Innovations in Science, Technology, Engineering, and Mathematics (STEM) are a cornerstone of the United States' economic prosperity, yet national and international assessments paint a poor picture of students' STEM competencies (reference 1; reference 2). These data have led to transformative approaches to improve STEM education including research-informed science standards (reference 3) and recommendations to maximize science teachers' instructional effectiveness (reference 4). One area that holds particular promise for improving STEM education is "spatial thinking" (reference 5). Over 60 years of longitudinal research has suggested that spatial skills are essential to success in STEM (reference 6) and intervention work indicates that developing students' spatial skills improves STEM outcomes (references 7; 8; 9). These findings have motivated efforts to infuse spatial skills into classroom-based interventions (references 10; 11). While such efforts aim to develop students' spatial skills, teachers are the key drivers of intervention effectiveness, and their beliefs and feelings can influence intervention fidelity (references 12; 13) and student outcomes (references 14; 15; 16). In order to develop effective classroom interventions to foster students' spatial skills, we must understand teachers' spatial thinking beliefs and perceptions. Here, we report the first attempt to probe elementary teachers' perceptions of their own spatial skills, beliefs about the importance of these skills for solving STEM problems and academic success, and self-efficacy in cultivating students' spatial skills during science instruction. There are, however, no extant measures to probe these questions. Therefore, we aimed to (a) develop such measures and examine their psychometric properties, (b) describe teachers' beliefs and perceptions related to spatial thinking in STEM and self-efficacy in cultivating students' spatial skills during science, (c) examine similarities and differences in beliefs and perceptions based on teacher characteristics, and (d) explore relations between general and science teaching self-efficacy, and teachers' spatial thinking self-efficacy. We administered five measures to 104 2nd, 4th, and 5th grade teachers drawn from a large, urban district outside of Washington DC: 1. "Teachers' Sense of Efficacy Scale Short Form" (TSES; reference 17) which probes efficacy in student engagement, instructional strategies, and classroom management on a 9-point scale ranging from "none at all (1)" to a "great deal (9)." 2. "Science Teaching Efficacy Belief Instrument Personal Scale" (STEBI; reference 18), which probes teachers' personal science teaching efficacy. Statements are rated on a 5-pt scale from "strongly disagree" to "strongly agree." 3. "Spatial Thinking Confidence and Anxiety Instrument" (developed for this study), which probes teachers' anxiety and confidence in solving spatial problems. All questions were rated on a five-point scale ranging from "Not at all" (0) to "Extremely" (4) for both anxiety and confidence. 4. "Beliefs about the Importance of Spatial Thinking for Solving Problems" (developed for this study), which probes the degree to which teachers feel spatial thinking is important for solving STEM and non-STEM problems that span K-16 content areas. Problems varied in terms of their reliance on spatial thinking (either high-reliance, medium-reliance, or low-reliance). Importance was rated on a 5-pt scale from not important (0) to extremely important (4). 5. "Spatial Thinking in Science Self-Efficacy" (developed for this study), which probes teachers' self-efficacy in cultivating and assessing students' spatial thinking skills during a variety of aspects of science instruction (e.g., observing students, planning a lesson, differentiating instruction). All items were scored on a 5-pt scale ranging from not at all well (0) to extremely well (4). We explored the structure of these measures and examined (a) how anxious and confident teachers were when solving spatial problems, (b) how important teachers believe spatial thinking skills are for solving academic problems, c) how well teachers feel they can cultivate students' spatial skills during science instruction, and (4) how spatial thinking self-efficacy during science instruction relates to general and science teaching efficacy? Most teachers in our sample identified as female (84.6%). As seen in Table 1, the plurality identified as African American/Black (44.2%) but there was a range of ethnicities and ages. Of the sample, 25.3% taught 2 grade, 34.1% taught 4 grade, and 40.7% taught 5 grade and 57.7% reported their highest degree in education was a master's degree. On average teachers had 11.8 years of experience (SD = 8.53 years; range 1-34 years). Approximately half of teachers (53.8%) worked in a Title I school, and more than half (56.7%) were specialists. Of the specialists, 44.1% taught math and 94.9% taught science. th All measures exhibit high internal consistency (see Table 2). Results showed teachers experienced low anxiety when solving spatial problems (M = 1.25, SD = 0.88) and felt confident that their skills can improve with practice (M = 2.94, SD = 1.06). Teachers were able to identify educational problems that rely on spatial problem-solving (indicating spatial skills were important for solving these problems) and believed that spatial skills are more important for older compared to younger students. Despite reporting high efficacy in their general teaching and science teaching (see Table 3), teachers reported significantly lower efficacy in cultivating students' spatial skills during science instruction (see Table 3). Table 3 illustrates the correlations between Spatial Thinking in Science Self-Efficacy and the TSES and the STEBI. Results indicate that these scales are related but that self-efficacy in cultivating spatial skills during science is distinct from general self-efficacy and science teaching self-efficacy. Results were fairly consistent across teacher characteristics (years of experience and teaching role as generalist or specialist) with the exception that only years of teaching science was related to teachers' efficacy in cultivating students' spatial thinking skills during science instruction. Together, these findings indicate elementary teachers believe these skills are important and malleable with practice and suggest targeted professional learning opportunities for cultivating teachers' capacity to develop students' skills through classroom-based instruction. Results will be discussed within the broader context of teacher beliefs, self-efficacy, and implications for professional development research and programs.