Renner Gerolf, Schroeder Anne, Irblich Dieter
Faculty of Special Education, Ludwigsburg University of Education, 71634 Ludwigsburg, Germany.
Werner Otto Institute, 22337 Hamburg, Germany.
J Intell. 2023 Jul 22;11(7):148. doi: 10.3390/jintelligence11070148.
Multidimensional intelligence test batteries such as the KABC-II are widely used in clinical practice. Although validity evidence should be provided for all intended uses of a test, data on the factorial validity of the KABC-II mostly relies on the standardization samples and raises some concerns about the adequacy of the factor structure. Confirmatory factor analyses of the KABC-II core subtests were conducted in a sample of 627 children who had been assessed in German Centers for Social Pediatrics. The standard structure of the KABC-II was superior to unidimensional models but, as in previous research, evidenced cross-loadings and a high correlation between and . was more closely related to than to . A four-factorial structure combining subtests from and to form a common factor emerged as a better representation of the data. showed a secondary loading on . On average, most subtest variance was accounted for by the general factor. Models with bonus points for fast responses generally fitted worse than those without. Clinicians should be aware that and measure both visual and fluid abilities. Scales of the KABC-II should not be interpreted as dimensions independent of the general factor.
诸如考夫曼儿童成套评估测验第二版（KABC-II）之类的多维智力测验组合在临床实践中被广泛使用。尽管应该为测验的所有预期用途提供效度证据，但KABC-II的因素效度数据大多依赖于标准化样本，这引发了对因素结构充分性的一些担忧。对627名在德国社会儿科学中心接受评估的儿童样本进行了KABC-II核心子测验的验证性因素分析。KABC-II的标准结构优于单维模型，但与先前的研究一样，存在交叉负荷以及[具体变量1]和[具体变量2]之间的高相关性。[具体变量1]与[具体变量3]的关系比与[具体变量2]的关系更密切。一种将来自[具体部分1]和[具体部分2]的子测验组合形成一个共同因素的四因素结构，作为数据的更好表示形式出现。[具体变量4]在[具体变量5]上显示出次要负荷。平均而言，大多数子测验方差由一般因素解释。有快速反应加分的模型通常比没有加分的模型拟合得更差。临床医生应该意识到[具体变量1]和[具体变量2]同时测量视觉和流体能力。KABC-II的量表不应被解释为独立于一般因素的维度。
