Age-Related Invariance of Abilities Measured With the Wechsler Adult Intelligence Scale-IV
Age-Related Invariance of Abilities Measured With the Wechsler Adult Intelligence Scale-IV
复制标题
DOI：
10.1037/pas0000290
复制
发表时间：
2016-11-01
影响因子：
3.6
通讯作者：
Weiss, Lawrence G.
中科院分区：
文献类型：
--
作者：
Sudarshan, Navaneetham J.;Bowden, Stephen C.;Weiss, Lawrence G.
人群间测量不变性的评估对于有意义的考试成绩比较至关重要，尤其是在教育评估或临床诊断需要重复测量的情况下。建立测量不变性的假设，即测试分数反映了相同的心理特质，在不同的人群或在不同的场合。韦氏成人智力量表第四版的检验。S.标准化样本显示，一阶5因素测量模型最适合16岁至69岁的9个年龄组。强度量不变性被发现为3的5个因素和部分截距不变性的其余2。相邻年龄组的成对比较支持这样的推论，即认知特质组的差异表现为测试成绩的组间差异。在教育和临床环境中，这些研究结果提供了理论和实证支持，以解释指数或子测试分数的变化，反映了相应的认知能力的变化。此外，在临床相关的情况下，可以使用子测试评分组合来比较各自认知能力的变化。该模型得到了加拿大标准化数据的支持，合并年龄组，但样本量不足以详细检查加拿大样本中的单独年龄组。
Assessment of measurement invariance across populations is essential for meaningful comparison of test scores, and is especially relevant where repeated measurements are required for educational assessment or clinical diagnosis. Establishing measurement invariance legitimizes the assumption that test scores reflect the same psychological trait in different populations or across different occasions. Examination of Wechsler Adult Intelligence Scale-Fourth Edition (WAIS-IV) U. S. standardization samples revealed that a first-order 5-factor measurement model was best fitting across 9 age groups from 16 years to 69 years. Strong metric invariance was found for 3 of 5 factors and partial intercept invariance for the remaining 2. Pairwise comparisons of adjacent age groups supported the inference that cognitive-trait group differences are manifested by group differences in the test scores. In educational and clinical settings these findings provide theoretical and empirical support to interpret changes in the index or subtest scores as reflecting changes in the corresponding cognitive abilities. Further, where clinically relevant, the subtest score composites can be used to compare changes in respective cognitive abilities. The model was supported in the Canadian standardization data with pooled age groups but the sample sizes were not adequate for detailed examination of separate age groups in the Canadian sample.