国外应用G理论进行概念图教学评估的实验研究

Application of the generalizability theory to G-map teaching assessment

文章分析了Shavelson&Yue Yin等应用G理论进行概念图评估的实验研究。首先对G理论、概念图评估、C技术、S技术进行了界定;然后,阐释了Shavelson等运用G理论对C与S概念图技术的比较实验研究。结果表明C和S在技术上是不等价的,S技术的概化系数比C技术大;若达到预期的概化系数,单次测试下S技术需要的概念数少于C技术。因此,对于大规模的终结性评估,S技术相对于C技术或许是更好的选择;而C技术则可能更适合于课堂上的形成性评估。

This paper introduces the results of a study made by Yue Yin and Richard J. Shavelson on the application of generalizability theory to the C and S technique research. The first part of this paper is on the concepts of generalizability, C - map assessment, C and S techniques. Then the G theory is applied to the technical qualities of C and S. Result shows that C and S are not technically equivalent. The G coefficients for S are larger than those for C. Furthermore, a D study shows that fewer items （concepts） are needed for S than C to reach desired level of G coefficients. Therefore, S might be a better candidate than C in the large- scale formative assessment, while C is preferred as a formative assessment in classroom.