van Zyl Llewellyn E, Olckers Chantal, Roll Lara C
Department of Industrial Engineering, University of Eindhoven, Eindhoven, Netherlands.
Optentia Research Focus Area, North-West University (VTC), Vanderbijlpark, South Africa.
Front Psychol. 2020 May 7;11:796. doi: 10.3389/fpsyg.2020.00796. eCollection 2020.
The purpose of this study was to examine the psychometric properties (i.e., factorial validity, measurement invariance, and reliability) of the Grit-Original scale (Grit-O) within the Netherlands. The Grit-O scale was subjected to a competing measurement modeling strategy that sequentially compared both independent cluster model confirmatory factor analytical- and exploratory structural equation modeling approaches. The results showed that both a two first order, bi-factor structure as well as a less restrictive two factor ESEM factorial structure best-fitted the data. The instrument showed to be reliable at both a lower- (Cronbach's alpha) and upper-level (composite reliability) limit. However, measurement invariance between genders could only be established for the B-ICM-CFA model. Finally, concurrent validity was established through relating the GRIT-O to task performance. The linear use of the Grit-O scale should therefore carefully be considered.
本研究的目的是检验荷兰境内坚毅原始量表（Grit-O）的心理测量特性（即因子效度、测量不变性和信度）。Grit-O量表采用了一种竞争性测量建模策略，该策略依次比较了独立聚类模型验证性因子分析和探索性结构方程建模方法。结果表明，一个二阶双因子结构以及一个限制较少的双因子ESEM因子结构最适合这些数据。该工具在较低水平（克朗巴哈α系数）和较高水平（组合信度）上均显示出可靠性。然而，仅能为B-ICM-CFA模型建立性别间的测量不变性。最后，通过将GRIT-O与任务绩效相关联来建立同时效度。因此，应谨慎考虑Grit-O量表的线性使用。
