Rusche Marianna Massimilla, Ziegler Matthias
Psychology Department, Faculty of Life Science, Humboldt Universität zu Berlin, 10117 Berlin, Germany.
J Intell. 2023 Feb 28;11(3):47. doi: 10.3390/jintelligence11030047.
Along with crystallized intelligence (Gc), domain-specific knowledge (Gkn) is an important ability within the nomological net of acquired knowledge. Although Gkn has been shown to predict important life outcomes, only a few standardized tests measuring Gkn exist, especially for the adult population. Complicating things, Gkn tests from different cultural circles cannot simply be translated as they need to be culture specific. Hence, this study aimed to develop a Gkn test culturally sensitive to a German population and to provide initial evidence for the resulting scores' psychometric quality. Existing Gkn tests often mirror a school curriculum. We aimed to operationalize Gkn not solely based upon a typical curriculum to investigate a research question regarding the curriculum dependence of the resulting Gkn structure. A set of newly developed items from a broad range of knowledge categories was presented online to 1450 participants divided into a high (fluid intelligence, Gf) Gf ( = 415) and an unselected Gf subsample ( = 1035). Results support the notion of a hierarchical model comparable to the one curriculum-based tests scores have, with one factor at the top and three narrower factors below (Humanities, Science, Civics) for which each can be divided into smaller knowledge facets. Besides this initial evidence regarding structural validity, the scale scores' reliability estimates are reported, and criterion validity-related evidence based on a known-groups design is provided. Results indicate the psychometric quality of the scores and are discussed.
除了晶体智力(Gc)之外,特定领域知识(Gkn)是后天习得知识的法则网络中的一项重要能力。尽管Gkn已被证明能够预测重要的生活结果,但目前仅有少数标准化测试可用于测量Gkn,尤其是针对成年人群体。更复杂的是,来自不同文化圈的Gkn测试不能简单地进行翻译,因为它们需要具有文化特异性。因此,本研究旨在开发一种对德国人群体具有文化敏感性的Gkn测试,并为所得分数的心理测量质量提供初步证据。现有的Gkn测试往往反映学校课程。我们旨在不仅仅基于典型课程来对Gkn进行操作化,以研究一个关于所得Gkn结构对课程依赖性的研究问题。一组来自广泛知识类别的新开发项目在网上呈现给1450名参与者,这些参与者被分为高(流体智力,Gf)Gf组( = 415)和未经过筛选的Gf子样本组( = 1035)。结果支持了这样一种分层模型的概念,该模型与基于课程的测试分数所具有的模型类似,顶部有一个因素,下面有三个较窄的因素(人文、科学、公民学),每个因素又可进一步细分为更小的知识方面。除了关于结构效度的这一初步证据外,还报告了量表分数的可靠性估计,并提供了基于已知群体设计的与效标效度相关的证据。结果表明了分数的心理测量质量并进行了讨论。