This study is intended to investigate the effect of school environment on the performance of teachers and students. In the literature review, the conceptualization of school environment is based upon the multi-perspective approach of Sergiovanni (1984), Bush (1986), Bolman and Deal (1991), and Cheng (1993b), that school environment is separated into five aspects: instructional, structural, human resource, political, and cultural. The leadership behavior of the principal of the school, conceptualized into five dimensions, is considered to be an important component of the school environment. Teacher efficacy and time-use, and students' learning efficacy and attainment test scores are assumed as dependent variables to assess their relationship to school environment.6.2 Implications
In a large scale survey of 61 schools, a total of 1407 teachers and 4715 students responded to some parts of the survey questionnaire. Despite these enthrusiastic response rate, only 55 schools, with an average of 7.4 teachers and 36.0 students per school had provided criterion related information useful for multi-level analysis. Based on the analysis of the data, the following major findings can be concluded.
First, for hypothesis testing of H1, no component variables of school environment is found to contribute significantly to the performance of teachers. The most important reason for the lack of effect is due to the fact that the between-school variations of the teacher performance variables are too small to detect any variations in the school environment.
Second, for hypothesis testing of H2, some component variables of school environment do contribute significantly to the performance of students. Three factors contribute significantly to the academic performance of the students and they are caring and support factor (a combination of three environment variables: esprit, intimacy, and student-centeredness), the formalization factor, and the pupil control factor (pupil control ideology minus organizational ideology).
Third, for hypothesis testing of H3, some of the component variables of teacher performance are correlated with some of the component variables of student performance at the school level. Teacher's time-use on discipline is found to correlate negatively with the three attainment test scores of the students.
Fourth, for hypothesis testing of H4, the contextual variables of the school do contribute significantly to the school environment, with the most important contribution coming from the age, size of the school, and the average teaching experience (also the average age of the teachers). The age of the school is correlated with the student centeredness variable. Hierarchy of authority and hindrance, which are signs of bureaucratization of a school, are positively correlated with the size of the school. The average teaching experience is positively related to participative decision and negatively correlated with hierarchy of authority.
Fifth, in past studies, leadership has been found to be an important factor in the maintenance of a cohesive social environment for the teachers to work in (Cheng, 1993b). However, in the present study, the strength of leadership factor (an integration of the five leadership dimensions: educational, structural, human resource, political, and symbolic) is not found to contribute significantly to the performance of teachers and students. One possible reason is because the sponsoring body of the sampled schools often reassigns a principal to different schools after the principal has served in one school for a few years. For this reason, a principal seldom has the opportunity to serve in a school for a long time and to allow leadership effects to take root. If this is the reason for the lack of leadership effect, then this study has obviously overlooked the issue of leadership stability--aminimum period of time required for leadership effects to take root.
Sixth, competition is highly correlated with affiliation and involvement, which shows that within the sample schools being surveyed, a competitive study environment can enhance better social relationship among the students and increase the students' engagement in study.
Eighth, among the criterion variables used for student performance, English attaimnent test score has the largest between-school variation (13.9%), which shows that the English attainment test score is more school dependent than the other student performance variables. This also shows the importance of the internal process of a school in the teaching and learning of a foreign language.
Ninth, interactions between environment factors and the personal characteristics of students do exhibit significant effects on the academic performance of students. This has supported Lewin's notion of person-environment interaction (Lewin, 1943).
Finally, the multi-level model is found to be a powerful model in the study of school environrnent as related to performance of students. The model is useful in the detection of school level effects, the interaction of personal characteristics and school level effects, and the in-depth analysis of the parameter variances.
From the above findings, several implications may be derived with regard to theory development and further study, to methodology development, and to practical concerns. Also, based on the practical implications, suggestions and recommendations are offered to school administrators, the Education Department, and teacher training institutes. They are summarized as follows.
(Implications for theory development and further study)
First, the delineation of school environment into multiple aspects is helpful to the researchers to systematically understand, analyze, and manage the internal process of the school organization. Similar approach can also be applied to the study of other types of organizations.
Second, although the leadership of principal can also be conceptualized into five dimensions, yet these dimensions are highly correlated. As in Cheng (1993b), it may be appropriate to use the strength of leadership instead of separate components in the anlaysis.
Third, the results of insignificant effects of the leadership factor on school environment in this study suggests that the stability of leadership effect may be important in research.
Fourth, contrary to the claims of encisting literature that competition is harmful to the students' social and intellectual development (Slavin, 1987; Gutierrez & Slavin, 1992), the present finding reveals that a competitive study environment in the academically above average schools may even enhance better social relationships among the students and more involvement in their studies. This implies that competition can have positive or negative consequences, and there may be other factors involved, such as family situations and personal characteristics, etc.
Fifth, the teacher's use of time (on discipline) factor, more than the other teacher performance variables, has significant between-school variations, but they are not accounted for by the school environment factors. This suggests that there are important organizational factors involved which were not considered in the present study.
Sixth, it has been found that teacher efficacy is not significantly related to learning efficacy at the school level. This implies that there may be a large gap between teachers' belief and teachers' action, and more studies on teachers' thought process are called for.
Seventh, students' learning efficacy has been found to interact with school environment and the effect contribute significantly to their academic performance. More in-depth study is needed to reveal the mechanism of this person-environment interaction.
Eighth, a similar study is called for to look at the school environment in the private primary schools, primary schools of other sponsoring bodies, as well as the secondary schools in Hong Kong.
(Implications for methodology development)
First, the social environment of a school includes human behavior, perception and belief, and these may change with time. Therefore, school environment has an evolving nature, and if one wishes to conduct more in-depth studies on this topic, one may need to conduct longitudinal research which may involve qualitative techniques.
Second, the present study has shown how a two-level model can be applied to study the effect of school environment on the performance of teachers and students. One can actually add one more level (classroom level) to the existing model. In this manner, one can understand how the school level effects influence the classroom process and how the latter influences the students.
Third, in order to make good prediction with the multi-level model, one will need to have survey instruments which are more sensitive to the variations in school environment as outcome variables.
Fourth, the application of the multi-level model in the study of school environment suggests that this method may be applied to study other school and social phenomena which involve the grouping of people. These may include research topics which involve physical grouping, such as school effects, classroom effects, voting behavior of people in various districts, etc., or nonphysical grouping, such as evaluation of instructional methods, student tracking, social mobility, etc.
Fifth, the multi-level model has been found to be a powerful tool in tne study of school effects. However, it is not a perfect tool. One of the reasons is because the statistical package is still at its early stage of development, many of the standard procedures in multiple regression, such as stepwise procedure, significance testing, etc., are unavailable in the present package. Another reason is because the model is basically a multi-level regression equation, which still has many of the limitations of the regression equation. In order to increase its versatility and applicability, one may combine this method with other multivariate statistical tools, such as MANOVA, path analysis, etc., to overcome some of its inadequacies in multi-level research.
(Implications for practical concerns and recomrnendations to practitioners, teacher training institutions, and Education Department)
First, it is fruitful to conceptualize social environment of a school from a multiple perspective, including the instructional, structural, human resource, political, and cultural aspects, and to study its relations with the performance of teachers and students. An implication to the policy makers is that they should not be mainly concerned about the inputs of schooling (finance, curriculum, and student allocation), but should also pay attentions about the internal process of the school, such as instructional approaches, school structure, etc. An implication to school administrators is that the school environment may be related to the contextual factors of the schools, such as age, size, experience of the teachers, etc., which is not under the control of the school administrators, but they have the authority to plan and implement suitable policies which can reduce the negative effects of the school contextual factors. Anothef implication is for the school administrators to keep in mind that although the school environment is complicated, the different aspects of school environment are inter-related, and they need to take a holistic view about school environment, and that school management should not be conducted in a piecemeal fashion. Second, since the principal is in a key role in the creation of a good school environment, which would be beneficial to the learning of the students and the success of the teachers, it implies that selection and training of principal should be given prominant attention by the policy makers and Education Department. The stability of leadership effect implies that in order for the leadership effects of principals to take root, the incumbents should not changed too often. Also, because a principal has multiple roles, he/she is an educational leader, a structural leader, a human relationship leader, a political leader, and a symbolic leader, it suggests that training programs for the principal should include the following five components:
Third, a caring and supportive climate is conducive to the learning of English language. The caring and supportive climate is also an important factor contributing to teachers helping students who are weak in learning efficacy to do well in Chinese. This implies that in the training program, emphasis should be placed on the development of a proper professional attitude--esprit, collegiality, and a genuine concern for both the students and the colleagues. This also implies that the school administrators should pay particular atention to the building of a professional, less bureaucratic, and humanized work environrnent which will nurture a caring and supportive climate arnong the teachers.
- sound and coherrent educational prinicples and philosophy;
- management training in planning, organizing, implementing, and evluaiing, which are important functions of the school structure;
- human relations training, such as motivational techniques, team techniques, etc.
- training in conflict resolution and the understanding of the internal politic of the school;
- trainings in educational philosophy and the methods to convey this philosophy to the other members of the school so that it can become part of the school culture.
Fourth, a school structure which places too much emphasize on writen policies and established procedures is counterproductive to the learning of Mathematics. Moreover, an overly formalized school structure produces more harmful effect on the academic performance of the male students than that of the female students. The same type of formalized school structure produces more deleterious effects on the Mathematics performance of students who have low learning efficacy than students who have high learning efficacy. These imply that a formalized school structure is an important organizational factor which will influence the behavior of its members. Formalization in the school process will facilitate a smooth operation according to the established goals of the school, and that communications may not be easily misunderstood; but when this is carried to an extreme, people may be required to abide by the school rules just for the sake of the writen policies. Therefore, school administrators should consider school structures which will fulfill the needs of the organization and the needs of the members within it.
Fifth, a cohesive and student-centered philosophy of teaching for the entire school is important for students to do well in school. Teachers in such schools tend to have a shared vision in education and are more concerned about the needs of the students than the need to the behavior of the students inside the classroom. Teacher's use of time on discipline also has important consequence to the academic performance of the students. These imply that in the training program, the teachers should be given a sound and coherrent instructional philosophy which puts the students at the center stage of the educational program, and should be given training in classroom management techniques, so that time will not be wasted on discipline. Also, the school adrninistrators should develop, within the schools, a sound and coherrent educational philosophy which puts students and learning at the center of the school program, and should give more classroom supervision to the teachers so that a proper balance in the use of time in teaching and on discipline can be attained. Furthermore, the Education Department should put more resource in the promotion of whole school approach in guidance, so that the teachers will progress away from a discipline-oriented instructional approach.
Sixth, for the schools with above average academic performance, a proper level of spirit of competition is related to a positive social climate among the students. This implies that in the training program, the teachers should be enabled to differentiate between positive competition and negative competition, that the former is associated with a positive social climate, and the latter is damaging to the morale of the students. The school administrators should also see to it that the spirit of competition is not too intense to such a point that is damaging to the morale of the weaker ones.
In conclusion, the present study is helpful to the understanding of the internal social environment of the school organization and its relations to the performance of teachers and students. Using the multi-level statistical model, the social environment of 55 aided primary schools were investigated in relation to the performance of 1407 teachers and 4715 students within the schools. Due to some limitations in the performance indicators and instruments, the relationship between school environment and performance of teachers was not discerned from the results, but the findings of the relationship between school environment and performance of students did have theoretical and practical implications in school management. Based on the findings and implications, suggestions were then provided to the policy makers, school administrators, and teacher training institutes. This study is a comparatively new attempt in organizational studies and the results have been moderately satisfactory.