Validating the "Chinese character difficulty scale" for primary students using rasch measurement

This study is a further refinement of an earlier paper on the development of Chinese Character Difficulty Scale. While the earlier study is concerned with the development of the scale, the current study provides further empirical evidence on the validity of the scale using Rasch measurement model on an expanded data set. In Chinese language development, characters serve as the basic building blocks in sentence construction and in reading comprehension for pupils aged from 6 to 9. Consequently, measures on Chinese character knowledge can be used as a proxy for the language ability of young children. The sample for this study comprised about 2000 students from 7 schools in primary 1 to 3. Thirty-six versions of test booklets were designed to provide compre?hensive coverage of all characters in the Revised List. Each booklet was made up of a list of 130 to 132 items (char?acters), with 50% linkage to the next booklet in the series. Rasch analysis showed that the scale has strong validity for use with Primary students. Factors contributing to difficulty level of characters include number of strokes, character form, and level of abstractness carried by the character. There was no difference between gender in their ability but there was significant year-level differences.