LARGE‐SCALE PREDICTOR VALIDATION IN PROJECT A: DATA COLLECTION PROCEDURES AND DATA BASE PREPARATION
描述了陆军选拔与分类项目（项目A）中同步验证样本的数据库，包括抽样程序、各岗位样本量分布、变量阵列及数据收集与缺失数据处理方法。
The data base for the Army Selection and Classification Project (Project A) contains two major samples referred to as the concurrent validation sample and the longitudinal validation sample. The former was drawn from a cohort that joined the Army in 1983/84, and the latter from a cohort that entered in 1986/87. This paper describes the data base resulting from the concurrent sample. The sampling procedure, the distribution of sample sizes over jobs, the total array of variables, and the data collection procedures are described. Also discussed are the extensive data editing procedures that were used to deal with missing data.