Back to: School ImprovementDimensions and Determinants of School
Social Climate in Schools Enrolling Middle Years Students
By James Ian MacIntosh
SSTA Research Centre Report #91-04: 62 pages, $14.
| The purpose of this study was to apply a conceptual framework
developed from the literature in an investigation of school social climate in schools
enrolling Middle Years students (Grades 7, 8, and 9). The framework encompassed 15
variables or determinants in four clusters: physical features, organizational factors,
aggregate teacher characteristics, and aggregate Student characteristics. The determinants
were derived from school effectiveness research. Students, teachers, and principals from four types of schools completed questionnaire. The quantitative data were analyzed using correlation analysis, analysis of variance, and path analysis. |
|
| Summary of Findings |
Back to: School Improvement
The SSTA Research Centre grants permission to reproduce up to three
copies of each report for personal use. Each copy must acknowledge the author and the SSTA
Research Centre as the source. A complete and authorized copy of each report is available
from the SSTA Research Centre.
The opinions and recommendations expressed in this report are those of the author and may
not be in agreement with SSTA officers or trustees, but are offered as being worthy of
consideration by those responsible for making decisions.
The purpose of this study was to apply a conceptual framework developed from the literature in an investigation of school social climate in schools enrolling Middle Years students (Grades 7, 8, and 9). The framework encompassed 15 variables or determinants in four clusters: physical features, organizational factors, aggregate teacher characteristics, and aggregate Student characteristics. The determinants were derived from school effectiveness research.
Students, teachers, and principals from four types of schools completed questionnaire. The quantitative data were analyzed using correlation analysis, analysis of variance, and path analysis.
Four determinants exhibited consistently positive effects on student-perceived school climate dimensions, while two determinants exerted Similar negative patterns. Three determinants exhibited consistently positive relationships with teacher-perceived school climate, while one determinant exhibited a negative pattern. Several determinants demonstrated inconsistency in direction of effect at both the student and teacher levels.
Variations in the strength of relationships and dissimilar patterns of influence suggest that teachers and students held disparate views regarding school social climate.
The path model of determinants of social climate provided a means to test the hypothesized network of relationships among variables
Both School effectiveness research and the earlier quality of school life studies, while different in emphasis and concern, have looked to the concept of social climate for explanation of the student outcome-environment relationship.
According to research findings, effective schools appear to have a critical mass of positive characteristics which, when found together, account for the differences in levels of student attainment. This combined effect is more powerful than any individual factor considered on its own, and these school-based variables were subject to manipulation. The resulting institutional phenomenon was referred to as "ethos" (Rutter et al., 1979) or "school climate" (Broakover et al., 1978), and this notion stimulated renewed interest in the psycho-sociological dimensions of schools among a growing number of researchers concerned with the means to create effective schools (Squires, Huitt, and Segars, 1983; Bickel, 1983; Purkey and Smith, 1985).
Previous to the interest in school social climate emanating from the school effectiveness literature, study of the phenomenon was sustained by a small cadre of theorists and researchers exploring the notion of quality of school life as an outcome of the process of schooling. To these investigators, school satisfaction for students was analogous to job satisfaction for adults and environmental assessment to determine the school setting impact on student attitudes and behavior was important in its own right. General satisfaction, commitment to task, and relation to authority figures were seen as important to children in school, as they were to adults at work, for daily mental health. Furthermore, positive
reactions to school may increase the likelihood that students will stay in school longer, develop a lasting commitment to learning, and use the institution to their advantage. This notion was supported by researchers who contended that, on the basis of their research, positive social environments and positive learning outcomes appear to go hand in hand.
The convergence of interest in school social climate by quality of life and school effectiveness investigators has given strong impetus to the most recent strand of school climate research. However, the means by which positive school climates develop, or how they may be implemented, remains unclear.
One of the ma]oz justifications of this study, is that it presents a framework for the examination of the determinants and dimensions of school Social climate and uses it in an investigation of school social climate in types of school enrolling Middle Years students. By attempting to map the determinants of school social climate and to identify normative patterns for types of schools, this study sought to broaden understanding of the school as a social environment.
Research in the area of school social climate has demonstrated that many variables associated with positive student outcomes are manipulable. This information can provide the basis for planning and change whether it is at the classroom or school level. Identifying normative school social climate patterns for schools with different grade-level patterns extends the utility of prescription for those practitioners who deal with students at all levels in the school organization.
A recurring theme in the literature has emphasized the importance of meeting the needs of Middle Years children in the school setting
Knowledge of the Middle Years student responses to the school social environment as revealed through their perceptions, combined with staff perceptions, may provide some practical insight for educational decision-makers. Purpose and Statement of Research Questions The purpose of this study was twofold. It was designed first to select and refine a framework for the investigation of school social climate and second to apply that framework in a study of social climate in different types of schools enrolling Middle Years students [Grades 7, 8, and 9) in the Province of Saskatchewan.
More specifically, the study was guided by the following research questions which were developed from the conceptual framework.
1. What relationships existed between the determinants described in the framework and school social climate as perceived by students?
2. What relationships existed between the determinants described in the framework and school social climate as perceived by teachers?
3. How similar were the patterns of relationships between the determinants described in the framework for student-perceived and teacher-perceived school social climate
4. Did student perceptions of School social climate differ significantly according to type of school enrolling Middle Years students?
5. Did teacher perceptions of school social climate differ significantly according to type of school enrolling Middle Years students?
In an effort to describe and to analyze viewpoints about school social climate in a systematic manner, a review of the relevant literature and research was conducted in which approaches were classified into three broad categories: the personal dimensions perspective, the environmental dimensions perspective, and the interaction perspective. The theoretical notions and research associated with the approaches found within each perspective were discussed and examined according to criteria developed for that purpose. This critical analysis of the perspectives provided a basis for concluding that the interaction perspective, as conceptualized in the tripartite model (Moos, 1973), provided the moat comprehensive rationale for the description of social climate within schools at the classroom level and, possibly, the school level.
Support was found in the literature for both the feasibility and the utility of differentiating social climate at the school level from that of the classroom setting Further examination revealed that the school climate factors identified by Rutter et al. (1979) and Johnston (1984) could be reconceptualized in terms of Mooe' (1973) tripartite model as potential dimensions or subscales within the three domains.
Attention was then given to an existing school social climate construct that was conceptualized as having three domains and thirteen variables or dimensions (Massachusetts Study 1977). In order to examine its rational validity and to monitor its heuristic worth, the construct was compared to criteria established using school social climate factor from the school effectiveness studies. This examination appeared to provide support and evidence that the school social climate construct, developed for the Massachusetts Study (1977), was congruent with the theoretical basis of the tripartite model (Moos, 1973) and had potential heuristic worth for the description and assessment of the school social climate phenomenon. While description of school social climate in itself is valuable, many researchers emphasized that the need existed to identify the determinants of the phenomenon and the relationship among those factors. Attention was then turned to the examination of the determinants of school social climate.
Moos (1979) proposed a unifying conceptual framework to rationalize the determinants of classroom social climate. This model was based on the interrelationship of five sets of classroom characteristics (overall context, physical and architectural features, organizational factors, teacher characteristics, and aggregate student characteristics). Since Moos (1979) applied this model to other social environments (e.g., university living groups) before testing it at the classroom level, it was reasonable to assume that the model had some generalizable applicability to various social environments. With this reasoning in mind, each of the five acts of determinants was examined and the model was modified for the purpose of this study for use at the school level. While Moos' (1979) model of the determinant of classroom social climate provided the guideline for theory development, the potential determinants were gleaned for the most part from the school effects literature.
This review of literature and research led to the selection and refinement of a framework for the investigation of school social climate. The review concluded with the statement of research questions and the
The model of the determinants of school social climate, as modified for this Study, is shown in Figure 1. This model specified both the determinants of School social climate and the social climate dimensions involved. Operational Definition of Terms Concepts central to this study are operationally defined in this section. The independent and dependent variables are described respectively. The intervening variables (the determinant clusters) are presented in the order in which they appear in the model to he used in this study.
Type of School
The nominal contextual variable, Type of School, was defined on the basis of grade-level organizational pattern. Grade-level organizational pattern specified the range of school years (or grades) found within one administrative unit and usually located within a single school facility. Because of the close relationship of school grade to pupil age, grade-level organizational pattern usually delimited the age group or groups that formed the student composition of the school.
For the purpose of this study, only schools whose grade-level organizational pattern encompassed grades seven to nine were of interest. The operational definition for Type of School was delimited to K to 12, K to 9, grade 7 to 9, and grade 7 to 12 schools.
School Social Climate
The dependent variable, school Social climate, was conceptually defined as feelings and opinions about various aspects of the school and how it operates, as perceived by students, teachers, and administrators. For the purposes of this study, school social climate was operationally defined in terms of three domains and thirteen dimensions or subscales each of which was measured by seven items. The three domains were considered as global areas of school social climate and the dimensions or subscales assigned to each domain were viewed as more specific areas of school social climate. The domains, dimensions, and items within each dimension are found in Appendix A.
Organization Factors
This determinant consisted of a set of variables that were influenced to a large degree by rational planning and decision-making at the school building level. Three factors, School Policies and Practices, Organization for Instruction, and Class Size, operationally defined this determinant. The principal in each school was asked to provide the necessary information regarding these factor.
School Policies and Practices
This variable concerned the nature and existence of school policy in areas that were identified in the school effects literature as potentially related to school social climate. The policy statements of interest were in the areas of teacher supervision, student evaluation, decision making, extra-curricular activities, student recognition end rewards, goal development, and mission statements.
Organization for Instruction
This factor was related to School practices for instructional delivery in three areas. Schools were differentiated on the basis of emphasis placed on subject specialization, use of streaming, and pull out resource rooms or segregated special education classes.
Class size
Class size was defined as the average number of students per classroom at the Middle Years (grades 7, 8, and 9) level.
Physical and Architectural Features
This determinant referred to a set of ecological factors which influenced school social climate through their level of attractiveness to the participants in that social environment. The determinant, physical and architectural features, was operationally defined in terms of a cluster of three variables: School Appearance, Display of Student Work, and Student Conditions. The descriptive information concerning these variables was supplied by the principal of each school.
School Appearance
School appearance pertained to aspects respecting the cleanliness, maintenance, lighting, state of decor, and absence of graffiti within the school.
Display of Student Work
Display of Student Work was operationalized through an assessment of the amount of student work given public recognition in the school.
Student Conditions This concerned the emphasis given to student access to physical areas of the school, equipment, recreational activities, and conveniences. It was operationally defined by a set of six questions requiring an assessment of specific conditions.
Aggregate Teacher Characteristics
This determinant consisted of a cluster of factors that were descriptive of the teacher cohort involved with offering the Middle Years (grades 7, 8, and 9) program in each school. Five variables, Teacher Mix, Teacher Stability, Level of Training, Teacher Gender Ratio, and Teacher Grade Preference, comprised the set of factors that operationally defined the determinant aggregate teacher characteristics.
Teacher Mix
This variable referred to the proportion of the staff with five or fewer years of teaching experience. It was calculated from data collected on the principal questionnaire.
Teacher Stability
Teacher stability was viewed as a function of teacher experience in the same school setting. It was operationalized by comparing the number of teachers with more than five years experience in their present school to the total staff. This measurement was shown as a proportion score for each school staff. It was calculated from data collected on the principal questionnaire.
Teacher Gender Ratio
The number of female teachers on staff was expressed as a proportion of the total staff. The information needed to make the above calculation was obtained on the principal questionnaire.
Level of Training
Level of training was defined by the professional certification held by the teachers assigned to the Middle Years level. Operationally, teachers were classified numerically on the basis of a non-degree, degree, and degree plus status. An aggregate average for each school was calculated from the information received on the teacher questionnaire.
Teacher Grade Preference
This variable referred to the level of satisfaction experienced by teachers who were assigned to teach at the Middle Years level. It was measured with a set of three questions on the teacher questionnaire and it was reported as a school average
Aggregate Student Characteristics
This determinant referred to a cluster of variables which characterized the Middle Years (grades 7, 8, and 9) student cohort in the school. The four factors selected to operationally define aggregate student characteristics were Student Mix Ratio, Student Gender Ratio, Student Satisfaction, and Student Sense of Academic Support.
Student Mix Ratio
Student Mix Ratio indicated the proportion of the total student body formed by the Middle Years (grades 7, 8, and 9) student cohort. It was calculated from data provided by the principal.
Student Gender Ratio
Student Gender Ratio indicated the proportion of the Middle Years (grades 7, 8, and 9) student cohort that was female. It was calculated from information provided by the principal.
Student Satisfaction
Student Satisfaction was defined in terms of an aggregate student response to a question about level of satisfaction with school in general. It was reported as a school average.
Student Sense of Academic Support
This construct referred to aggregate student feelings about functioning adequately in the school social environment. It was operationalized through the incorporation of a modified version of the 12 item scale developed by Brookover et al. (1977) into the student questionnaire used in this study.
Validation of Instruments
The conceptual underpinnings of the student social climate instrument were supported in a qualitative study using grade 7 and 8 Students. The questionnaire was modified for use in a Saskatchewan context and its psychometric properties, along with those of Student Sense of Academic Support and Student Satisfaction, were tested in a pilot study. At the same time, a modified version of the teacher social climate instrument originally developed for the Massachusetts Study (1977) was piloted. The School Questionnaire for Principals, which collected information on the determinants, was also assessed in a pilot Study at a later date.
Sample
The sample for this study was selected from four types of schools enrolling grade 7, 8, and 9 students (K - 9, K - 12, 7 - 9, and 7 - 12 schools). Invitations to participate were sent to principals whose schools met the criteria related to grade-level organization and size. In total, 88 schools participated in the study.
Organization and Collection of Data
Each principal, who agreed to participate, received student questionnaires, teacher questionnaires for the Middle Years staff (assigned to teach at the Grade 7, 8, or 9 level), and a principal questionnaire. Instructions were included for the selection of a random sample of students and for the administration of the questionnaires. As well, procedures to help guarantee the anonymity of respondents were part of the instructional format.
The student questionnaire was designed to measure school social climate at the student level and two aggregate student characteristics, Student Sense of academic Support and Student Satisfaction. The teacher questionnaire measured school social climate from the teacher perspective and two aggregate teacher characteristics, Teacher Grade Preference and Level of Training. The principal questionnaire gathered information pertaining to the remaining determinants.
quality of Data Following the collection of the study data, the psychometric properties of the social climate instruments were examined with respect to item and subscale characteristics and reliability estimates. The ability of the social climate dimensions to discriminate among schools at the student and teacher level was tested The various analyses of the social climate instruments Suggested that the data were of good quality and could be reliably interpreted.
Analysis of Data
The descriptive statistics associated with the determinants were calculated. Path analysis and analysis of variance (ANOVA) were selected as the major analytical procedures. Path analysis required that theory predict the relationship among the variables before statistical analysis.
This section focuses on school social climate as perceived by Middle Years students in the Province of Saskatchewan. It begins by examining the literature and research for evidence that social climate at the school level can be differentiated from the classroom level and that such a differentiation is not only valid, but useful and definitive. The development of a social climate intrument for the Saskatchewan context is then described. The section concludes with a discussion about the provincial profile for school social climate as perceived by Middle Years Students.
The theoretical underpinnings of 'Moos' (1973) tripartitie model presented a useful framework for the examination of social climate in a variety of contexts including classroom social climate. This framework held that three domains were salient for any type of social environment. Furthermore, not only can various types of environments he compared and contrasted with one another, it is possible to describe unique climate profiles for every social environment within each classification. The tripartite model focused on the nature and intensity of personal relationships within the social environment (relationship domain); the potential within the environment for personal growth and the development of esteem (personal growth or goals orientation domain); and the extent to which the environment is orderly and clear in its expectations, maintains control, and is responsive to change (system maintenance and change domain). The variables or dimensions that comprise each domain will vary from one type of social environment to another. Considerable fuzziness exists in the literature respecting school social climate, as it was often confused or confounded with classroom social climate. For example, Johnaon and Johnson (1970) considered school climate to be nothing more than the sum total of all the classroom climates found within the school, while other writers (Tunney, 1980; Sagor, 1981) used the terms interchangeably. AS well, some researchers ( Ellett and Walberg, 1979; Marjoribanks, 1980) have adopted approaches that measure school environments with a classroom social climate instrument in which the word class is simply replaced by school, thereby obfuscating or ignoring the distinctive features of both social environments. Furthermore, Chavez Chavez (1984) concluded that school social climate research per se was problematic because a variety of classroom climates, positive or negative may exist within the same school. In their discussion of classroom and school social climate, Rentoul and Praaer (1982) raise the issue of whose perceptions were appropriate. They differentiated between classroom level and school level research in social climate by advancing the argument that while students' perceptions were commonly used in measuring classroom environments, they were seldom used in measuring school climate. For this reason, Rentoul and Praser (1982) used teachers' perceptions to gain insight into school social climate.
However, support was found in the literature for focusing on social climate at the school level as a potentially important area of study. Several researchers noted the pervasive influence that school social climate had on teachers. Copeland (1979) found that teachers are socialized into the normative pattern of the school, thus making teachers more alike within schools, while Scott (1977) held that students play a Significant part in socializing teachers into the school culture. Boocock (1973) and Anglin (1979) argued that the teacher's role is imbedded in the culture of the school. Furthermore, Rutter et al. (1979) observed that it was easier to be an effective teacher in some schools than others and that some teaching practices that worked well in some schools did not work successfully in others. Mackenzie (1983) noted that it would be difficult, if not impossible, to provide effective instruction in a disorderly, disorganized, and disoriented school.
Stewart (1979) reasoned that the school environment could be considered as distinct from and more global than classroom environment, while Trickett (1978) regarded the classroom as a reflection of the broader school culture, but not synonymous with it. Based on his research evidence, Rutter et al. (1979) commented that differences between schools as social institutions were the primary determinants of differences in outcomes attained by students Examined collectively, various sources of evidence gave support to the conclusion reached by Anderson (1982) and Mackenzie (1983) that classes nested within schools have social climates that are directly or indirectly influenced by the wider school social climate This being the case, it appeared that school social climate as a construct had considerable heuristic worth.
Factors That Differentiate School Social Climate
The school effectiveness literature, particularly the work of Rutter et al. (1979) and Johnaton (1984), helped to clarify the nature of school social climate and to identify the factors which influence its development.
Rutter et al. (1979) observed that individual classes and the total school constituted different social groups with different skills involved in their management. Since the school environment is less controlled and directed, it is characterized by less direct formal organizational press and by peer group influences which are considerably less curtailed than in the classroom setting. With these points in mind, it is important to note that the effectiveness of any group norms in the school social environment depends implicitly on students perceiving membership in the group which holds the norm ( Garbarino and Asp, 1981; Boocock, 1973).
According to Rutter et al. (1979), the chief mechanisms for the establishment of normal and values in a secondary school setting are likely to related be teacher expectations about children and their work, the model presented by teachers and other students, and the feedback that students receive on what is acceptable performance at school. Furthermore, it is far more likely that students will share the goals of the school and be influenced by the norms and values to which they are exposed if certain factors, related to the general conditions for students and attitudes toward them, were operating positively in that social environment. Rutter et al. (1979) also noted that these important process variables do not result usually from the actions of teachers individually, but were matters of school policy and collective responses.
The conclusions reached by Rutter et al. (1979) regarding the factors which influence the development of school ethos or social climate received support in the work of Johnston (1984). While studying effective middle schools, Johnaton (19S4) identified factors which approximate the mechanisms and influences documented by Rutter et al. (1979). Based on his observations, Johnston (1984) hypothesized that school social climate could be subdivided into four clusters of climate variables (physical, academic, organizational, socio-emotional) which interact to create a culture or complex milieu which, in turn, influences every activity of the school.
The climate clusters identified by Johnston (1984) may be readily reclassified according to the domains of the tripartite model. The relationship domain (Moos, 1974) corresponded closely with Johnston's (1984) socio-emotional cluster, while the personal growth domain and the system maintenance domain of Moos' (1973) model resembled Johnston's (1984) academic and organizational clusters respectively. While the terminology used by Johnston (1984) and Moos (1973) did not correspond with that selected by Rutter et al (1979), a closer inspection of the descriptors showed striking similarities, and the mechanisms and influences identified by Rutter et al. (1979) can be rearranged to approximate the three domains. The relationship among these three approaches is further described in Table l.
As suggested in the review of the literature, support was found for using the tripartite model (Moos, 1973) as the basic theory in conceptualizing school social climate. In addition, when examined, the social climate construct developed for the Massachusetts Study (1977) and operationalized in the form of a student questionnaire appeared to be congruent with that theoretical framework. However, the decision was made to test the conceptual model and to examine the psychometric properties of the student questionnaire in the Saskatchewan setting. Therefore a confirmatory study was undertaken to examine the conceptual underpinnings of the tripartite model and a pilot study was conducted to test a modified version of the student questionnaire. (For explanation and description concerning the domains, dimensions, and items of the student questionnaire in its final form, the reader may turn to Appendix A).The purpose of the confirmatory study was to collect qualitative data about school social climate from the student perspective and to analyze these findings in terms of the tripartite model. For this reason, four grade 8 classrooms, two from rural and two from urban settings, were selected to provide the initial data of interest.
The students were then asked to participate in two activities: writing their thoughts about their school environment and then discussing their ideas in a classroom meeting that would be taped. students were asked to spend 10 to 15 minutes in the written activity. The subsequent classroom discussion was basically open ended with the investigator encouraging students both to initiate discussion and to respond to comments from the other students and from the investigator. Each of the group discussions lasted approximately one half hour
In the analysis stage, the written comments of the students were broken down into main ideas or themes. Each item was then examined for its compatibility with the broad taxonomy of the tripartite model (Moos, 1973). The vast majority of the comments and observations made by the students fell within the parameters set by the tripartite model. Written and spoken themes centered around Such topics as teacher support, student friendship, involvement, sense of community (relationship domain), purposes of schooling, teacher expectations, opportunities available (goal orientation or personal development domain), level of order, rules and fairness, influence of students in decision-making (system maintenance and change domain) Areas outside of the tripartite model included references to the physical facilities (or lack of them), Size of school, location of school, bussing, and attendance at other schools. From this analysis of qualitative data, it appeared logical to conclude that the tripartite model reasonably conceptualized school social climate in terms meaningful and salient to students. In addition, both urban and rural students tended to write and to talk about similar notions when asked about the school social environment.
Since the confirmatory study essentially supported the conceptual validity of tripartite model, as perceived by students, the decision was made to examine the social climate instrument developed for the Massachusetts Study (1977) in terms of its items and subscales and their applicability to the Saskatchewan context. In that instrument, 3 domain and 13 dimensions were used to represent fundamental aspects of school Social climate.
Since the student questionnaire used in the Massachusetts Study (1977) had been developed for administration in large high Schools, the reading level of the original instrument was assessed using three different methods. The Davidson Speed Reader, a computerized program, assigned grade 6 reading level to the student questionnaire. A second assessment, using Frye's Readability Indices, estimated that the student questionnaire had a grade 4 reading level. Finally, the contents of the instrument were discussed with grade 7 and 8 students who indicated no difficulties in understanding the items. Based on this information, the conclusion was reached that simplification of language was not a ma]or concern since the estimated reading level appeared to be within the capability of the vast majority of students in the grade levels of interest.
The student questionnaire was then examined for content unrepresentative of smaller schools or lover grade levels. Part of this procedure involved an initial administration of the questionnaire to a class of grade 7 and 8 students (n = 14), after which the clarity and relevance of the questionnaire were discussed with the students. In all, nine items were reworded to increase clarity and to make the questionnaire consistent with the Saskatchewan context. As well, the response format was changed from a four to five point scale, which included a "neither disagree nor agree" midpoint to eliminate the forced choice format. Finally, this process led to clarification in instructions respecting questionnaire administration.
Following this development and modification stage, a pilot study was conducted in which the student questionnaire was administered to 160 students on two occasions, eight weeks apart, to test its psychometric properties When taken together, the internal consistency estimates, the repeated measure results, and the test-retest correlation coefficients indicated that the student questionnaire exhibited an adequate general level of consistency as a measurement instrument. These conclusions were later supported by the reliability estimates found for the major study data.
Further examination of the psychometric properties of the student questionnaire supported its validity. All items correlated highest with the dimension to which they were assigned, and each dimension correlated highest with the domain to which it was assigned. This finding that two measurement techniques, the conceptual assignment of items to dimensions and dimensions to domains and the statistical treatment which indicated item/dimension/domain convergence, were highly related is evidence of convergent construct validity. It should be noted that all social climate dimensions are positively correlated with each other. These findings suggested that a student who rated his or her school favorably on one aspect of social climate tended to rate it high on another aspect. Similarly, schools rated high on one aspect were likely to be rated high on other aspects.
At the student level, intercorrelations ranged from r - .265 (Grouping and Influence Distribution) to r ~ .697 (Goal Direction and Learning Orientation) with the average intercorrelation among all 13 subscales being r - .504. Based on these intercorrelations, the proportion of variance shared by one social climate dimension with another ranged from under 9 per cent to approximately 48 per cent with an average variance of slightly over 25 per cent for all subscales. Since the subscales are considered to be different aspects of the same general conceptual notion, a degree of intercorrelation was both expected and desirable. Therefore, the finding that the 13 subscales levels appeared to measure distinct, although correlated, dimensions of the psycho-social environment further supported the validity of the social climate construct developed from the tripartite model.
It is therefore reasonable to assume that data gathered by using this questionnaire will provide a reasonably accurate and stable reflection of the school social climate dimensions as perceived by students. Furthermore, when those student perceptions are pooled and averaged on a school basis, a school profile emerges that is unique and discernible from other school environments
The standard deviation within a school can be regarded as a measure of the consistency of perceptions within that school In this regard, smaller standard deviations indicated greater congruence on the part of participants in their perceptions of the social climate dimensions. On the other hand, larger standard deviations were related to more diversity of opinion concerning perceived aspects of school climate. In addition, smaller standard deviations were usually found in conjunction with higher scores for the social climate dimensions, while larger standard deviations tended to be associated with lower subscale scores. From these observations, it appeared that congruity in shared perception at the student or teacher level was characteristic of Schools with more favorable ratings of social climate, while wider disagreement among participants about the quality of social climate was apparent in schools with less favorable ratings of social climate.
Variation Among Schools
The student instrument was examined to assess the degree to which it discriminated among schools. The 95 per cent confidence interval for the mean was calculated for the social climate dimensions for each school at both the student level. The confidence intervals were then inspected for schools which were discreet. from one another given this conservative estimate. Based on these observations, the conclusion was reached that each subscale at the student level discriminated among the schools.
The sample means, standard deviations, and the range for the student
social climate dimensions and the four types of schools (7 - 9, K - 9, 7 -12, K - 12.), using the school as the unit of analysis. The results of ANOVA revealed significant differences in the pattern of scores for three social climate dimensions, Accessibility and Receptivity, Equal Treatment, and Options. These data, along with data on the social climate dimensions which did not attain significance is reported in Table 3.
As the ANOVA analyses indicated, significant differences among types of schools were limited to three social climate dimensions. In this respect, the findings reflected that 7 - 9 schools were found to be different from K - 12 schools in Accessibility and Receptivity and Bqual Treatment and from K - 9 schools in Options. Also, in the case of Options, 7 - 12 schools differed from K - 12 and K - 9 schools, and a similar relationship existed between K - 12 and K - 9 schools. However, it appeared that for the majority of social climate dimensions, variation between schools in general was large enough to override differences between schools when classified by Type of School. In other words, failure to find significant differences for the remaining social climate dimensions indicated that aggregate student perceptions tended to be independent of Type of School and the variation within the four types of schools was greater than the variation among them.
Notwithstanding the results of the ANOVA, reasons existed for further examination of the social climate data when classified by Type of School First, it should be noted that, because of varying size of the samples and certain other characteristics of the data, the possibility existed that the level of significance actively in force was more conservative than the .050
chosen, thus making it inordinately difficult to find significance differences. Second, inspection of the descriptive data associated with the social climate dimensions served to clarify somewhat the relationship existing between Type of School and the social climate dimensions at the student level and to identify certain trends that could be considered interesting.
In this respect, it was noteworthy that 7 - 9 schools exhibited the highest mean rating for all social climate dimensions, with the exception of Groupings, Goal Direction, and Options. However, the difference between means, although consistent, appeared to be more incremental than substantial in magnitude for the majority of the social climate dimensions.
When the variation in scores is examined on the basis of Type of Schools, interesting comparisons and contrasts emerge. The K - 9 and 7 - 9 schools are strikingly similar in the restricted range each exhibited in terms of the social dimensions. When variation is expressed in terms of whole sample standard deviations, the average ranges for the social climate dimensions for the K - 9 and 7 - 9 schools are 2.06 and 2.24 respectively. The 7 - 12 schools have a similar pattern in variation for the dimensions of the Relationship Domain (2 54), but exhibit variation well in excess of 3.0 standard deviations for the remaining dimensions.
On the other hand, the greatest variation in scores (6.3 standard deviation) was found among the K - 12 schools in all instances. In addition, with the exception of Groupings, the range of scores for K - 12 schools encompassed both the highest and lowest scores recorded for the social climate dimensions at the student level. Figure 3 illustrates the parameters discussed for the dimensions of the relationship domain
The analysis of the determinants of school social climate an the basis of Type of School was beyond the scope of this study. However, the dramatic differences that existed among K - 12 schools are of particular interest and importance when one considered the number of schools which fit into this category. It is worth noting that the relationships among the determinants and their saliency for school social climate may be even more critical to K - 12 schools than the other categories that were examined.
In this regard, Organizational Factors, such as School Policies that pay careful attention to specific goals, appropriate programs, and positive conditions for Middle Years students as a group, may be key variables in the establishment of more positive school social climates. In addition, it was shown that Student Sense of Academic Support was the most influential determinant of School social climate for the total sample, and that Teacher Grade Preference (teacher satisfaction with a Middle Years assignment) influenced significantly both school social climate and Student Sense of Academic Support. Bearing in mind the above relationship, it perhaps should be a priority to match teacher and grade level at the Middle Years, particularly in K - 12 schools where there appears to be a greater potential for a mismatch. The broad focus of the K - 12 school with its many agendas makes a teacher selection process and assignment, that takes into account a preference for Middle Years students, extremely difficult to achieve without conscious administrative attention and commitment
1. The subscales, which represented the dimensions of school Social climate conceptualized and operationalized for this study, discriminated Significantly among schools at both the student and teacher levels.
2. Various reliability indicators revealed that the school social climate subscales provided stable and consistent measurement at both the student and teacher level.
3. Students who rated their school social climate more favorably tended to agree with one another, while wider disagreement among students about the quality of social climate was more apparent in schools with less favorable ratings of social climate
4. In general, the range of participant perception was greater at the teacher level than at the student level. This result suggests that wider disagreement about the quality of social climate was mare evident at the teacher level than at the student level.
5. Examination of the correlation matrix for the social climate dimensions at the teacher and student levels revealed that school social climate for teachers and students appeared to be rather distinct, although related, phenomena. This finding gave support to the notion found in the literature that the "world of the student- and the "world of the teacher" are basically different, despite the fact they occur in the same physical setting.
6. At the student level, significant differences among types of schools
were limited to three social climate dimensions, Accessibility and Receptivity, Equal Treatment and Options. However, it was noted that 7 "9 schools had the highest means for 10 of 13 social climate dimensions and K - 12 schools had the greatest variation in scores for all social climate dimensions.
7. With the exception of Options, the variation within the four types of schools was greater than the variation among them for teacher perceived school social climate. However, K - 9 schools recorded the highest mean for 11 of 13 social climate dimensions and K - 12 schools once again shoved the greatest variation in scores.
l. In general, the hypothesized relationships presented in the path model accounted for substantial amounts of the variance for all social climate dimensions (27 to 72 per cent) at both the student and teacher levels, although its influence was usually stronger at the student level. The path model represented those determinants conceptualized as having both direct and indirect effects on the social climate dimensions.
2. The determinants tended to exhibit a different pattern at the teacher level from what was observed at the student level. In this respect, it was clear that some of the principal determinants of social climate at the student level were substantially reduced in impact at the teacher level and, at times, were opposite in effect. On the other hand, certain determinants had greater importance at the teacher level than previously observed at the student level.
3. Student Sense of Academic Support emerged as the dominant determinant for student level social climate. In addition, this determinant had a strong influence on Student Satisfaction, which in turn had a consistently positive influence in general on the social climate dimensions at the student level. Therefore, Student Sense of Academic Support had an indirect effect on the social climate dimensions as well as a direct one. In other words, student feelings about functioning adequately within the school social environment played a major role in both the determination of school social climate and student satisfaction with school.
4. Organization for Instruction exhibited a positive influence on Student Sense of Academic Support. In other words, higher levels of subject specialization, pull-out resource rooms, and ability grouping were found in association with higher aggregate scores in Student Sense of Academic Support.
5. Class Size was found to have a positive impact on Student Satisfaction; that is, schools with larger average class sizes tended to have higher aggregate scares in student satisfaction with school in general. It should be noted that the average size (21.02 students) approximated the ideal class size described in the literature. In addition, the restricted range of class size may have caused a ceiling effect Given these circumstances, this finding should he treated with caution since it may not be representative of a larger and more heterogeneous population of schools. 6. Teacher Grade Preference was the strongest determinant of school social climate at the teacher level. However, this determinant also exhibited a consistently positive influence at the student level where it had both a direct effect itself and an indirect effect via its impact on Student Sense of Academic Support By its positive influence at both levels of social climate, Teacher Grade Preference appeared to serve as an important linkage between the two levels and to underscore the importance of teacher satisfaction with a Middle Years teaching assignment in the determination of the perceptions of school social climate in total.
7. In turn, Teacher Grade Preference was influenced by Level of Training and Student Mix Ratio. These relationships suggested that teachers who were satisfied with a Middle Years teaching assignment tended to be less qualified and tended to be found in schools with higher ratios of Middle Years students.
8. With the exception of one social climate dimension, Student Mix Ratio, was negatively associated with aggregate student perceptions of the Social climate dimensions involved. Of particular interest were the findings that schools with higher proportions of Middle Years students tended to be viewed by those students as having greater intergroup rivalry and friction, somewhat less emphasis on learning and skill development, and a lover commitment toward purpose, goal direction, and challenge.
9. Student Gender Ratio was present more often as an influence at the teacher level than at the student level. In addition, its impact was consistently positive, thus indicating that teachers tended to perceive the social climate dimensions more favorably when there were higher proportions of girls in the Middle Years cohort.
10. School Policies had a direct influence on several student perceived social climate dimensions While its impact was not substantial, it was, nevertheless, consistently positive, thus associating higher levels of student oriented policies with more favorable aggregate student ratings in the social climate dimensions involved. In addition, School Policies exhibited a weak, but positive, indirect impact on Student Satisfaction via Display of Work and School Appearance.
11. When found as an influence at the student level, Teacher Gender Ratio had a consistently inverse relationship with the social climate dimension involved, thus indicating that higher proportions of female Staff tended to be associated with less favorable aggregate student ratings. This finding should be interpreted with caution since schools with higher proportions of female staff were usually smaller schools and were overrepresented in the sample. In these instances, the proportional representation of Staff could be changed substantially between schools by a gender difference of only one or two teachers
12. With the exception of one social climate dimension, Teacher Stability exerted a consistently negative influence on the social climate dimensions at the teacher level. These results suggested that higher ratios of long term staff were associated with lower aggregate scores of the dimensions involved. This relationship is the inverse from that generally described in the literature and requires further consideration.
1. The findings respecting Type of School suggest that there is no clearly preferable way of organizing grade patterns. In some ways, 7 - 9 schools appeared to offer appropriate social environments. As a group, they had the highest mean score for 10 of the 13 dimensions and differed significantly from other groups on 3 dimensions. They also tended to attract teachers with a preference for grade 7 to 9 students. On the other hand, high proportions of Middle Years Students in the student cohort was negatively associated with same social climate dimensions. In this respect 7 - 9 schools were characterized by more peer group rivalry. It should be noted that K - 12 schools had the greatest variation in scores and, in all but one dimension, had the highest and lowest aggregate student scores. These findings suggest that grade-level pattern alone should not be the deciding factor in the organization of schools to serve Middle Years students.
2. It is clear from the findings that teacher satisfaction with a Middle Years teaching assignment has an important impact on social climate at the teacher and student levels, and particularly on Student Sense of Academic Support, a critical determinant at the student level. Based on these findings, it is recommended that senior officials take into account grade preference in the recruitment and placement of teachers
3. In-school administrators are often involved in the assignment of teaching loads. The findings concerning teacher preferences have implications for the in-school administrator in that teacher placement is often made on the basis of Subject specialization which may not take grade preferences into account.
4. Student attitudinal characteristics (Student Sense of Academic Support and Student Satisfaction) play a fundamental role in the development of school Social climate at the student level. There is modest evidence that student oriented policies relate positively to student climate. These findings may provide Boards and administrators with a point of reference, in planning staff inservice or in the development of policies and practice.
5. The findings of this study point to the association of positive attitudes at both the student and teacher level in the development of positive school climate. The saliency of Student Sense of Academic Support indicates the need for teacher who can build an environment in which students feel competent. Enjoying the grade-level and providing student oriented practices appear related to developing positive attitudes.
6. Several key findings of this study may provide the impetus for further investigations that Boards of Education may deem worthy of support. Some examples are presented for consideration.
(a) Teacher stability (longevity in the school setting) was associated negatively with school social climate, particularly at the teacher level. The inconsistency of this finding with notions presented in the school effects literature prompts further study to examine the impact of teacher stability in the Saskatchewan context.
(b) The strong relationship found for teacher grade preference at the teacher level and its corresponding impact at the student level raises questions related to recruitment practices for teachers of Middle Years students. A study of recruitment practices for Middle Years teachers may provide knowledge and understanding necessary for effective hiring practices.
(c) The relationship between teacher preference for a Middle Years teaching assignment and student perceptions of functioning adequately in the social environment draws attention once again to the teacher-pupil relationship. A study which seeks to identify and highlight the patterns of caring that exist for Middle Years students may provide insights concerning supportive school environments that foster academic growth.
(d) The lack of congruency between student perceived and teacher perceived School social climate suggests that the world of the school is different for teachers and students. An investigation that focuses on an explanation of the phenomenon may advance understanding of its relationship to school effectiveness.
7. Teachers who preferred Middle Years teaching assignments tended to have lover academic qualifications. This finding suggests that teachers who have a commitment to Middle Years students do not tend to actively pursue upgrading. Boards of Education could encourage the Colleges of Education to increase their emphasis on undergraduate and graduate programs that will have relevance for Middle Years teachers and applicability to Middle Years classrooms.
Back to: School Improvement