| Summary: | 碩士 === 國立臺中教育大學 === 教育測驗統計研究所 === 101 === Large-scale assessment is one of the important approaches that the government used to evaluate the effect of policy of education. The purpose is to monitor population progress. Therefore, population statistics are what the large-scale assessment focus on. Plausible value method is proposed to be a great method that measures population statistics accurately so it is used to provide students’ achievement data by some significant large-scale assessment programs, nowadays. As the NCLB legislation calls, the issues of vertical scaling gets more concerned. Vertical scaling is the way test publishers used to longitudinally evaluate achievement that spans grade levels. For the test structure getting more complicate, researchers have agreed that multidimensional traits come into plays in test. Also, when calibrating items from multiple test forms for the purpose of measuring students across a range of grade levels, the IRT assumption of unidimensionality would appear to be impossible.
For the foregoing points, this research is aimed to analysis if: (1) the number of dimensions of the test, (2) the number of item for each dimension, and (3) the method that used to estimate parameters whether or not impact on the recovery of ability parameters of subjects and the recovery of population statistics, based on multidimensional IRT with the vertical scaling design.
The result indicates that plausible value method recovers the standard deviation very well but not outstands in recovering the population means. When using MIRT vertical design, parameters are estimated better when the number of items is more.
|
|---|
