Despite the popularization of integrated approaches to teaching science, technology, engineering, and mathematics (STEM) in policy documents, standards, and classrooms over the past several years, research related to the teaching of K-12 integrated STEM education continues to be impeded by the lack of observational tools available to education researchers. The work presented here uses a new observation protocol - the STEM Observation Protocol (STEM-OP) - designed for measuring the degree of integrated STEM teaching in K-12 science and engineering classrooms. The STEM-OP includes 10 items with four descriptive levels for each item (scored 0-3): 1) Relating content to students' lives, 2) Contextualizing student learning, 3) Developing multiple solutions, 3) Cognitive engagement in STEM, 5) Integrating STEM content, 6) Student agency, 7) Student collaboration, 8) Evidence-based reasoning, 9) Technology practices in STEM, and 10) STEM career awareness. In this study, we used the STEM-OP to explore current practices in integrated STEM education, including comparisons across science content areas (Physical, Earth, and Life Science) and grade levels (elementary, middle, and high school). Our data set included a total of 2,030 video-recorded classroom observations from K-12 science classrooms where integrated STEM teaching was enacted through the use of an engineering design challenge to learn and apply science and mathematics content. Our results suggest that current K-12 science teachers miss opportunities in their lessons to relate content to students' lives, develop multiple solutions, use evidence-based reasoning, engage students in technology practices, and promote STEM career awareness. However, results from crosstab and non-parametric analyses reveal that various components of integrated STEM education occur more frequently and at higher levels in Physical Science and elementary classrooms compared to Life/Earth Science and middle/high school classrooms, respectively. Our work illustrates various places where integrated STEM education could be focused, including a better representation of engineering through developing multiple solutions and using evidence-based reasoning. Our work also highlights the importance of providing K-12 teachers with more opportunities to engage in professional development related to integrated STEM education. Implications for this work include those for K-12 teachers, teacher educators, classroom coaches, and administrators.