The van Hiele theory and van Hiele Geometry Test have been extensively used in mathematics assessments across countries. The purpose of this study is to use classical test theory (CTT) and cognitive diagnostic modeling (CDM) frameworks to examine psychometric properties of the van Hiele Geometry Test and to compare how various classification criteria assign van Hiele levels to students. The findings support the hierarchical property of the van Hiele theory and levels. Using conventional and combined criteria to determine mastery of a level, the percentages of students classified into an overall level were relatively high. Although some items had aberrant difficulties and low item discrimination, varied selection of the criteria across levels improved item discrimination power, especially for those items with low item discrimination index (IDI) estimates. Based on the findings, we identify items on the van Hiele Geometry Test that might be revised and we suggest changes to classification criteria to increase the number of students who can be assigned an overall level of geometry thinking according to the theory. As a result, practitioners and researchers may be better positioned to use the van Hiele Geometry Test for classroom assessment.