论文部分内容阅读
我们运用教育测量学的原理和方法,对近年的高考试卷进行了信度、效度、题目的难度,区分度四个方面的分析。信度是反映测验的稳定性和可靠性指标,表明一个测验反映受测者稳定水平的程度。国外一般要求信度在0.90以上,常达0.95,我们分析的部分试卷的信度,只有少数几份,如79年的物理、数学,80年的化学,81年的英语,历史较高外,其他均不高。效度是反映测验的准确性和有效性的指标,表示对它所要测量的东西实际测得有多好。国外一般是以被录取的考生的大学一年级各科平均成绩作为入学考试有效性的标准,二者相关达0.4—0.7,我们分析了16所大学24个班资料发现相关值如此之低,特别有许多负值,在一定程度上说明高考分数不能有效预测学生在大学的学习成绩。
We use the principles and methods of educational measurement to analyze the four aspects of reliability, validity, difficulty of the topic, and degree of differentiation. Reliability is a measure of the stability and reliability of the test, indicating the degree to which a test reflects the stability of the testee. The general reliability of foreign requirements is above 0.90 and often reaches 0.95. The reliability of some of the papers we analyzed is only a few, such as 79 years of physics, mathematics, 80 years of chemistry, 81 years of English, and higher history. Others are not high. Validity is an indicator that reflects the accuracy and validity of a test and shows how good it actually measures what it is measuring. In foreign countries, the average scores of the first-year college students of the admitted candidates are used as the criteria for the validity of the entrance examination. The correlation between the two is 0.4-0.7. We analyzed the data of 24 classes of 16 universities and found that the relevant values are so low. There are many negative values, to a certain extent, that the college entrance examination scores can not effectively predict students’ academic performance in college.