| Theories of intelligence have been one of the most famous topic since the birth of them. With the emergence of theories, tests of intelligence are booming. Nevertheless, the existing tests in China are numbered, it is imperative to develop an native multi-dimensional intelligence scale aiming at preschool and primary school children as well as with favorable diagnostic capability.In view of this, the developing work of the Multidimensional Intelligence Diagnosis Scale for Children of China (MIDSC) began in 2006 and has lasted for 10 years. In order to make full use of it,the standardization of the MIDSC is necessary.Firstly, the present study modified the MIDSC (the third edition) and finalized it (the fourth edition). The modification of the MIDSC (the third edition) was based on the reanalysis of the data got in 2010, including analyzing the mean and standard deviation of raw data, reliability analysis, and validity analysis. According to the analysis results, the study made some changes to items, administering, scoring, materials, sequence of items, instructions, and the software that inputted data. The MIDSC (the fourth edition) adapts to the children aged from 4 to 11 years old and retains original test framework. Namely, the whole scale consists of verbal, practical and working memory subscales, with verbal subscale including picture vocabulary, words using, arithmetic, and understanding (supplementary subscale), practical subscale including drawing, picture classify, puzzle playing, and picture reasoning, as well as working memory subscale including words reciting, action imitation, and digit span.Secondly, the present study selected participants from six cities (Handan, Quanzhou, Shanghai, Chongqing Xi’an, Jinzhou) of China and obtained a total of 785valid data. Based on these data, item analysis, reliability analysis and validity analysis was performed to the MIDSC (the fourth edition). The item analysis included difficulty analysis and discrimination analysis. Reliability of the scale was examined using Cronbach’s alpha coefficient, split-half coefficient and test-retest reliability coefficient. Validity analysis of the scales included criterion validity and structure validity analysis. The results of item analysis showed that 1)the difficulty of items was appropriate,2)the index of discrimination was high, and the scores of items were significantly related to the scores of subtests. The reliability analysis presented the results that 1)the alpha coefficients ranged from 0.667 to 0.963 with a mean of 0.866, 2)the Guttman split-half coefficient ranged from 0.852 to 0.989 with a mean of 0.945, 3)the test-retest reliability coefficient was 0.815. In addition, the validity analysis indicated that 1)the correlations of other tests were between 0.579 to 0.732,2)the difference between the normal and exceptional children was significant,3)the raw scores of full scale became higher along with the growth of the age,4)the correlations between one subtest and the relative subscale were higher than the correlations between the subtest and the rest of subscales, the correlation coefficients between subscales were medium to low, and the correlations between subtests and full scale were significant,5)the fitness of indirect hierarchical model (CMIN/DF =3.271,CFI=0.942,RMSEA=0.054) and direct hierarchical model (CMIN/DF =2.530,CFI=0.969,RMSEA=0.044) was good. In a word, MIDSC (the fourth edition) has good reliability and validity,Finally, the present study developed a national norm for MIDSC. The norm contains 11 transformation tables matching raw scores with scale scores, and 4 transformation tables matching scale scores with intelligence quotient (IQ). |
PDF Full Text Request