The purpose of this national study was to construct an instrument that reliably and effectively measures the teacher self-efficacy of preservice and in-service middle school science teachers to teach engineering practices and to apply the engineering design process in the science classroom. Two hundred fifty-nine respondents provided feedback for the 46-item instrument, Efficacy to Teach Engineering Practices (E2TEP), using a 6-point Likert scale. An accumulation of evidence was gathered in test content, response processes, internal structure, relationship of the construct to other variables, and consideration of the consequences of test use. The unitary concept of validity underpinned this mixed-methods study. Qualitative components integrated an 11-member expert group with backgrounds in science, engineering, STEM, invention education, and career technical education to assist in instrument item development. Five cognitive interviews from the target audience further refined the instrument. Quantitative components such as an exploratory factor analysis (EFA) revealed an eight-factor solution to factors necessary to effectively assess middle school preservice and in-service science teachers’ self-efficacy in teaching engineering practices and applying the engineering design process in the science classroom. Moreover, a scree plot with a total variance of 66 percent explained provided additional evidence of validity. Several independent samples t-tests and an ANOVA were performed to verify additional statistically significant differences between groups. Cronbach alpha reliability tests revealed an overall instrument reliability coefficient of .943. Reliability scores and interpretations of an accumulation of validity evidence supports that the E2TEP measures what it is intended to measure.