American K-12 students are widely believed to have lower academic achievement in STEM (Science, Technology, Engineering, and Math) subjects than in most developed countries. However, because education is constitutionally a state matter, student performance can vary widely from jurisdiction to jurisdiction. States also have diverse education systems, so adopting education policies from the top performers without other considerations may not always improve student academic achievement. Hence, this study aimed to group states into clusters based on their similarity in education characteristics to promote more effective policy diffusion. To accomplish that, principal component analysis was applied to 26 variables (representing four different education-related aspects) to reduce the dimensionality of the data at the elementary school, middle school, and the whole K-12 levels. Six principal components at each level, covering more than 80% of the total variance of the initial datasets, were used in hierarchical clustering. Comparing the resulting dendrograms and examining the disparities in the National Assessment of Educational Progress (NAEP) scores in math and science, eight to eleven state clusters were proposed for each education level. In each cluster or subcluster, the states with the highest NAEP score were assumed to have effective policies that others in the same group should adopt. The policy diffusion pathways among states are visualized in the forms of six interstate policy diffusion maps (IPDMs) for either math or science at each education level. These IPDMs allow states to narrow the search for model jurisdictions that may possess desirable K-12 STEM education policies.
","abstract_html":"<p>American K-12 students are widely believed to have lower academic achievement in STEM (Science, Technology, Engineering, and Math) subjects than in most developed countries. However, because education is constitutionally a state matter, student performance can vary widely from jurisdiction to jurisdiction. States also have diverse education systems, so adopting education policies from the top performers without other considerations may not always improve student academic achievement. Hence, this study aimed to group states into clusters based on their similarity in education characteristics to promote more effective policy diffusion. To accomplish that, principal component analysis was applied to 26 variables (representing four different education-related aspects) to reduce the dimensionality of the data at the elementary school, middle school, and the whole K-12 levels. Six principal components at each level, covering more than 80% of the total variance of the initial datasets, were used in hierarchical clustering. Comparing the resulting dendrograms and examining the disparities in the National Assessment of Educational Progress (NAEP) scores in math and science, eight to eleven state clusters were proposed for each education level. In each cluster or subcluster, the states with the highest NAEP score were assumed to have effective policies that others in the same group should adopt. The policy diffusion pathways among states are visualized in the forms of six interstate policy diffusion maps (IPDMs) for either math or science at each education level. These IPDMs allow states to narrow the search for model jurisdictions that may possess desirable K-12 STEM education policies.</p>","abstract_has_math":false,"creators":["Tran, Vinh Q"],"institution":null,"degree_name":"Philosophy, PhD","degree_level":"Open Access Dissertation","degree_discipline":"School of Educational Studies","degree_department":null,"school":null,"contributors":["Dr. Thomas F. Luschei","Dr. Susan J. Paik","Dr. Mark Abdollahian"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2022,"date_issued":"2022-01-01T08:00:00Z","date_published":"2022-01-01T08:00:00Z","updated_at":"2026-07-24T01:39:59Z","subjects":["Hierarchical clustering; K-12 education; NAEP; NCES; Policy diffusion; STEM education","Education"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://scholarship.claremont.edu/cgu_etd/423","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Dr. Thomas F. Luschei","Dr. Susan J. Paik","Dr. Mark Abdollahian"]},{"key":"dc:creator","label":"Author","values":["Tran, Vinh Q"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2023-02-23T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["School of Educational Studies"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Open Access Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Philosophy, PhD"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Hierarchical clustering; K-12 education; NAEP; NCES; Policy diffusion; STEM education","Education"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://scholarship.claremont.edu/cgu_etd/423"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["American K-12 students are widely believed to have lower academic achievement in STEM (Science, Technology, Engineering, and Math) subjects than in most developed countries. However, because education is constitutionally a state matter, student performance can vary widely from jurisdiction to jurisdiction. States also have diverse education systems, so adopting education policies from the top performers without other considerations may not always improve student academic achievement. Hence, this study aimed to group states into clusters based on their similarity in education characteristics to promote more effective policy diffusion. To accomplish that, principal component analysis was applied to 26 variables (representing four different education-related aspects) to reduce the dimensionality of the data at the elementary school, middle school, and the whole K-12 levels. Six principal components at each level, covering more than 80% of the total variance of the initial datasets, were used in hierarchical clustering. Comparing the resulting dendrograms and examining the disparities in the National Assessment of Educational Progress (NAEP) scores in math and science, eight to eleven state clusters were proposed for each education level. In each cluster or subcluster, the states with the highest NAEP score were assumed to have effective policies that others in the same group should adopt. The policy diffusion pathways among states are visualized in the forms of six interstate policy diffusion maps (IPDMs) for either math or science at each education level. These IPDMs allow states to narrow the search for model jurisdictions that may possess desirable K-12 STEM education policies.
"]},{"key":"dc:title","label":"Title","values":["Constitutionally Confederated: Developing an Interstate Policy Diffusion Framework for K-12 STEM Education"]}]}],"canonical_facts":{"dc:contributor":["Dr. Thomas F. Luschei","Dr. Susan J. Paik","Dr. Mark Abdollahian"],"dc:creator":["Tran, Vinh Q"],"dc:date.available":["2023-02-23T08:00:00Z"],"dc:description.abstract":["American K-12 students are widely believed to have lower academic achievement in STEM (Science, Technology, Engineering, and Math) subjects than in most developed countries. However, because education is constitutionally a state matter, student performance can vary widely from jurisdiction to jurisdiction. States also have diverse education systems, so adopting education policies from the top performers without other considerations may not always improve student academic achievement. Hence, this study aimed to group states into clusters based on their similarity in education characteristics to promote more effective policy diffusion. To accomplish that, principal component analysis was applied to 26 variables (representing four different education-related aspects) to reduce the dimensionality of the data at the elementary school, middle school, and the whole K-12 levels. Six principal components at each level, covering more than 80% of the total variance of the initial datasets, were used in hierarchical clustering. Comparing the resulting dendrograms and examining the disparities in the National Assessment of Educational Progress (NAEP) scores in math and science, eight to eleven state clusters were proposed for each education level. In each cluster or subcluster, the states with the highest NAEP score were assumed to have effective policies that others in the same group should adopt. The policy diffusion pathways among states are visualized in the forms of six interstate policy diffusion maps (IPDMs) for either math or science at each education level. These IPDMs allow states to narrow the search for model jurisdictions that may possess desirable K-12 STEM education policies.
"],"dc:identifier":["https://scholarship.claremont.edu/cgu_etd/423"],"dc:subject":["Hierarchical clustering; K-12 education; NAEP; NCES; Policy diffusion; STEM education","Education"],"dc:title":["Constitutionally Confederated: Developing an Interstate Policy Diffusion Framework for K-12 STEM Education"],"thesis:degree_discipline":["School of Educational Studies"],"thesis:degree_level":["Open Access Dissertation"],"thesis:degree_name":["Philosophy, PhD"]},"updated_at":"2026-07-24T01:39:59Z"}