Lee Ong Kim
National Institute of Education, Nanyang Technological University, Singapore.
J Appl Meas. 2003;4(1):10-23.
The longitudinal measurement of ability growth requires that the measures that are taken at the various time points be obtained using the same yardstick. Tests given for this purpose, therefore, need to be equated. The popular practice still in use for the purpose of equating tests, is the grade-equivalent. This paper compares the observation of children's growth in reading using grade-equivalents with that using Rasch Simultaneous Vertical Equating procedure. It is found that grade equivalents differ much more between two different test forms compared to ability measures obtained using Rasch Simultaneous Vertical Equating. It is also found that the spread of students' grade-equivalents, increased over the years as they grow while the standard deviations of their Rasch measures remain relatively constant over the same period of time. Student responses to the ITBS Form 7 and the CPS90 and CPS 91 were used. A total of 5,623 students were tracked over eight years.
能力增长的纵向测量要求在不同时间点所采用的测量方法要使用同一标准。因此，为此目的所进行的测试需要进行等值处理。目前仍在用于测试等值处理的普遍做法是年级当量。本文比较了使用年级当量与使用拉施同时垂直等值程序来观察儿童阅读能力增长的情况。研究发现，与使用拉施同时垂直等值法获得的能力测量结果相比，两种不同测试形式之间的年级当量差异要大得多。还发现，随着学生逐年成长，他们的年级当量分布有所增加，而在同一时间段内，他们的拉施测量标准差则保持相对稳定。使用了学生对ITBS表格7以及CPS90和CPS 91的回答。在八年时间里对总共5623名学生进行了跟踪。
