Kurt F. Geisinger
Date of this Version
Usher-Tate, B. (2020). Estimating School-Level Achievement in Belize. University of Nebraska-Lincoln.
This dissertation consists of five chapters: introduction, literature review, methodology, results, and discussion with final thoughts. The research design of this dissertation study attended to structures, cultures, and characteristics associated with, and specific to, the Belize education system. The processes for data collection and types of analyses were appropriate, yielded meaningful results, and served as a segue for national application. The Belize Educator Survey was developed to capture the educators’ voices and illuminate their relationship to educational achievement in Belize. The Belize Educator Survey was piloted and revised with direct input from educators and experts who work in the Belize education system. In this dissertation study, the Belize Educator Survey was the primary data collection instrument for a district-wide, full-coverage survey approach in one of the six districts in Belize. All primary school educators in the district were invited to participate (N = 524), and the response rate was 60.11% (n = 315).
One of the end goals of this study was to create a pragmatic way of estimating School-Level Achievement that incorporated data about all grade levels and include the voice of all educators associated with the school. Two methods, the Weighted-Indicator Estimation Protocol (WISP) and a Multilevel Achievement Estimation Protocol (MAEP), were compared. The Weighted-Indicator Scores Protocol estimation uses a combination of classic statistical analyses, while the Multilevel Achievement Estimation Protocol version relies on a multilevel approach that estimates the within- and between-school statistics simultaneously. The 2018 Primary School Examinations school-level performance served as the primary criterion-referenced variable. Analyses of variances, comparisons of standard errors of the mean, and rank-order matchings show that the school-level estimates derived from the Weighted-Indicator Scores Protocol and Multilevel Achievement Estimation Protocol methods are statistically different.
Possible extensions of this study could identify or develop additional measures of academic performance that align to grade-level expectations in the Belize education system. The statistically significant negative associations between educators’ perceptions, School Size, and Location are also worthy of investigation for strategic planning purposes.
Advisor: Kurt F. Geisinger