at Tertiary Level inthe People’s Republic of China (Unpublished doctoral dissertation). The Univer-sity of Hong Kong, Hong Kong.

Zhang, J., & Collis, B. (1995). Comparison of teaching models in the West and inChina. E-Journal of Instructional Science and Technology, 1(1), 3, Retrieved from.http://doc.utwente.nl/26914/.

Zhao, W. (2007). Transformation in teaching practice of Chinese teachers blendingwestern and Chinese educational approaches for orphan children in China: A mixedmethods study (Unpublished doctoral dissertation). The University of Nebraska,Nebraska.