This study is aimed at developing a tool to measuring school police's job performance, and verifying its credibility and validity. This study collected the sample to 250 subjects using a significance collection method to parents of elementary schools school police work. But excluding dishonestly responding data, this used total 210 people for the final analysis. This study executed experts' meeting, exploratory and confirmatory factor analysis, and item analysis to verify reliability, and verification of internal consistency and reliability. This study extracted and confirmed 7 factors for specialization, attitude, program, prevention activity of school violence, effect, basic duty, teaching for school time and home time, and correlation with related people. Through the above research approach and procedure, this study arrived at the following conclusion. First, validity to school police's job performance and measurement tool to job performance is very high. That is, school police's job performance was explored and confirmed through correlation with specialization, attitude, program, prevention activity of school violence, effect, basic job, teaching for school time and home time, and related people, and of them, explanation of attitude is the highest. Second, reliability to the measurement scale for school police's job performance is very high. That is, school police's job performance has high correlation between items, sub-factors, and total scores, and reliability of internal consistency, Cronbach's ${\alpha}$ value, is over .708. Putting them together, measurement scale for school police's job performance is a questionnaire sheet with very high level of validity and reliability.