Principals in Saskatchewan Rural Schools: Their Leadership Behaviors and School Effectiveness

OBJECTIVE
To compare the behaviors of rural school principals, examining their leadership behaviors and any relationship with school effectiveness.

METHODOLOGY
The sample was limited to school principals and their schools in the province of Saskatchewan, Canada during the 1988-89 academy year at three types of schools: elementary, high, and kindergarten to grade twelve, in two types of settings (rural and urban). Random sample generated responses from 62 principals (69% response rate) and 450 constituents in the rural setting and the urban sample included 29 (72.5% response rate) principals and 200 constituents, completing the LPI-Self and LPI-Observer (respectively), and a modified version of the San Diego County Office of Education Effective Schools Survey. 86.5% of the principals were men, nearly all held college degrees, and the majority (48.3%) were between 36-45 years of age, with tenure as a principal being about 3-7 years (48.3%). Most of the constituents were teachers (72%), men (72%), between 36-45 years of age (40%), with 3-7 years (44%) at the school.

KEY FINDINGS
Significant differences were found in the leadership behaviors of principals in effective and noneffective schools (NOTE: Principals and teachers significantly differed in their perceptions of whether a school was effective or not). Principals perceived significant differences in their Inspiring behaviors. Constituents (teachers) saw principals' behavior as significantly different in terms of Challenging, while combining principal and constituent effectiveness assessments resulted in significant differences for the leadership behavior of Inspiring. No significant differences were found in the LPI-scores of principals based upon rural or urban setting or upon type of school-- these findings were consistent across principal, constituent, and principal-constituents combined groupings. Stepwise regression analysis found Inspiring and Enabling to account for 48.2 percent of the variance around principals' effectiveness (as measured by constituents; F = 40.86, p < .0001).