[ Proceedings Contents ] [ Schedule ] [ Abstracts ] [ WAIER Home ]
Teacher and student perceptions of classroom interactions: A multi-level modelTony RickardsUniversity of Western Australia
Michael Newby
Darrell Fisher |
The purpose of the study was to compare students' perceptions of teacher-student interactions with those of their teachers by administering the Questionnaire on Teacher Interaction (QTI) to teachers and students in 80 lower secondary classes in schools in Tasmania and Western Australia. There are three possible versions of the QTI. Students completed the student version which assesses the students' perceptions of the teacher-student interactions in a specific class. Their teachers completed the teacher actual version of how they perceived their interactions with their students in those same classes. The teachers also indicated how they thought ideal teachers would interact with students by responding to the teacher ideal version. Previous statistical analysis had confirmed the reliability and validity of the QTI for secondary school students. Two multilevel models were proposed: the teacher ideal interaction influences the teacher actual interaction; and the teacher actual affects the student actual and vice versa. Using structural equation modelling techniques, both models were found to be reasonable fits to the data. The results would seem to confirm the underlying basis of the QTI in that the teachers' actual perceptions of their interactions with students affects the students' perceptions, which in turn affect the teachers' perceptions.
Learning environment research on teacher-student interaction began in The Netherlands where Wubbels, Créton, and Hooymayers (1985) developed a model to map teacher-student interactions extrapolated from the work of Leary (1957). This model was used in the development of an instrument, the Questionnaire on Teacher Interaction (QTI), to gather students' and teachers' perceptions of teacher-student interactions (Wubbels, Brekelmans, & Hooymayers, 1991; Wubbels & Levy, 1993). This model maps these interactions with the aid of an influence dimension (Dominance, D - Submission, S) and a proximity dimension (Cooperation, C - Opposition, O). These dimensions are represented in a coordinate system divided into eight equal sectors.
Every instance of interactional teacher behaviour can be placed within this system of axes. The closer the instances of behaviour are in the chart, the more closely they resemble each other. The sections are labelled DC, CD, CS, SC, SO, OS, OD and DO according to their position in the coordinate system. For example, the two sectors DC and CD are both characterised by Dominance and Cooperation. In the DC sector, however, the Dominance aspect prevails over the Cooperation aspect, whereas in the adjacent sector CD Cooperation prevails over the Dominance aspect (Wubbels, Créton, Levy, & Hooymayers, 1993). Table 1 shows the names of the behaviours (eg., leadership behaviour, helping/friendly behaviour, understanding behaviour) given to each sector. These sector names are the names given to the eight scales of the QTI. Table 1 clarifies further the nature of the QTI by providing a scale description and a sample item for each of the eight scales.
Scale name | Description of scale (The extent to which the teacher...) | Sample item |
Leadership | ...leads, organises, gives orders, determines procedure and structures the classroom situation. | This teacher talks enthusiastically about his/her subject. |
Helping/ friendly | ...shows interest, behaves in a friendly or considerate manner and inspires confidence and trust. | This teacher helps us with our work. |
Understanding | shows confidence and understanding and is open with students. | This teacher trusts us. |
Student responsibility/freedom | ...gives opportunity for independent work, gives freedom and responsibility to students. | We can decide some things in this teacher's class. |
Uncertain | ...behaves in an uncertain manner and keeps a low profile. | This teacher seems uncertain. |
Dissatisfied | ...expresses dissatisfaction, looks unhappy, criticises and waits for silence. | This teacher thinks that we cheat. |
Admonishing | ...gets angry, express irritation and anger, forbids and punishes. | This teacher gets angry unexpectedly. |
Strict | ...checks, maintains silence and strictly enforces the rules. | This teacher is strict. |
The information obtained by means of the questionnaire includes perceptions of the behaviour of the teacher towards the students as a class, and reflects relatively stable patterns of behaviour over a considerable period. One advantage of the QTI is that it can be used to obtain the perceptions of interpersonal behaviour of either students or teachers (Rickards and Fisher, 2000; Fisher & Rickards, 2000). When the QTI is administered to both teachers and their students, information is provided about the perceptions of teachers and the perceptions of their students of the teacher-student interactions occurring in the classroom. of that teacher. The wording of the questionnaire is varied slightly when used to obtain teachers' self-perceptions. For example the question "This teacher talks enthusiastically about his/her subject", becomes "I talk enthusiastically about my subject" in the teacher self-perception version, and "This teacher would talk enthusiastically about his/her subject" in the teacher ideal version.
Another use of the QTI in The Netherlands involved investigation of relationships between perceptions on the QTI scales and student learning outcomes (Wubbels, Brekelmans, & Hooymayers, 1991). Regarding students' cognitive outcomes, the more that teachers demonstrated strict, leadership and helpful/friendly behaviour, then the higher were cognitive outcomes scores. Conversely, student responsibility and freedom, uncertain and dissatisfied behaviours were related negatively to achievement.
When teachers described their perceptions of their own behaviours, they tended to see it a little more favourably than did their students. On average, the teachers' perceptions were between the students' perceptions of actual behaviour and the teachers' ideal behaviour. An interpretation of this is that teachers think that they behave closer to their ideal than their students think they do. Variations in the students' attitudes toward the subject and the lessons have been characterised on the basis of the proximity dimension: the more cooperative the behaviour displayed, the higher the affective outcome scores (Wubbels, Brekelmans, & Hooymayers, 1991). That is, student responsibility and freedom, understanding, helping/friendly and leadership behaviours were related positively to student attitudes. Uncertain, dissatisfied, admonishing and strict behaviours were related negatively to student attitudes. Overall, previous studies have indicated that interpersonal teacher behaviour is an important aspect of the learning environment and that it is related strongly to student outcomes.
Levy, Créton, and Wubbels (1993) analysed data from studies in The Netherlands, the USA and Australia involving students being asked to use the QTI to rate their best and worst teachers. Students rated their best teachers as being strong leaders and as friendly and understanding. The characteristics of the worst teachers were that they were more admonishing and dissatisfied.
Levy, Wubbels, Brekelmans, and Morganfield (1997) investigated a sample of 550 high school students in 38 classes comprised of three primary investigation groups, namely 117 Latinos, 111 Asians and 322 from the United States. The primary focus was the language and cultural factors in students' perceptions of teacher communication style. This study focused on identifying ways in which the students' culture relates to student perceptions of their teachers. The results from this study suggested that the students' cultural background is indeed significantly related to the perceptions that they had of their teachers' interaction behaviour. The study also concluded that teachers do not seem to be aware of cultural differences in their interactions with students in their classes in the same way as their students were, despite altering their behaviour in classes with different cultural compositions.
The Australian version of the QTI containing 48 items was used in a pilot study involving upper secondary science classes in Western Australia and Tasmania (Fisher, Fraser, & Wubbels, 1993; Fisher, Fraser, Wubbels, & Brekelmans, 1993; Fisher, Henderson, & Fraser, 1995). This pilot study strongly supported the validity and potential usefulness of the QTI within the Australian context, and suggested the desirability of conducting further and more comprehensive research involving the QTI.
Wubbels (1993) used the QTI with a sample of 792 students and 46 teachers in Western Australia and Tasmania. The results of this study were similar to previous Dutch and American research in that, generally, teachers did not reach their ideal and differed from the best teachers as perceived by students. It is noteworthy that the best teachers, according to students, are stronger leaders, more friendly and understanding, and less uncertain, dissatisfied and admonishing than teachers on average. When teachers described their perceptions of their own behaviours, they tended to see it a little more favourably than did their students. On average, the teachers' perceptions were between the students' perceptions of actual behaviour and the teachers' ideal behaviour. An interpretation of this is that teachers think that they behave closer to their ideal than their students think that they do.
Scale | Teacher Actual | Teacher-Ideal | Student-Class mean |
Leadership | .88 | .81 | .93 |
Helping/ friendly | .92 | .86 | .96 |
Understanding | .88 | .83 | .95 |
Student resp/ freedom | .79 | .65 | .82 |
Uncertain | .78 | .69 | .87 |
Dissatisfied | .84 | .78 | .93 |
Admonishing | .79 | .75 | .87 |
Strict | .72 | .62 | .78 |
n = 1659 student s and 72 teachers in 80 classes. |
Figure 1: Model of Teacher_Ideal - Teacher_Actual interactions
In it there are two latent variables Teacher_ideal and Teacher_actual. The latent variable Teacher_ideal affects the responses to the questions relating to the scales of the Teacher Ideal version of the QTI and the variable Teacher_actual relates to the responses to questions relating to the scales of the Teacher Actual version of the QTI. It is hypothesised that there is a causal relationship between Teacher_ideal and Teacher_actual.
Relationship | B | beta |
Teacher_Actual <-- Teacher_Ideal | 4.45 | 0.85** |
Leadership_Ideal <-- Teacher_Ideal | 0.80 | 0.28* |
Helping_Ideal <-- Teacher_Ideal | 1.00 | 0.87^{#} |
Understanding_Ideal <-- Teacher_Ideal | 0.76 | 0.86** |
Student_Resp_Ideal <-- Teacher_Ideal | 0.00 | 0.00 |
Uncertain_Ideal <-- Teacher_Ideal | -0.74 | -0.27* |
Dissatisfied_Ideal <-- Teacher_Ideal | -0.85 | -0.60** |
Admonishing_Ideal <-- Teacher_Ideal | -0.05 | -0.04 |
Strict_Ideal <-- Teacher_Ideal | -0.97 | -0.72** |
Leadership_Actual <-- Teacher_Actual | 0.10 | 0.54** |
Helping_Actual <-- Teacher_Actual | 0.18 | 0.88** |
Understanding_Actual <-- Teacher_Actual | 0.16 | 0.75** |
Student_Resp_Actual <-- Teacher_Actual | -0.03 | -0.06 |
Uncertain_Actual <-- Teacher_Actual | -0.14 | -0.53** |
Dissatisfied_Actual <-- Teacher_Actual | -0.15 | -0.55** |
Admonishing_Actual <-- Teacher_Actual | 0.03 | -0.02 |
Strict_Actual <-- Teacher_Actual | -0.19 | -0.60** |
# used for identification, **p < .01, *p < .05 |
There are many indices for measuring how well a model fits the data (Marcoulides & Hershberger, 1997; Schumaker & Lomax, 1996) and the ones given in Tables 4 and 6 are the chi^{2} degrees of freedom ratio (chi^{2}/ df), the Goodness of Fit Index (GFI), the Adjusted Goodness of Fit Index (AGFI), the Root Mean Square Residual (RMR), and the Root Mean Square of the Error of the Approximation (RMSEA). The ratio chi^{2}/ df is a frequently used measure, and a value of less than 2 is considered to show a very good fit. The GFI measures the relative amount of variance and covariance in the data accounted for by the proposed model. Values in excess of 0.90 are considered to indicate a good fit. The AGFI makes allowance for the complexity of the model and again a figure above 0.90 is considered a good fit. The RMR measures the amount of remaining variance not explained by the model and in figure of 0.05 indicates a good fit. The RMSEA is a measure of the discrepancy of the fitted model per degree of freedom, and a value of 0.05 shows a close fit with a value of 0.08 representing reasonable errors (Jöreskog, 1993).
Table 4 contains the values of these indices for this model. Although all of them are outside the acceptable limits, most are just outside so this would indicate that the model is a reasonable if not a good fit.
Coefficient | Value |
chi^{2} | 292.58 |
df | 106 |
chi^{2}/df | 2.761 |
GFI | 0.853 |
AGFI | 0.791 |
RMR | 0.127 |
RMSEA | 0.083 |
Figure 2: Model of Teacher_Actual - Student_Actual
In it there are two latent variables Teacher_actual and Student_actual. The latent variable Teacher_actual affects the responses to the questions relating to the scales of the Teacher Actual version of the QTI and the variable Student_actual the responses to the questions relating to the scales of the Student Actual version of the QTI. Further, it is hypothesised that there is a causal relationship between Teacher_actual and Student_actual, and also in the reverse direction, Student_actual and Teacher_actual.
Relationship | B | beta |
Teacher_Actual <-- Student_Actual | 0.52 | 0.26* |
Student_Actual <-- Teacher_Actual | 3.75 | 0.72** |
Leadership_Actual <-- Teacher_Actual | 0.10 | 0.54** |
Helping_Actual <-- Teacher_Actual | 0.18 | 0.88** |
Understanding_Actual <-- Teacher_Actual | 0.16 | 0.75** |
Student_Resp_Actual <-- Teacher_Actual | -0.03 | -0.06 |
Uncertain_Actual <-- Teacher_Actual | -0.14 | -0.53** |
Dissatisfied_Actual <-- Teacher_Actual | -0.15 | -0.55** |
Admonishing_Actual <-- Teacher_Actual | 0.03 | -0.02 |
Strict_Actual <-- Teacher_Actual | -0.19 | -0.60** |
Leadership_Student <-- Student_Actual | 0.82 | 0.52** |
Helping_Student <-- Student_Actual | 1.00 | 0.87^{#} |
Understanding_Student <-- Student_Actual | 0.74 | 0.80** |
Student_Resp_Student <-- Student_Actual | 0.00 | 0.00 |
Uncertain_Student <-- Student_Actual | -0.94 | -0.54** |
Dissatisfied_Student <-- Student_Actual | -0.84 | -0.65** |
Admonishing_Student <-- Student_Actual | -0.03 | -0.02 |
Strict_Student <-- Student_Actual | -0.92 | -0.70** |
# used for identification, **p < .01 |
Table 6 contains the values of the fit indices for this model. As with the previous model, all of them are outside the acceptable limits, but most are just outside so this would indicate that the model is a reasonable if not a good fit.
Coefficient | Value |
chi^{2} | 301.45 |
df | 104 |
chi^{2}/df | 2.899 |
GFI | 0.843 |
AGFI | 0.782 |
RMR | 0.115 |
RMSEA | 0.087 |
The fit indices indicate that both models are reasonable if not good fits to the data. One observation that must be made is that in both models the regression coefficients for the observed variables Student Responsibility/Freedom (Submission-Cooperation) and Admonishing (Opposition-Dominance) on the corresponding latent variable (Teacher_Ideal, Teacher_Actual, Student_Actual) are not significant. In fact, the regression coefficient is either zero or close to it in most cases. This would suggest that these observed variables are not factors that contribute to the classroom interaction. Of the other observed variables, the coefficients for Leadership, Helping/Friendly, Understanding contribute positively and Uncertain, Dissatisfied, Strict contribute negatively. These results are in line with expectations so would tend to confirm these scales as contributing to the measurement of classroom interaction.
The significance of this study is that it is one of the first that proposes a multi-level model of classroom interaction. It demonstrates that the way a teacher perceives an ideal classroom interaction should influences the teacher's perception of the actual classroom interaction. Also, the teacher's perception of the actual classroom interaction influences how the students perceive the classroom interaction, and vice versa. It is this that highlights the potential usefulness of the QTI as a tool for self-reflection and a catalyst for change in individual classrooms (Rickards & Fisher, 2000) that can be driven by the teacher for the teacher and their students. If teachers modify the way in which they interact with students in response to students' perceptions of the classroom interaction then this should have a positive effect on those perceptions, which in turn should affect the teacher's perception of the classroom interaction.
Fisher, D., Fraser, B., & Wubbels, T. (1993). Interpersonal teacher behavior and school environment. In T. Wubbels & J. Levy (Eds.), Do you know what you look like? Interpersonal relationships in education (pp. 103-112). London, England: Falmer Press.
Fisher, D., Fraser, B., Wubbels, T., & Brekelmans, M. (1993, April). Associations between school environment and teacher interpersonal behavior in the classroom. Paper presented at the annual meeting of the American Educational Research Association, Atlanta, GA.
Fisher, D., Henderson, D., & Fraser, B. (1995). Interpersonal behaviour in senior high school biology classes. Research in Science Education, 25 (2), 125-133.
Fisher, D., Rickards, T., Goh, S., & Wong, A., (1997). Perceptions of interpersonal teacher behaviour in secondary science classrooms in Singapore and Australia. Journal of Applied Research in Education, 1(2), 2-11.
Fraser, B. J. (1994). Research on classroom and school climate. In D. Gabel (Ed.), Handbook of research on science teaching and learning, (pp. 493-541). New York: Macmillan.
Fraser, B. J. (1998). Science learning environments: Assessments effects and determinants. International Handbook of Science Education. B. J. Fraser, & Tobin, K. Dordrecht, The Netherlands, Kluwer Academic Publishers. 1: 527-564.
Fraser, B., & Walberg, H. (Eds) (1991). Educational environments: Evaluation, antecedents and consequences. Oxford: Pergamon Press.
Jöreskog, K.G. (1993). Testing structural equation models. In K.A. Bollen and J.S. Long (Eds), Testing structural equating models (pp. 294-316). Newbury Park, CA: Sage.
Leary, T. (1957). An interpersonal diagnosis of personality. New York: Ronald Press.
Levy, J., Créton, H., & Wubbels, T. (1993). Perceptions of interpersonal teacher behaviour. In T. Wubbels, & J. Levy, (Eds.), Do you know what you look like? Interpersonal relationships in education (pp. 29-45). London: Falmer Press.
Levy, J., Wubbels, T., Brekelmans, M., & Morganfield, B. (1997). Language and cultural factors in students' percepti ons of teacher communication style. International Journal of Intercultural Relations, 21, 29-56.
Marcoulides, G. A., & Hershberger, S. L. (1997). Multivariate statistical methods: A first course. Mahwah, NJ: Lawrence Erlbaum Associates.
McRobbie, C., & Fraser, B. (1993). Associations between student outcomes and psychosocial science environment. Journal of Educational Research, 87, 75-85.
Newby, M., & Fisher, D.L. (2000). A model of the relationship between computer laboratory environment and student outcomes in university courses. Learning Environments Research, 3(1), 51-66.
Riah, H., Fraser. B., & Rickards, T., (1997, June) Interpersonal teacher behaviour in chemistry classes in Brunei Darussalam's secondary schools. Paper presented at the International Seminar on Innovations in Science and Mathematics Curricula, Universiti Brunei Darussalam, Brunei.
Rickards, T., Riah. H., & Fisher, D., (1997, June) A comparative study of teacher-student interpersonal behaviour: Looking into secondary science classrooms in Brunei and Australia. Paper presented at the International Seminar on Innovations in Science and Mathematics Curricula, Universiti Brunei Darussalam, Brunei.
Rickards, A. W. J. (1998). The relationship of teacher-student interpersonal behaviour with student sex, cultural background and student outcomes. Science and Mathematics Education Centre. Perth, Curtin University: 207.
Rickards, T., & Fisher, D. (2000). Three perspectives on perceptions of teacher-student interaction: A seed for change in science teaching. The annual meeting of the National Association for Research in Science Teaching, New Orleans, Louisiana.
Rickards, T., (1999, August) Student perceptions of interpersonal behaviours: An on-line education perspective. Paper presented at annual meeting of the Queensland Secondary Information Technology Educators state conference, Brisbane, Australia.
Rickards, T., & Fisher, D. L. (1997, July). A report of research into student attitude and teacher student interpersonal behaviour in a large sample of Australian secondary mathematics classrooms. Paper presented at the annual meeting of the Mathematics Education Research Group of Australia, Rotorua, New Zealand.
Rickards, T., & Fisher, D. L. (2000, April). A comparison of teacher and student perceptions of classroom interactions: A catalyst for change. Paper presented at the American Educational Research Association Annual Meeting, New Orleans, LA.
Schumacker, R.E., & Lomax, R.G. (1996). A beginner's guide to structural equation modeling. Mahwah, NJ: Lawrence Erlbaum Associates.
Taylor, P., Fraser, B., & Fisher, D. (1997). Monitoring constructivist classroom learning environments. International Journal of Educational Research, 27 (4), 293-302.
Teh, G., & Fraser, B. (1995). Development and validation of an instrument for assessing the psychosocial environment of computer-assisted learning classrooms. Journal of Educational Computing Research, 12, 177-193.
Watzlawick, P., Beavin, J., & Jackson, D. (1967). The pragmatics of human communication. New York: Norton.
Wubbels, T. (1993). Teacher-student relationships in science and mathematics classes (What research says to the science and mathematics teacher, No. 11). Perth: National Key Centre for School Science and Mathematics, Curtin University of Technology.
Wubbels, T., & Levy, J. (1991). A comparison of interpersonal behaviour of Dutch and American teachers. International Journal of Intercultural Relations, 15, 1-18.
Wubbels, T., & Levy, J. (Eds) (1993). Do you know what you look like? Interpersonal relationships in education. London, England: Falmer Press.
Wubbels, T., Brekelmans, M., & Hooymayers, H. (1991). Interpersonal teacher behavior in the classroom. In B. Fraser & H. Walberg (Eds.), Educational environments: Evaluation, antecedents and consequences, (pp. 141-160). Oxford: Pergamon Press.
Wubbels, T., Créton, H., & Holvast, A. (1988). Undesirable classroom situations. Interchange, 19, 25-40.
Wubbels, T., Créton, H., & Hooymayers, H. (1985, April). Discipline problems of beginning teachers. Paper presented at annual meeting of American Educational Research Association, Chicago, IL.
Wubbels, T., Créton, H., Levy, J., & Hooymayers, H. (1993). The model for interpersonal teacher behaviour. In T. Wubbels & J. Levy (Eds), Do You Know What You Look Like? Interpersonal Relationships in Education (1st ed., pp. 13-28). London: The Falmer Press.
Please cite as: Rickards, T., Newby, M. and Fisher, D. (2001). Teacher and student perceptions of classroom interactions: A multi-level model. Proceedings Western Australian Institute for Educational Research Forum 2000. http://www.waier.org.au/forums/2001/rickards1.html |