Perceptualand Motor Skills, 1990, 70, 759-770. @ Perceptual and Motor Skills 1990 DEVELOPMENT OF HANDWRITING IN PRIMARY SCHOOL: A LONGITUDINAL STUDY ' LISA HAMSTRA-BLETZ AND ANKE W. BLOTE Leiden University Summary.-To investigate the development of handwriting in primary school chil- dren a study was conducted in which the scripts of 127 pupils were rated yearly, starting in Grade 2. Children were followed-up on for 3, 4, or 5 yr. The BHK-scale for children's handwriting was used for rating the scripts on 13 characteristics and for measuring the speed of writing. The shifts in frequency of occurrence of the various characteristics over time were analyzed using Correspondence Analysis. A one-dimen- sional solution was chosen, yielding a time dimension. Results were interpreted in terms of (a) children's growing motor ability, (b) a deterioration of the form aspects of the script as they had uuc~allylearned those, and (c) the development of a personal writing style with regard to estheric aspects. A strong relation was found between the speed of writing and grade. A large number of scales for the evaluation of children's handwriting are available. Some of them, mostly the older ones, are holistic in nature, the rest are analytic scales (Harris, 1960). O n the holistic scales a global judge- ment of a script is given, e.g., how legible is it in comparison with a given standard. On the analytic scales various characteristics of the handwriting are rated individually. Next a sum score is calculated, which is either evalu- ated irrespective of age or grade (see, e.g., De Ajuriaguerra, 1979) or is rated using age or grade norms (see, e.g., Phelps, Stempel, & Speck, 1985). What is lacking is the kind of scale on which the script is rated with respect to a set of characteristics using individual age or grade norms for each feature and for which a sum score is calculated afterwards. In the latter case a given feature is evaluated in relation to the performance of other subjects of the same age or grade. Such an age-related differentiation seems sensible, as common observation shows that in the development of handwriting shifts between writing characteristics occur. What are needed, however, are empiri- cal data on these changes in children's writing. As yet very IittIe quantitative information is available. Furthermore, research has been of a cross-sectional nature. As a consequence no conclusions can be drawn about the writing development of individud subjects. Moreover, if the age-groups in these studies differ from each other on more variables than age alone, this can have some negative consequences for the reliability of the results. 'This study was supported in part b a grant from the Prawentiefonds, The Netherlands (Project 28.815). We thank Peter van A r Heijden for his critical comments. We are gateful to Ailsa Leadbetter for correcting the English text. Correspondence concerning this article should be addressed to Lisa Hamstra-Bletz, Department of Psychology, Section of Developmental and Educational Psychology, Leiden University, Wassenaarseweg 52, 2333 AK Leiden, The Netherlands.