Purpose: The role of Teacher Innovative Behavior (TIB), in responding to systemic problems in educational systems and promoting "intrapreneurial" behavior has been recognized in recent times. A robust instrument that can help administrators and teacher educators gauge the levels of TIB among their teachers will facilitate the promotion of innovative behavior. Design/methodology/approach: This study tested a multidimensional innovative behavior inventory (IBI), innovation support inventory (ISI) and innovation output (IO) in a developing nation (India) context with public school teachers (n = 34,754), for reliability, validity, measurement invariance and structural invariance across caste, gender and subject groups. Findings: The IBI, ISI and IO showed good reliability and validity along with full measurement invariance at configural, metric and scalar levels. With respect to the structural parameters, the inventories exhibited invariance of factor variance and covariance, but not of factor means. Practical implications: Teacher innovative behavior (TIB) is seen by developing country education administrators as a tool to address difficult problems. With better measurement, it will be possible to identify teachers who need training in creativity and entrepreneurial behavior, teachers who might have developed innovative practices that could be used for teacher development, and ways of promoting competition among teachers. Originality/value: The study validates inventories, which were earlier tested in non-educational domains, for use with public school teachers of a developing country across gender, caste and subject groups.