Date of Award


Document Type


Degree Name

Doctor of Philosophy


Higher Education Administration EdD

First Advisor

Rudolf E. Klimes

Second Advisor

Lyndon G. Furst

Third Advisor

Daniel A. Klein


Problem. A major component of state-level educational accountability systems is the specification of goals upon which other components of the systems are established. The use being made of these goals at the local-school-district level in states where they exist has not been formally investigated.

This study examines the extent of use being made of state- level educational goals in the development of local-district goals and programs in the states of Colorado, Maryland, Michigan, and Rhode Island.

Method. The writer developed a six-item survey instrument to examine specific uses of state-level goals by local districts. This was sent to 766 local school-distrlct superintendents in Colorado, Maryland, Michigan, andRhode Island. There were 608 or 80 percent usable returns available for analysis. The subjects were asked to indicate a rank selection ranging from "always" to "never" regarding Che use of state goals in each category.

Six hypotheses were developed for the study. The first and second hypotheses related to the use of state-level goals in local district goal and program development. Hypothesis three related to the question of state and local district goal compatibility. The fourth and fifth hypotheses were concerned with the structure and adequacy of the state's assessment program. Hypothesis six related to state-level goals providing direction and purpose for local educational programs. The Kolmogorov-Smimov nonparametrlc test was used to test the data.

Results. Null hypotheses specifying there would be no significant difference in the number of times each of five ranks identified as, "always," "usually," "sometimes»" "infrequently," and "never," would be selected by respondents were formulated. All six null hypotheses were rejected at the .05 level of significance, with the point of maximum divergence between what was expected and what was observed occurring at the rank specified as "usually".

In subgroups of each of the states and small-sized school districts (less than 3,000 pupils), medium-sized school districts (3,001 to 10,000 pupils),and large-sized school districts (over 10,000 pupils) the points of maximum divergence between the expected and observed were statistically significant for at least four and usually six of the hypotheses tested. The consistent exception was Rhode Island where the six null hypotheses were retained.

Conclusions. (1) State-level educational goals are influential in the determination of local-district goals and the development of local- district programs in the states of Colorado, Maryland, and Michigan. (2) Superintendents of local school districts in the states of Colorado, Maryland, and Michigan feel their state departments of education view local-district and state-level goal compatibility as desirable. (3) Superintendents of local school districts in the states of Colorado, Maryland, and Michigan feel their state-level assessment programs are structured to evaluate educational outcomes in light of state-level goals and that they do so adequately. (4) Superintendents of school districts in Rhode Island do not support the value of educational goals in local districts to the same extent as those in Colorado, Maryland, and Michigan. (5) School district student population has no influence on the extent to which state-level goals are considered useful at the local-district level in the states in which the study was conducted.

Subject Area

Education and state, State departments of education