Date on Master's Thesis/Doctoral Dissertation


Document Type

Doctoral Dissertation

Degree Name

Ph. D.

Cooperating University

Western Kentucky University

Department (Legacy)

Department of Leadership, Foundations, and Human Resource Education

Committee Chair

Miller, Stephen K.

Author's Keywords

Leadership; Instruction; High school principal; Scholastic audit; Curriculum; Kentucky education reform


Educational leadership--Kentucky; Educational accountability--Kentucky; Educational change--Kentucky; School improvement programs--Kentucky


In 1983 the National Commission on Excellence in Education released A Nation at Risk, which triggered an extended era of school reform culminating in today's accountability movement. In Kentucky the school improvement process, in which principals play an integral part, is based on the Standards and Indicators for School Improvement (SISI--with nine standards and 88 indicators overall) and the Scholastic Audit, a measurement tool based thereon. Murphy's (2004) mediated model of instructional leadership recognizes that complex school reform necessitates both direct and indirect skills. The central research question captures the purpose of this study, testing Murphy's model: Based on Scholastic Audits, what are the effects of instructional leadership on Kentucky accountability scores, as mediated by curriculum and instruction? Secondary data from 83 high schools in Kentucky are examined to determine the direct and indirect effect of leadership on student achievement. Descriptive statistics were calculated for nine demographic factors and the composite Academic Index (AI), an accountability measure. Factors for Standard 7 (Leadership), Standard 1 (Curriculum), and Standard 3 (Instruction) were derived from exploratory factor analysis of the 11, 7, and 8 indicators, respectively, for the standards. Cronbach's alpha and inter-scale correlations were also calculated. Simultaneous multiple regressions examined relationships among demographic factors, three standards, and AI. The full mediated effects model was tested with hierarchical regression. The three sets of indicators each produced a single factor solution explaining 45.4%,52.9%, and 48.2%, respectively of the variance. Cronbach's alpha ranged from .837 to .873. Standard 7 (Leadership) explained 39%, 40%, and 20%, respectively, of the variance on Curriculum, Instruction, and the AI. Standards 1 and 3 combined explained 40% of the AI. Demographic factors explained 65% of variance on AI; each of the three standards was essentially independent of demographic controls. In final hierarchical model, Leadership added .06 to the variance explained in Step 2. Step 3 with Curriculum and Instruction added produced an additional.05; all that influence came from Standard 3 (Instruction). The final model had an effect size of 76%, establishing empirical support for Murphy's mediated effects model of instructional leadership for high schools in Kentucky.