Organization, Information and Learning Sciences ETDs

Publication Date

Summer 8-1-2023

Abstract

Chapter 2 introduces an innovative method integrating Geographic Information Systems and Machine Learning to evaluate educational policy outcomes. This approach overcomes the limitations of traditional statistical methods and effectively analyzes complex relationships between educational data and demographic factors.

The study in Chapter 3 developed a prediction model for school report cards in New Mexico using census data and applying the methods described in Chapter 2. The accurate predictions suggested that the school accountability system disproportionately assigned failing grades to schools serving minoritized students. The study also confirmed that status measures depend more on demographics than growth measures.

The study in Chapter 4 used eye-tracking data to analyze collaborative decision-making during complex problem-solving. It revealed relationships between students’ status and excluding agency and predicted, through cross-recurrence analysis and machine learning, instances of excluding agency with 77% accuracy. The study further reveals the evolution of shared and excluding agency dynamics over time.

Degree Name

Organization, Information and Learning Sciences

Level of Degree

Doctoral

Department Name

Organization, Information & Learning Sciences

First Committee Member (Chair)

Vanessa Svihla

Second Committee Member

Karl Benedict

Third Committee Member

Scott Hughes

Fourth Committee Member

Meeko Oishi

Language

English

Keywords

School Accountability, Geographic Information Systems, Machine Learning, Random Forest, Quantitative Critical Theory, Collaboration in Learning Groups, Excluding Agency, Cross-Recurrence Quantification Analysis, Eye-Trackers

Document Type

Dissertation

Download

Available for download on Friday, August 01, 2025

Share

COinS