中国幼儿智力量表的编制Ⅰ.编制策略、项目筛选、信度和常模

Chinese Intelligence Scale for Young Children (CISYC):Item Analysis, Reliability, and National Norms

目的：编制一套具有中国特色、适用于３～６岁半儿童的智力量表。方法：根据１９９０年全国第４次人口普查资料分层按比例取样，分别建立城市（ｎ＝１７２８）和农村（ｎ＝１５９５）两套年龄常模。结果：常模样本基本反映全国六大区人口结构，儿童性别、儿童父亲的受教育程度和职业分布与人口资料吻合，具有较好的代表性；条目分析表明测验条目对城市儿童适中，相对农村儿童稍偏难，但均具有较好的区分度；信度研究表明各分测验、成份分和智商的信度均符合测量学要求。结论：中国幼儿智力量表基本达到了设计目的和预期水平。

Objective: To develop an intelligence scale for Chinese children aged 3 to 6.5 year old.Method: CISYC consisted of Verbal- Comprehension/Generalization and Space- Cognition/Reasoning two subscales which each included four subtests; another two subtests in the scale were additional subtests. Two national norms had been established for urban (n=1 728) and rural (n=1 595) children, and the individuals of both standardization samples were stratified according to the 1990 census variables of age, sex, region, education and paternal occupation.Results:Item analysis showed the average difficulties across 9 subtests were 0.53 for urban and 0.43 for rural samples, the indexes of item discrimination were from 0.25 to 0.44 in urban children and 0.20 to 0.47 in rural children, depending on different subtests. The reliability coefficients of subtest were 0.57 to 0.93, and 75.0% of them were above 0.80. The values of reliability coefficient were 0.89 to 0.95 for subscales and 0.94 to 0.96 for IQ. The stability coefficients ranged from 0.53 to 0.84 for subtests and 0.82 to 0.87 for subscales and IQ. Conclusion:The techinical features of norms and reliability accord with the standards for educational and psychological testing.