Validating a measure of growing pattern understanding in Chinese preschool children

Abstract

Recent research has pointed to the importance of patterning skills in early mathematics development. However, there is a lack of psychometrically robust instruments to measure patterning skills, especially growing pattern understanding. In this study, we developed and validated an instrument assessing growing pattern understanding among preschool children. One hundred and thirty-eight preschoolers aged approximately five years in Hong Kong were assessed individually on 10 growing pattern items at two time points over a one-year period. Confirmatory factor analysis showed that a two-factor model (i.e., increasing and decreasing patterns) fit the data best in the second year of preschool (PS2), but a one-factor model (i.e., growing patterns) fit the data best in the third year (PS3). Scale reliability was adequate at both time points. At the item level, Rasch analysis showed that all the items were within an acceptable range of difficulty at both time points. Information curves also suggested that the test items provided the most information (i.e., having high reliability) at high ability levels at PS2 but were a sensitive measure for children of both low and high abilities at PS3. No differential item functioning was detected across boys and girls at PS2 or PS3. Also, expected relations were found between growing pattern understanding and children's performance on repeating patterns and English expressive vocabulary, supporting convergent and discriminant validity. Overall, the results suggest that the growing pattern task has sound psychometric properties and is appropriate for use in preschool children aged five and six years.