The objective of this paper is to develop a brief screening instrument of posttraumatic stress disorder (PTSD) for young victims of natural disasters. Data were derived from flood victims in 1998 and 1999 in Hunan, China. A representative population sample of 6,852 subjects 7-15 years of age was selected. Among them, 6,073 (88.6%) were interviewed. Multistage sampling was used to select the subjects and PTSD was ascertained with Diagnostic and Statistical Manual of Mental Disorders: 4th Edition (DSM-IV). We randomly assigned 80% (4,851) of the study subjects to construct the screening instrument (construct model) and the remaining 20% (1,222) subjects to examine the model (validation model). Logistic regression analysis and receiver operating characteristics curves were utilized to select a subset of symptoms and cutoff point from the pre-structured questionnaires. A seven-symptom instrument for PTSD screening was selected. Scores of 3 or more on this instrument were employed to define positive cases of PTSD with a sensitivity of 96.9%, specificity 99.0%, positive predictive value (PPV) 82.6%, and negative predictive value (NPV) 99.8%. The brief screening instrument developed in this study is highly valid, reliable, and predictable.