University of Oregon, Eugene.
Am J Speech Lang Pathol. 2018 Nov 21;27(4):1363-1374. doi: 10.1044/2018_AJSLP-17-0221.
There is a lack of quick, reliable, and valid standardized reading comprehension assessments appropriate for postsecondary readers. We attempted to address this gap by designing Read, Understand, Learn, & Excel (RULE), a reading comprehension measure that employs sentence verification and recall tasks to assess reading comprehension. This article describes the exploratory study undertaken to construct RULE and then examines the preliminary concurrent validity and alternate form reliability of this measure.
The RULE measure was first developed by designing reading stimuli, test items for the sentence verification task, and directions for the recall test for 2 forms based on previous work (Griffiths, Sohlberg, Kirk, Fickas, & Biancarosa, 2016). Thirty undergraduate students who identified themselves as typical readers were administered the RULE measure as well as the Nelson-Denny Reading Test (Brown, Fishco, & Hanna, 1993). Students also completed questionnaires and participated in informal interviews to provide information regarding study and learning habits and academic background.
There was preliminary evidence of alternate form reliability between the sentence verification task sets of 2 chapters of RULE (r = .38, p < .05). Preliminary evidence for concurrent validity between RULE and the Nelson-Denny Reading Test was provided by correlation coefficients in the low to moderate range (.03-.38).
RULE design and preliminary findings of concurrent validity and alternate form reliability provide "proof of concept" for an ecologically valid testing format that assesses comprehension skills appropriate for the postsecondary level. Suggestions for strengthening validity and reliability of the tool are provided, and clinical contributions of RULE are discussed.
缺乏适用于高等教育读者的快速、可靠和有效的标准化阅读理解评估方法。我们试图通过设计阅读、理解、学习和卓越(RULE)来解决这一差距,这是一种使用句子验证和回忆任务来评估阅读理解的阅读理解衡量标准。本文描述了构建 RULE 所进行的探索性研究,然后检验了该衡量标准的初步并行有效性和备用形式可靠性。
RULE 衡量标准首先通过设计阅读刺激物、句子验证任务的测试项目以及基于先前工作的回忆测试的说明来开发(Griffiths、Sohlberg、Kirk、Fickas 和 Biancarosa,2016)。30 名自认为是典型读者的本科生接受了 RULE 衡量标准以及 Nelson-Denny 阅读测试(Brown、Fishco 和 Hanna,1993)的测试。学生还完成了问卷并参加了非正式访谈,提供了关于学习和学习习惯以及学术背景的信息。
RULE 的两个章节的句子验证任务集之间存在备用形式可靠性的初步证据(r =.38,p <.05)。RULE 与 Nelson-Denny 阅读测试之间的并行有效性的初步证据是通过低到中等范围的相关系数提供的(.03-.38)。
RULE 的设计以及并行有效性和备用形式可靠性的初步发现为评估适合高等教育水平的理解技能的生态有效测试格式提供了“概念验证”。提供了加强工具有效性和可靠性的建议,并讨论了 RULE 的临床贡献。