Psychol Assess. 2018 Aug;30(8):1038. doi: 10.1037/pas0000652.
Reports an error in "Revisiting Carroll's survey of factor-analytic studies: Implications for the clinical assessment of intelligence" by Nicholas F. Benson, A. Alexander Beaujean, Ryan J. McGill and Stefan C. Dombrowski (, Advanced Online Publication, May 24, 2018, np). In the article "Revisiting Carroll's Survey of Factor-Analytic Studies: Implications for the Clinical Assessment of Intelligence," by Nicholas F. Benson, A. Alexander Beaujean, Ryan J. McGill, and Stefan C. Dombrowski (, Advance online publication, May 24, 2018, http://dx.doi.org/10.1037/pas0000556), the majority of values in the ωH and ωHS columns of Table 4 were incorrect and have been amended. These revisions required text in the fourth paragraph of the Results section to be changed from "Moreover, the ωHS value for is relatively high and very close to the and ωH values for g" to "Moreover, the ωHS values for and are relatively high, exceeding the ω and ωH values for g." All versions of this article have been corrected. (The following abstract of the original article appeared in record 2018-23627-001.) John Carroll's three-stratum theory (and the decades of research behind its development) is foundational to the contemporary practice of intellectual assessment. The present study addresses some limitations of Carroll's work: specification, reproducibility with more modern methods, and interpretive relevance. We reanalyzed select data sets from Carroll's survey of factor analytic studies using confirmatory factor analysis as well as modern indices of interpretive relevance. For the majority of data sets, we found that Carroll likely extracted too many factors representing Stratum II abilities. Moreover, almost all factors representing Stratum II abilities had little-to-no interpretive relevance above and beyond that of general intelligence. We conclude by discussing the implications of this research with respect to the interpretive relevance and clinical utility of scores reflecting cognitive abilities at all strata of the three-stratum theory and offer some directions for future research. (PsycINFO Database Record
报告尼古拉斯·F·本森、A·亚历山大·博让、瑞安·J·麦吉尔和斯特凡·C·多布罗夫斯基所著的《重温卡罗尔的因素分析研究综述：对智力临床评估的启示》（《，高级在线出版物，2018年5月24日，np）中的一处错误。在尼古拉斯·F·本森、A·亚历山大·博让、瑞安·J·麦吉尔和斯特凡·C·多布罗夫斯基所著的《重温卡罗尔的因素分析研究综述：对智力临床评估的启示》（《，高级在线出版物，2018年5月24日，http://dx.doi.org/10.1037/pas0000556》）一文中，表4的ωH和ωHS列中的大多数值有误，现已修正。这些修订要求结果部分第四段中的文本从“此外， 的ωHS值相对较高，非常接近g的 和ωH值”改为“此外， 和 的ωHS值相对较高，超过了g的ω和ωH值”。本文的所有版本均已更正。（原始文章的以下摘要出现在记录2018 - 23627 - 001中。）约翰·卡罗尔的三层次理论（及其发展背后数十年的研究）是当代智力评估实践的基础。本研究探讨了卡罗尔研究工作的一些局限性：规范、用更现代的方法进行可重复性研究以及解释相关性。我们使用验证性因素分析以及现代解释相关性指标重新分析了卡罗尔因素分析研究综述中的选定数据集。对于大多数数据集，我们发现卡罗尔可能提取了过多代表第二层能力的因素。此外，几乎所有代表第二层能力的因素在一般智力之外几乎没有解释相关性。我们通过讨论这项研究对于反映三层次理论各层次认知能力的分数的解释相关性和临床效用的影响来得出结论，并为未来研究提供一些方向。（PsycINFO数据库记录
