The three decades of teacher efficacy research have been characterized by debate regarding the conceptualization and measurement of the construct. Teachers' sense of efficacy refers broadly to a belief in one's ability to influence valued student outcomes (Wheatley, 2005). Initially assessed using two Likert scaled items, teaching efficacy during the research of the 1980s was dominated by Gibson and Dembo's (1984) Teaching Efficacy Scale, which focused on a teacher's belief in their ability to bring about positive student outcomes, and beliefs about the likely outcomes of teaching. In the late 1980s and 90s, this model was criticized for its characterization of teacher efficacy as a global construct, and failure to couch efficacy in the specific contexts of teaching and tasks involved in the process. Recent efforts have begun to examine efficacy beliefs in specific subject areas and for specified tasks and outcomes. Two examples are the Teaching Confidence Scale and Bandura's Teacher Efficacy Scale (Woolfolk-Hoy & Spero, 2005), which focus on aspects of instruction such as classroom management and creating a positive climate. To fill the void for a multi-faceted physical education teaching efficacy scale, we initiated this project. The Physical Education Teaching Efficacy Scale (PETES) was grounded on the teaching efficacy literature, existing scales, and recommendations for creating such instruments. In developing items, we used the NASPE (2001) Standards for Initial Programs in Physical Education Teacher Education and the teaching effectiveness literature as a framework. Items were written to address each of the 10 NASPE standards using wording that paralleled the standard and its outcomes, and also provided context for efficacy judgments. The initial items were evaluated in two steps by three experts in physical education teacher education and a sample of preservice teachers. The final instrument was composed of 80 efficacy items, plus those gathering demographic information. Physical education teacher education majors (N = 497) at 10 institutions completed the survey during the 2006-07 school year. Factor analysis resulted in inclusion of 59 items and 7 factors, which accounted for 61.59% of the cumulative variance and reliability coefficients (Cronbach's alpha) ranging from 0 .79 to 0.96. Factors focused on efficacy about content knowledge, applying scientific knowledge in teaching PE, accommodating skill level differences, teaching students with special needs, instruction (including management, motivation, and communication), using technology, and assessment. Our results suggest this instrument addresses many aspects of teaching physical education and meets research criteria for validity and reliability.Keyword(s): measurement/evaluation, professional preparation, research