An evaluation of children's structural drawing strategies

Abstract

Inspecting children's structural drawing is developmentally and psychologically important. Today's digital availability of such data from electronic tablets, inspires automatic analysis; however, converting such data to an informative feature vector for further analysis and identification of related indicators, needs appropriate algorithms. This study presents simple, fast methods for detecting X, O and 4 basic lines' drawing strategies. The functionality of the algorithms is tested on an available database subtending the performances of 74 (6-7years) pupils. Results demonstrate typical behaviors such as counterclockwise popularity in circle drawing, and other tendencies in the population's structural pattern drawing's performance. Algorithms clearly reveal all different strategies used by participators. Consequently the suitability of the algorithms, in effectively detecting strategic features, is clarified. In addition, according to the data labels in database, some features are suggested to serve as attributes distinguishing handwriting ability. © 2012 Penerbit UTM Press. All rights reserved.