Francis D J, Fletcher J M, Rourke B P
Department of Psychology, University of Houston, TX 77204-5341.
J Clin Exp Neuropsychol. 1988 Dec;10(6):779-99. doi: 10.1080/01688638808402814.
The discriminant validity of left- (LH) and right-hand (RH) sensorimotor measures from a comprehensive neuropsychological battery was tested using confirmatory factor analysis. A group of children (primarily learning disabled) was divided into analysis (n = 488) and cross-validation (n = 400) samples and the following measures were taken for both the LH and the RH Tactile Perception, Finger Agnosia, Fingertip Number Writing, Tactile Form Recognition, Finger Tapping, Grip Strength, Grooved Pegboard, and Mazes. RH- and LH scores from these tests, and scores from five WISC subtests (Information, Similarities, Vocabulary, Block Design, and Object Assembly) were analyzed in a series of four nested confirmatory factor models. Models distinguished between LH and RH skill factors, and/or between simple- and complex-skill factors. Models were compared using incremental fit ratios and chi 2 difference tests (Bentler & Bonnett, 1980). Model comparisons revealed little evidence for discriminant validity of LH and RH measures, but strong support for distinctions between measures of simple- and complex-skills. These findings were replicated in the cross-validation sample.
使用验证性因素分析对来自综合神经心理测验组的左手（LH）和右手（RH）感觉运动测量的区分效度进行了测试。一组儿童（主要是学习障碍儿童）被分为分析样本（n = 488）和交叉验证样本（n = 400），并对LH和RH进行了以下测量：触觉感知、手指失认、指尖数字书写、触觉形状识别、手指敲击、握力、有槽钉板和迷宫测试。对这些测试的RH和LH分数，以及五个韦氏儿童智力量表子测验（知识、相似性、词汇、积木图案和图形拼凑）的分数，在一系列四个嵌套的验证性因素模型中进行了分析。模型区分了LH和RH技能因素，和/或简单技能与复杂技能因素。使用增量拟合比和卡方差异检验（Bentler & Bonnett，1980）对模型进行比较。模型比较显示，几乎没有证据支持LH和RH测量的区分效度，但有力支持了简单技能与复杂技能测量之间的区别。这些发现也在交叉验证样本中得到了重复。
