P. R. Parrish, R. A. Stodden. Aligning assessment and instruction with state standards for children with significant disabilities. Teaching Exceptional Children, 2009: 46-56.
Rasch Validation of the Inclusive Curriculum-Based SCALE Instrument in Hong Kong
曾君兰, 谢宗义, 李启明
课程本位评量, 智障学生, 等级量尺分析
Curriculum-Based Assessment; Pupils with Intellectual Disabilities; Rasch Analysis
《Advances in Education》, Vol.3 No.2, 2013-04-30
背景：过去的研究结果显示，需要为有特殊教育需要的学生发展一些具实用性的学与教套件，包括相关的评估、课程与教学策略，藉此促进全纳教育的发展。为响应此号召，香港9所特殊学校的教师团队为智障生研发了一套以连接特殊学校与主流学校的课程的评估工具，称为“融通学习成效量表(SCALE)”。目的：本研究目标，在于以等级量尺模式检视融通量表的信效度，验证近期研发的融通量表在评估具特殊教育需要的学生于香港主流课程中的四个主要学习领域的学习能力表现水平时是否达到其研发目的。方法：本研究采用了等级量尺模式来检视融通量表的信效度。依据受试者个人能力差异的逻辑函数判断答案正误的机率，检测使用融通量表之观察数据与等级量尺模式的预期数据匹配的适合程度、点数测量相关系数、个人/题项信度及主成因素构念向度分析。结果：从统计数据显示出融通量表拥有极高信效度，能将学生分为13个不同的学习能力水平级别，且题项难度范围足够将学生能力分为8~11个学习能力表现水平级别。所有的题项分数都与预期总分具非常高的相关性，证明融通量表具单一构念向度，也就是说，融通量表仅仅测量了学生的学习能力表现水平，而不是测量他们学习的其他方面。结论：从评估学习的角度，本文进一步讨论课程评估测量法的使用如何在绩效责任上产生作用；以及对于协助教师指导学习能力水平各有差异的学生的影响Background: Past research has reflected the needs to develop measures in the area of assessment, curriculum and teaching strategies for pupils with special educational needs. The SCALE, an assessment instrument to align assessment with the central curriculum for pupils with intellectual disabilities was developed by a team of teacher leaders from nine special schools in Hong Kong. Aim: The objectives of this study are to critically evaluate the validity and reliability of the SCALE attainment scales for each strand in four key learning areas (KLA) of the central curriculum of Hong Kong using Rasch analysis. Method: The SCALE data was calibrated using the Rasch measurement model. By calculating the probability of a response by right or wrong answers in terms of a logistic function of the difference between the ability of the person taking the test, Rasch analyses on the assessment data were carried out in the form of model fit statistics, point-measure correlation coefficient, person / item reliability and principal component factor analysis on dimensionality. Results: The Rasch statistics indicate that SCALE possesses a high degree of validity and reliability. It can classify pupils into at least 13 levels of abilities, and that the range of item difficulty is capable of stratifying pupils’ abilities into at least 8 to 11 attainment levels. The result also supports the uni-dimensionality of the SCALE. That is, SCALE solely measures the attainment levels of the pupils but not other aspects of their learning. Conclusion: The use and the impact of SCALE in providing evidenced-based outcome data to document progress of pupils across their years in education, and, to inform school-based curriculum planning are further discussed.