Scholastic Speech Aptitude Test For 3～4 Grades

As the Chinese education transforming from examination-oriented to qualification-oriented education, it is quite urgent to have a forward-looking objective supplementary measuring pattern to find out the students' capacities, so that we can teach them in accordance with their aptitude and give full play to the students' unique aptitude. However, even though researches on achievement test have achieved great success in the Chinese education field, but there is few research touching upon predictability tests.To estimate the pupils' basic ability to study Chinese, the speech aptitude test on primary school pupils which is studied in this research mainly determined subtests and each program according to the ability to understand and to use the Chinese language, to read relatively more difficult sentences and paragraphs, as well as to comprehend main context and deep meaning. From the test results, the subtests obviously have closer relationships with the total score than with each other. That means they all contribute to the total score, and there is high consistency between each subtest and the orientation of the overall test.To enable the test to distinct the difference among test targets in the special quality of being tested, this research mainly analyze the programs according to their degree of difficulty and discrimination. The results show that, generally speaking, most of the subtests' degree of difficulty distribution is quite satisfying. The tests which have a degree of difficulty of 0.30 to 0.69 take up 63.4 percent of the total program. The proportion of each degree of difficulty is quite suitable too, each subtest with an average degree of difficulty of about 0.5. Considering the situation of different grades, the degree of difficulty is relatively high in the grade three, with a pass rate lower than 0.5. The pass rate of grade four is about 0.5 and grade five higher than 0.6.From the perspective of discrimination index, the proportion of the programs in each subtest of each grade which has effective degree of discrimination accounts for more than 91%. On the whole, the proportion of the programs which the discrimination indexes are greater than 0.2, isabout 94%. Each subtest's average discrimination index is from 0.42 to 0.66. From the relevant of the program and the total score, the programs which the correlation coefficient is above 0.2 is above 90% , the correlation coefficient of each subtest program and total score is from 0.31 to 0.45.This research investigates the reliability of the test from the aspects of test-retest reliability, test-retest equivalent forms reliability, and homogeneity a coefficient. The results indicate that, the test-retest reliability of the total score are all above 0.77. While the test-retest equivalent forms reliability of the total seal is 0.44,and the a coefficient of total score are all 0.92 .From the perspective of validity research, the results show: (1) construct validity. The differences of each subtest and the total score of each grade are prominent, the test score increases with the grade and shows a tendency to rise steadily;the school students in both urban and city have a prominent difference in statistic on the whole. The differences of each subtest and the total score between men and women are not prominent. (2) criterion-related validity. Both Volume A and B have a prominent relevant with the score of Chinese General Examination. It is proved that this test can predict the score of Chinese study immensely and also has a better validity of prediction.The concurrent correlations with basic skill test from 0.40 to 0.77, and were significant.