Bazarova Saodat I, Engelhard George
Division of Educational Studies, Emory University, 1784 North Decatur Rd., Suite 240, Atlanta, GA 30022, USA.
J Appl Meas. 2004;5(2):142-59.
Using the Mantel-Haenszel (MH) Procedure, we analyzed data for 7,087 American and 4,022 Russian Grade 8 students from the Third International Mathematics and Science Study (TIMSS) to compare mathematics achievement in the two countries on each of the 124 multiple-choice items. The results of the analyses indicate that the performance of the students on individual multiple-choice mathematics items vary by country. The results also suggest that the relationship between country and item performance differ as a function of content area. A total score of a country's achievement does not provide the whole picture of achievement dynamics; it averages out potentially important information on student achievement and the causes of their performance relative to other countries. The dynamics of achievement across countries will not be revealed unless the analyses are done at the item level.
我们使用曼特尔 - 亨泽尔（MH）方法，对来自第三次国际数学和科学研究（TIMSS）的7087名美国八年级学生和4022名俄罗斯八年级学生的数据进行了分析，以比较两国学生在124道多项选择题上的数学成绩。分析结果表明，学生在各个数学多项选择题上的表现因国家而异。结果还表明，国家与题目表现之间的关系因内容领域而异。一个国家的总成绩并不能全面反映成绩动态；它平均了有关学生成绩以及他们相对于其他国家的表现原因的潜在重要信息。除非在题目层面进行分析，否则各国之间的成绩动态将无法显现。
