The purpose of this paper is to describe the development/reliability/validity of a teacher instrument to assess the organizational coupling structure (OCS) of schools. A secondary purpose is to describe the implications of this work for understanding school coupling structure as a complex, "paradoxical" measurement and theoretical construct. The study continues the development and refinement of a multidimensional inventory--the 98-item Organizational Coupling Structure Inventory-Teacher Form (OCSI-T)--derived from a previous pilot work. In the spring of 1989, all teachers in 96 elementary and secondary schools (ESS) were mailed the OCSI-T, a demographic information form, and the 8-item Index of Perceived Organizational Effectiveness. Usable data were received from 1,843 teachers (an overall response rate of 61%) from 73 schools. Standardized student achievement and school average daily attendance data were also collected. Descriptive statistics, instrument structure, reliability, and validity assessments were conducted. The OCSI-T is a reasonably valid and reliable inventory for measuring five dimensions of OCS in schools from the teacher perspective. The OCSI-T dimensions were largely statistically independent and showed moderate to high internal consistency reliabilities. The coupling perspective appears more useful in understanding effective schools as organizations than in understanding effective schools in terms of school productivity or indices of school holding power. The findings relative to manipulative control support the paradoxical view of the coupling construct. Overall, the OCSI-T coupling dimensions did not account for considerable variation in school achievement. (RLC)