Instructional Leadership and School Culture 102 Examining Instructional Leadership: A Study of School Culture and Teacher Quality Characteristics Influencing Student Outcomes Florida Journal of Educational Matthew Ohlson, Ph.D. Administration & Policy University of Florida Summer 2009 Volume 2, Issue 2 First, to examine the influence of teacher input characteristics and teacher perceptions of school culture on student absences. Second, to examine the influence of teacher input characteristics and teacher perceptions of school culture on out-of-school suspensions. Data was obtained for the 2006-2007 school year from 23 urban public elementary schools in Florida. Using the school as the unit of analysis, data was collected examining student absences and suspensions during the 2006-2007 school year reported by the Florida Department of Education's School Indicators Report. Surveys were administered to examine collaborative leadership, teacher collaboration, unity of purpose, professional development, collegial support, and learning partnership, identified by Gruenert and Valentine (1998) as the six components of the collaborative culture of a school. Surveys also documented teacher input characteristics such as years teaching, percent out of field, and highest degree obtained. Correlation using multiple regression was used to analyze the data. As the Unity of Purpose factor increased, the model predicted that student absences would decrease by 22.56%. In addition, the model predicted that when either the average years of experience for teachers within a school increased or when the Collaborative Leadership factor increased, student suspensions would decrease by 0.413%. and 4.81% respectively. Keywords: Instructional Leadership; Performance Framework; School Culture; Teacher Quality; Collaborative Leadership Introduction & Literature Review This research is a report of a statistical modeling study of the relationship between school leadership, school culture, teacher quality and the influence these variables have upon student outcomes. The research is based upon data gathered during the start of the 2007 school year and includes twenty-three public elementary schools in the state of Florida. The findings offer valuable insight into the characteristics of quality teaching, instructional leadership and school culture that demonstrate greatest impact on student attendance and suspensions and thus may influence educational policy, teacher training, educational leadership, and school reform initiatives. For policy makers to provide comprehensive teacher evaluation programs, it is paramount to understand what constitutes quality teaching. Linda Darling-Hammond (2000) found that student achievement increased and dropout rates decreased when teachers were certified in their field, obtained their master's degrees, and were enrolled in graduate studies. In addition, she contends that teacher preparation and certification had the Instructional Leadership and School Culture 103 strongest correlation for student achievement, more than any other school based factors. Furthermore, teaching in-field, in math and science particularly, led to increased student achievement (Goldhaber and Brewer, 1997). This need for school-based evaluation methods and support for professional development stresses the importance of effective educational leadership. Effective instructional leadership is generally recognized as the most important characteristic of school administrators (Hoy & Hoy, 2009). Cosner and Peterson (2003) go so far as to claim that promoting teacher professional development is the most influential educational leadership behavior. Principals and administrators are needed to lead educational improvement, foster effective change efforts, lead the implementation of new standards, and are central to shaping strong, professional school cultures (Deal & Peterson, 1998). The relationship between effective teaching and effective leadership is reinforced in the vital role of school culture. Among the numerous definitions of school culture, Deal and Peterson (1990) and Schein (1985) affirm that school culture refers to the deep patterns of values and beliefs and traditions that have been formed over the course of the school's history and which are understood by members of the school community. Peterson (2002) suggests that culture is built within a school over time as teachers, school leaders, parents and students work together. It is the school culture that often influences the staff development and professional growth that takes place within a school. Fullan and Steiglebauer (1991) contend that the key to successful change is not only a change in organizational structure but also more importantly a change in the culture. A positive school culture may have a significant influence on the academic and social success of the students within schools (Squires & Kranyik, 1996). When a school exhibits characteristics of a positive school culture, there are fewer suspensions, increased attendance rates, and increased achievement on standardized test scores (Anson et al. 1991, Becker &Hedges 1992). Purpose of the Study The purpose of the study is to determine if teacher quality characteristics and school culture components influence student attendance and suspension rates. Specifically this study will address the following questions: 1. Is there a relationship between the characteristics of teacher quality (the percentage of classes taught by out-of-field teachers, the percentage of teachers with advanced degrees, and the average years of experience for teachers within a school) and student attendance and suspension rates? 2. Is there a relationship between school culture factors, as measured by the School Culture Survey, and characteristics of teacher quality (the percentage of classes taught by out-of-field teachers, the percentage of teachers with advanced degrees, and the average years of experience for teachers within a school)? 3. Is there a relationship between school culture factors, as measured by the School Culture Survey, and student attendance and suspension rates? Instructional Leadership and School Culture 104 Conceptual Framework The theoretical basis for the research is grounded in the educational performance framework (Goldhaber & Brewer, 1997; Levin, 1998). A performance framework (Figure 1.2) may be conceptualized as having three main parts: inputs, processes, and outputs (Rouse & Putterill, 2003). As it relates to this study, "inputs" apply to the characteristics the individual brings to the workplace. Examples include the highest degree earned, the number of years teaching, teaching within field and state certification. Relevant literature will be examined to determine the ways in which these input characteristics are associated with quality teaching and influence student achievement. "Processes" refer to pedagogical development and practice in and outside of the classroom. Examples include the nature of collaboration with peers, administrator evaluations, and professional development activities. Aggregated at the school level, these variables form the heart of educational leadership (Leithwood, Jantzi, & Dart, 1990). Educational leadership research is examined to determine the influence upon student achievement and teacher working conditions. In addition, the components of school culture are analyzed to offer a more cohesive understanding of the relationship between effective educational leadership and a collaborative school culture (Leithwood, 1992. "Outputs" are the immediate and recurring indicators of students within a school. Examples include attendance rates and suspension rates. The relevant research will be examined to determine the influence these outcomes have upon student achievement. Furthermore, an analysis of these output variables will determine significant input and processes influence in previous research. Analysis will include the relationship between years of teaching experience and the level of student suspensions and the impact collaborative school culture may have upon reducing excessive absences. Methodology In order to answer the two research questions, data was obtained from 23 schools in Florida for the 2006-2007 school year. Sample schools were selected based upon their participation in a statewide school improvement program directed by the Lastinger Center for Learning at the University of Florida. The goal of the Lastinger Center for Learning is to offer comprehensive and continuous support to these schools as they work towards enhanced teacher efficacy and improved student achievement. These goals are addressed through the use of job-embedded professional development, teacher inquiry, collaborative culture building practices, and the use of data to help shape school policies. Teachers voluntarily completed the school culture survey at schools in various districts throughout the state including Collier, Miami-Dade, Pinellas, and Duval. The survey, administered in person by an individual external to the daily operation of the school, had an 85% response rate. The nature of the data positioned the school as the unit of analysis. Data Sources The outputs construct was operationalized by both student absences and suspensions during the 2006-2007 school year as reported by the Florida Department of Education's Florida School Indicators Report (n.d.a). Specifically, student absence data reflected the Instructional Leadership and School Culture 105 percentage of a school's population that was absent 21 or more days. Suspension data reflected the percentage of a school's population who were suspended outside of school during the 180 day academic school year. This percentage utilized an unduplicated- headcount to ensure the same student was not counted twice (Florida School Indicators Report, n.d.b). The inputs construct was operationalized by three variables reported in the Florida School Indicators Report. They include a measure of the percentage of classes taught by out- of-field teachers, the percentage of teachers with advanced degrees, and the average years of experience for teachers within a school. The process construct was operationalized by variables obtained from Gruenert and Valentine's (1998 ) School Culture Survey, which was administered to 23 elementary schools in three districts during the Spring of 2007. The School Culture Survey was inclusive of 6 factors and has been validated. Each factor had between four and eleven questions, to which respondents reported their perception on a five-point Likert scale, from 1 (strongly disagree) to 5 (strongly agree). This study incorporated four of the six factor scores reflective of the relationship between administrators and faculty: Collaborative Learning, Teacher Collaboration, Professional Development, and Unity of Purpose. "Professional Development" describes the degree to which teacher's value continuous personal development and school-wide improvement. Teachers remain knowledgeable about current and effective practices from workshops, seminars, colleagues, observations, and other professional resources. Continual growth, improvement and learning are the foundation of this component of school culture. Effective educational leaders support teachers with the resources and opportunity to partake in meaningful professional development (Lumpkin, 2008). The Unity of Purpose element of school culture describes the degree to which teachers work toward a common mission for the school. This involves an active collaboration amongst parents, teachers, students, support staff, administrators, and the local community toward setting and achieving a common goal. It is these common goals, plans, visions and values that become the focal point of everyone's efforts (Levin, 2001: Kotter 1996: Schein, 1992). The Collaborative Leadership component describes the degree to which school leaders establish and maintain collaborative relationships with the school staff. The educational leaders seek the input of the school community, value their ideas, and provide venues for their ideas to be expressed. In addition, collaborative leadership promotes new ideas, new risks and a sense of shared trust that allows for innovation to take place (Byrk & Schneider, 2002). Teacher Collaboration examines the extent to which teachers engage in constructive dialogue and conversations that further the educational mission, vision and goals of the school. Ideally, teachers throughout a school will work collectively and collaboratively, including such activities as mutual classroom observations, lesson modeling, grade-level and team planning, and evaluation and assessment of teaching practices (Bambino, 2002). Collegial support details the degree in which teachers work together in an effective and trusting manner. Collegial support may serve as a positive influence upon school culture and student achievement, with the presence of teacher collegiality, mutual respect amongst stakeholders, and a shared responsibility of meeting the needs of students (Blackmore, et al, 1996). Instructional Leadership and School Culture 106 Treatment of the Data Models inclusive of all variables were entered in SPSS to determine the proportion of the variance associated with the dependent variable. Stepwise regression was employed to determine the order of the variables as a theoretical basis for order of entry was not substantiated. In both models, assumptions of linearity, normality, independence and equal conditional variances were checked utilizing scatter and residual plots (Ruiz-Primo & Shavelson, 1996). Tolerance and variance inflation factors methods were applied to check for collinearity. Cook's Distance and DFBETAs were calculated to check for the influence of data points on the analysis. The decision to reject or fail to reject the null hypothesis was based upon a type I error rate of .05. Non-directional hypotheses were employed in the analysis, as significance, either positive or negative, was of interest to the researchers. Limitations The use of the school as the unit of analysis framed the discussion as an examination of instructional leadership, as opposed to individual level analysis of teachers. It may, however, be possible to utilize the performance framework at the teacher level, however, data to undertake such an analysis was not available to the researchers. Results The purpose of the study was to examine the association of outputs to inputs and processes at 23 schools in Florida. Analysis of the data provided results for each of the two research questions developed for this study. Student Absences To test the first research question a stepwise, multiple regression analysis was performed with the dependent variable as student absences. It was determined that the influence of outliers was not present, as all independent variables were within the acceptable ranges, thereby maintaining the sample size at 23 schools. Descriptive statistics, presented in Table 1, indicated a fairly tight dispersion of scores for the four school survey factors and the average years of experience for teachers within these schools. As it pertains to schools whose percentage of students were absent more than 21 days, the mean of 8.83% is questionable as one standard deviation for this sample was 6.29. As such results of the analysis should be viewed with caution. As it pertained to the process variables operationalizing the process construct, the standard deviations for the percentage of teachers with advanced degrees and the average years of experience indicated a relatively close distribution. The percentage of classes taught by out- of- field teachers had to be removed from the analysis as the distribution was too wide to be included. Two-tailed, zero-order correlations between students with 21 or more absences and one independent variable - Unity of Purpose (-.489) - was significant at the 0.05 level. The other variables are presented in Table 2. The stepwise, multiple regression analysis resulted in the standard regression equation, Instructional Leadership and School Culture 107 ST_AB = 36.055 - 23.564 (UP) + 17.274 (CL) (Eq. 1) where ST_AB represents the percentage of students in a school with 21 or more absences, UP represented the Unity of Purpose factor, and CL represented the Collaborative Leadership factor (Table 3). This model accounted for 43.9% of the variance in the percentage of students with greater than 21 absences (Table 4). The Analysis of Variance (ANOVA) revealed that the model was statistically significant, F (2, 20) = 7.821, p < 0.01 (Table 5). These results suggest that a large percentage of variance in the percentage of students with 21 or more absences in a school was associated with a school's Unity of Purpose and Collaborative Learning. The associations, however, differed for the two independent variables in that as the Unity of Purpose factor increased by a score of one, it would be predicted that student absences would decrease by 22.56%. Conversely, as Collaborative Leadership in a school increased by 1 Likert-scale point, the model predicted that the percentage of students with 21 or more absences would increase by 17.27%. Out-of-School Suspensions To test the second research question a stepwise, multiple regression analysis was performed with the dependent variable as out-of-school suspensions. It was determined that the influence of outliers was present in two schools, reducing the sample size to 21. Descriptive statistics, presented in Table 6, again indicate a fairly tight dispersion of scores for the four school survey factors. As it pertains to out-of-school suspensions, the mean of 1.93% is questionable as one standard deviation for this sample was 2.29. As such results of the analysis should be viewed with caution. As it pertained to the process variables operationalizing the process construct, the standard deviations for the percentage of teachers with advanced degrees and the average years of experience maintained a relatively close distribution when compared to their respective statistics for the previous model inclusive of all 23 institutions. The percentage of classes taught by out-of-field teachers had a mean of 1.01% with a standard deviation of .91, again a questionable data point. Two-tailed, zero-order correlations between a school's percentage of out-of-school suspensions and four independent variables were significant at the 0.05 level, with Collaborative Leadership being the strongest (.000). The correlations for all variables are presented in Table 7. The stepwise, multiple regression analysis resulted in the standard regression equation, SUS = 22.842 - 4.811 (CL) - 0.413 (T_AVGYR) (Eq. 2) where SUS represents the percentage of school's population who were suspended out-of- school, CL represents the Collaborative Leadership factor, and T_AVGYR represents the average years of experience for teachers within a school (Table 8). This model accounted for a 74.9% of the variance in the percentage of students suspended out-of-school (Table 9). The Analysis of Variance (ANOVA) revealed that the model was statistically significant, F (2, 18) = 26.813, p < 0.01 (Table 10). These results suggest that a large percentage of variance in the percentage of school's Instructional Leadership and School Culture 108 population who were suspended out-of-school was associated with a school's Collaborative Leadership and average years of experience for teachers within a school. The associations, however, differed for the two independent variables in that as the Collaborative Leadership factor increased by a score of one, it would be predicted that student suspensions would decrease by 4.81%. Additionally, as the average years of experience for teachers within a school increased by 1 Likert-scale point, the model predicted that the percentage of students suspended out-of-school would decrease by 0.413%. Discussion Analysis of the data provided results for this pilot study on the influence of instructional leadership on student outputs. The purpose of the study was examined via the examination of the research questions addressing the association of a school's outputs to its inputs and processes. A discussion of the results follows. Student Absences Analysis of the data revealed that two process factors were significantly associated with a school's percentage of students with 21 or more absences during a school year. One factor was the perceived Unity of Purpose, (degree to which teachers work toward a common mission for the school) and the other was a perceived level of Collaborative Leadership (degree to which school leaders establish and maintain collaborative relationships with the school staff ) at a school. Unity of Purpose. The decrease in excessive school absences in relation to an increase in a "Unity of Purpose" felt by school faculty may have profound implications for a school community. When school leaders and teachers embrace a common mission and vision for teaching and learning, it decreases the likelihood of students being away from the learning environment. In turn, this increase in attendance will increase the likelihood of student achievement (Bedi & Marshall, 1999). Many schools have fallen prey to the "Christmas Tree" phenomenon in which new programs, new initiatives, and new goals are continually introduced and never given the support and resources necessary to impact teaching and learning. The end result is a culture that is reluctant to change in fear that this change will soon pass. Peterson (2002) contends that a school with a negative or toxic culture does not value professional learning and ultimately hinders the success of the teachers and students. In contrast, the development of a high-performance learning culture is influenced by school vision and mission embraced by the school community. With a "Unity of Purpose," teachers, students, and parents understand the vision and mission as being more than just a slogan. Instead it becomes the basis or foundation for all decision-making and the compass guiding the school community. The data illustrates the power of having a unity of purpose. If students feel that they are headed towards their goals and those goals are aligned with the teaching goals of the faculty, there may be a stronger desire to attend school. Collaborative Leadership. Collaborative Leadership had a negative impact on student attendance whereas the excessive absence rates increased as the Collaborative Leadership factor increased.. This relationship may be the result of many factors including the small sample size or the unique characteristics of the participating schools (demographics, openness to research and data collection). Instructional Leadership and School Culture 109 Yet, questions do arise including the notion that if a school leader is more collegial and has developed stronger relationships with the teachers does that impact the principal's authoritative role? Is there a breakdown in adhering to attendance, discipline, and academic policy when there is a greater emphasis placed upon relationship building within a school? Are students more prone to absences if school administration or school principals focus more of their time, resources, and energy on establishing and maintaining collaborative relationships within a school rather than policy or management issues? This is an area in need of further research. Collaborative Leadership was also found to impact "out of school suspensions." The data suggest that an increase in a culture that demonstrates collaborative leadership will decrease the number of out of school suspensions. This relationship is attributed to the fact that when teachers and school administration have collaborative relationships they may be more willing and able to address discipline issues before they require the severity of an out of school suspension. Furthermore, school administrators who are more involved with teachers rather than in their office and out of the classroom context may be more likely to help teachers address behavioral issues. These school administrators may take the time to offer strategies and professional development opportunities that will augment the skill set of the school's teachers. In addition, collaborative relationships between administration and teachers may increase the likelihood of those school administrators actually being present and involved in the classroom setting/context. The mere presence of the school principal in a classroom may decrease the likelihood of behavioral issues. Another aspect of this discussion is the subjective nature of these factors in comparison to the student absences. Unlike the previous data that showed a seemingly detrimental relationship (absences increase as collaborative leadership increases), the out of school suspensions are often determined by the teacher and school administration. There is an opportunity for judgment and discrepancy by the school faculty where school absences are determined solely by the action of the student and his or her family. When a student is disciplined, teachers and administration may meet to discuss the offense and offer a course of action. If the school administration and teacher have a collaborative relationship they may be more willing to work together to come up with more constructive solutions in lieu of out of school suspensions, an option that removes the student from school and thus drastically impacts his or her academic growth. Average Years of Teaching Experience. The data implies that teacher experience also influences the number of out of school suspensions. As the number of years of teaching experience increase, the number of out of school suspensions decreases (albeit smaller than the other relationships). When a teacher has more experience, he or she may be better equipped to address the complexities and issues surrounding student disciplinary issues and classroom management in a manner other than out of school suspensions. Research suggests that school suspension has been found to significantly correlate with poor academic achievement (Raffaele-Mendez, Knoff, & Ferron, 2002) and solutions should be sought as alternatives to simply remove the student from the school setting. The out of school suspension should be a last resort and with experience, teachers can augment their teaching strategies in a manner that increases student engagement and student achievement- factors that decrease behavioral issues (Klem & Connell, 2004). Instructional Leadership and School Culture 110 Implications It appeared, from this study, that the factors most prominently associated with student-based school outputs were the "Unity of Purpose" and "Collaborative Leadership" process factors related to characteristics of a collaborative school culture. The Unity of Purpose element of school culture describes the degree to which teachers work toward a common mission for the school. This involves an active collaboration amongst parents, teachers, students, support staff, administrators, and the local community toward setting and achieving a common goal. The Collaborative Leadership component describes the degree to which school leaders establish and maintain collaborative relationships with the school staff. The educational leaders seek the input of the school community, value their ideas, and provide venues for their ideas to be expressed. The role of school culture within this research may well have a lasting and wide- ranging influence on school improvement and school reform initiatives. Working collaboratively with school leadership and teachers to strengthen the culture of the school, with the intent of improving teaching practice and student learning, is a promising school reform strategy (Ross et al, 2007). Furthermore, the analysis of each factor for a significant correlation between variables will shape current and future school improvement and reform policies and initiatives at the local, state, and national levels. District personnel, administration, and school communities as a whole will be able to develop more targeted and effective plans and policies for bringing about change in their school culture; and ultimately, student achievement. The sample population also serves to enhance the significance of the findings. The schools participating in the research represent elementary schools in urban and rural districts throughout the state of Florida with high rates of minority students, teacher turnover, significant free and reduced lunch student populations, and historically low student achievement scores. This research seems likely to enrich the knowledge base available to policy makers, administrators, and district personnel serving teachers and students in districts throughout the nation that mirror the same attributes of the sample population. The methods chosen for this study provide insight into the impact school culture and the characteristics of effective teaching may have upon student outcomes. Rather than focus directly on standardized test scores, this study focuses on two distinct variables- attendance rates and out- of-school suspensions. These two factors have been found to influence student achievement as well as student graduation rates (Sheldon, 2003). This research will also have major implications for institutions of higher education in relation to the two teacher quality characteristics being studied (Liston, et al., 2008); (1) teachers with advanced degrees, (2) teachers certified in field. Numerous studies suggest that teachers' degree levels consistently demonstrated strong relations with increased student achievement (Greenwald, Hedges, & Laine, 1996). Yet, in this study, these variables do not show to influence student outcomes. Thus, this research may have major implications for alternative certification programs such as Teach for America and Troops to Teachers that fill teaching vacancies in areas with dire teaching shortages. This data supports previous research that suggests teachers with advanced degrees resulted in lower student achievement in certain subjects and that emergency certification teachers performed as well as traditionally certified teachers (Walsh, 2007). Instructional Leadership and School Culture 111 In conclusion, this research has helped to illustrate the importance of a collaborative school culture. Specifically, a school that embraces a common mission and a school leader that establishes a collaborative relationship with faculty may witness significant improvements in relation to suspensions, attendance, and ultimately; student achievement. Thus, by understanding the importance of these school culture factors and by discovering the practices and strategies taking place in schools that positively influence these variables, we can more effectively shape school improvement initiatives. References Bedi, Arjun S., and Jeffery H. Marshall. (1999). School attendance and student achievement: Evidence from rural honduras. Economic Development and Cultural Change, 47:3, 657-682. Berube, B., & Dexter, R. (2006). Supervision, evaluation and NCLB: Maintaining a most highly qualified staff. Catalyst for Change, 34(2), 11-17. Corbett, D., Wilson, B., & Williams, B. (2002). Effort and excellence in urban classrooms: Expecting—and getting—success with all students. New York: Teachers College Press. Cosner, S., & Peterson, K. (2003). Building a learning community. Leadership, 32 (5), 12-15. Danielson, C., & McGreal, T. (2000). Teacher evaluation to enhance professional practice. Alexandria, VA: Association for Supervision and Curriculum Development Darling-Hammond, L. (1996). What matters most: A competent teacher for every child. Phi Delta Kappan, November, 1996. Darling-Hammond, L. (2000). Teacher quality and student achievement: A review of state policy evidence. Education Policy Analysis Archives, 8 (1); at: http://epaa.asu.edu/epaa/v8n1. Deal, T. E., & Peterson, K. D. (1998). Shaping school culture: The heart of leadership. San Francisco: Jossey-Bass. Deal, T. E., & Peterson, K. D. (1990). The principal's role in shaping school cultures, Washington, DC, US Department of Education. DiPaola, M. F., & Walther-Thomas, C. W. (2002). Principals and special education: The critical role of school leaders. Center of Personnel Studies in Education, University of Florida. Retrieved March 25, 2008, from http://copsse.org/. Edweek.org. (2004). Teacher quality. Retrieved January 10, 2008, from http://www.edweek.org/rc/issues/teacher-quality/. Florida School Indicators Report. (n.d.a.). Tallahassee, FL: Florida State Board of Education. Retrieved May 14, 2008 from http://data.fldoe.org/fsir/default.cfm. Florida School Indicators Report. (n.d.b.). Tallahassee, FL: Florida State Board of Education. Retrieved May 28, 2008 from: http://data.fldoe.org/fsir/indicator_desc_cfm. Fullan, M. (2001). Leading in a culture of change. San Francisco: Jossey-Bass. Fullan, M. (2003). The moral imperative of school leadership. Toronto: Corwin Press and the Ontario Principals' Council. Fullan, M.G., & Steigelbauer, S. (1991). The new meaning of educational change. New York: Teachers College Press. Gitomer, D. H. (2007). Teacher quality in a changing policy landscape: Improvements in the teacher pool. Princeton, NJ: Educational Testing Service.