Predicting and Modeling Superintendent Turnover

Abstract

This study examines the longevity in office of superintendents in 292 school districts, using data from the period 1975-99. Random samples of districts from across the United States, as well as all school districts from North Carolina, were studied. Data were analyzed by survival analysis techniques, using information on superintendents, districts, and school boards as predictors. Superintendent tenure has not changed significantly since 1975-79, averaging 6–7 years over the whole period. Significantly related to survival in office were level of school board involvement in management, support for needed construction, merger of school systems, district poverty level, and superintendent postgraduate education.