Performance of school-aged children on a Chinese handwriting speed test

294 Occupational Therapy International, 4(4), 294 303, 1997 Whurr Publishers Ltd Performance of school-aged children on a Chinese handwriting speed test MEI HUI TSENG School of Occupational Therapy, National Taiwan University, Taiwan I-PING HSUEH School of Occupational Therapy, National Taiwan University ABSTRACT Slow handwriting speed is one of the major problems encountered by school-aged children having handwriting difficulties. These children are often referred to occupational therapists for assessment and intervention. However, to date, no norms of handwriting speed have been published for the Chinese writing system for school-aged children. The purpose of this study was to document handwriting speed performance of Chinese children from Grades 2 to 6. The results of this study will provide occupational therapists with an objective guideline when screening children who have handwriting speed problems. Participants were 1525 children from Grades 2 to 6 from three elementary schools in the Taipei area. Participants were asked to copy texts from a Chinese textbook that they had studied in the second semester of the first grade. The test lasted for five minutes. Results showed that handwriting speed increased with age; and the rate of increase was found to be greatest among Grades 2, 3, and 4. Furthermore, girls wrote faster than boys in Grades 3, 4, and 5. This study s baseline data on handwriting speed for the Chinese writing system provide further substantial information for future research and clinical practice. Key words: handwriting speed, Chinese writing system, school-aged children, occupational therapy. INTRODUCTION Handwriting has remained an important skill for school-aged children in spite of the introduction of word processors and augmentative communication systems (Søvik, 1975; Amundson & Weil, 1996). Mastering handwriting skills is one of the primary goals of elementary school education. A student s ability to write not only legibly but also at an efficient speed is

Chinese handwriting speed test 295 important for functional written communication as well as for educational development (Phelps, Stempel & Speck, 1985; Tseng & Cermak, 1993; Amundson & Weil, 1996). A study by McHale and Cermak (1992) examined the amount of time allocated to fine motor activities and the type of fine motor activities that school-aged children were expected to perform in the classroom. They found that 31 60% of the children s school day consisted of fine motor activities. Of these fine motor tasks, 85% of the time was employed in paper and pencil tasks. Based on the results of this study, Amundson and Weil (1996) pointed out that children spend a quarter to a half of their classroom time involved with paper and pencil tasks each day at school. In addition, the majority of these tasks have time constraints; therefore, an efficient writing speed is critical if the students are to accomplish an acceptable amount of work in the classroom and meet the standards of the teacher and the curriculum (Amundson & Weil, 1996). School-aged children who have handwriting difficulties are often referred to occupational therapists. Typical handwriting difficulties for these children are illegibility, inability to keep up with written class assignments, and the lack of automaticity of handwriting (Oliver, 1990; Cermak, 1991; Amundson & Weil, 1996). As to overall handwriting performance, speed is probably the simplest overall measure of proficiency in writing. At present, there are norms of handwriting speed only for the English writing system (Ayres, 1912; Freeman, 1954; Groff, 1961; Ziviani, 1984; Phelps et al., 1985; Phelps & Stempel, 1987; Larsen & Hammill, 1989). No norms of handwriting speed to date have been published for the Chinese writing system for school-aged children. In Taiwan, due to a lack of standardized evaluation tools, the occupational therapist and teacher often determine either subjectively or by clinical judgment whether the child s written productivity is adequate within the given time constraints. This practice not only hampers a meaningful comparison of an individual s performance to that of a normative sample but also the documentation of a child s progress. Progress in this field is therefore unnecessarily slow. The Special Education Act in Taiwan was passed in April 1997. This Act mandates that occupational therapy be a part of related services for children with special needs, which in turn will lead to an increase in the demand for occupational therapy in school settings. Thus, the need to provide a scale for measuring the handwriting speed of Chinese school-aged children has become urgent. The purpose of this study was to document handwriting speed performance of Chinese school-aged children from Grades 2 to 6 so that occupational therapists will be able to screen school-aged children who have handwriting speed problems using this norm as an objective guideline. Furthermore, this study s baseline data on handwriting speed for the Chinese writing system will be available for future research purposes.

296 Tseng and Hsueh LITERATURE REVIEW Speed and legibility are considered to be two major characteristics of handwriting. Handwriting speed is related to age and gender (Groff, 1961; Strickling, 1974; Ziviani, 1984). Compared with legibility, measurement of handwriting speed is much more objective. Freeman (1954) suggested using the letters written per minute to measure and establish the norm of the handwriting speed. This idea was proposed by Ayres in 1912. In Ayres s (1912) study, children of Grades 4, 5 and 6 were required to copy three sentences presented on the blackboard until they were familiar with them. Subsequently, children were asked to copy the same three sentences for two minutes and the number of letters written per minute were recorded. The findings showed that handwriting speed increased with age. Freeman (1954) found that the school-aged children s handwriting speed increased from 30 letters per minute in Grade 2 to 80 letters per minute in Grade 8, by using similar methods to those of Ayres (1912). However, his results for Grades 4, 5 and 6 were slower than those of Ayres. Groff (1961) argued that these earlier studies had the following methodological limitations: (1) handwriting speed based on copying sentences, which had been practised before the test, was misleading because of pretest effects; (2) the norms of handwriting speed appeared to be artificially inflated when compared with natural writing behaviour. Accordingly, Groff asked 5834 American children in Grades 4, 5 and 6 to read a standard sentence until they were familiar with it and recognized it. Subsequently, children were required to reproduce the passage, and performance was recorded as letters written per minute. Although this method gave considerably slower times than those of Ayres, Groff argued that his findings were more natural and realistic. Ziviani (1984) established the norms for handwriting speed for Australian school-aged children by asking the children to copy cats and dogs as many times as they could on the lines provided in two minutes. She found that school-aged children s handwriting speed increased from 32.60 letters per minute in Grade 3 to 52.15 letters per minute in Grade 7, and that girls wrote significantly faster than boys. Her findings support Groff s (1961) work. The various scales that have been developed to outline the handwriting speed of children of various ages are summarized in Table 1. In contrast with alphabetical and syllabic languages, Chinese may best be characterized as a morphemic language with each written symbol (a character) being a bound or unbound morpheme (Kao, Van Galen & Hoosain, 1986). Individual Chinese characters are often made up of constituent components (radicals), which are placed in different configuration within the generally square space occupied by the character (Kao et al., 1986). In the process of writing Chinese characters, pen-lifts and sharp turns are usually required. The writer is also required to follow specific stroke sequences and pay attention to details of character formation. Furthermore, in Taiwan, it is common practice for school-aged children to write Chinese characters in confined

Chinese handwriting speed test 297 spaces, ie., boxes on a sheet of grid paper. Suen (1983) demonstrated that it takes more time to write characters when constraints such as fixed stroke sequences and boxes are imposed. TABLE 1: Mean handwriting speed: comparison of data from various studies Grade Ayres Freeman Groff Ziviani Present (1912)* (1954)* (1961)* (1984)* study** 2 not available 30 not available not available 7.27 3 not available 40 not available 32.60 11.35 4 55 50 35.06 34.24 15.77 5 64 60 40.65 38.41 16.69 6 71 67 49.60 46.18 18.10 7 not available 74 not available 52.15 not available 8 not available 80 not available not available not available *mean handwriting speed in letters/minute **mean handwriting speed in characters/minute No norms for handwriting tests to date have been published for Chinese school-aged children. Hence, establishing these norms for handwriting speed should provide baseline information for future research and clinical practice, and, given the differences in writing systems, it should be of interest to see how they compare with their Western counterparts. Since it is not until the second semester of the first grade that Chinese school-aged children start to learn Chinese characters, norms of handwriting speed were established for Grades 2 and up. METHODS Participants Participants were 1525 children in Grades 2 6 from three elementary schools in the Taipei area. The schools were randomly selected and represented heterogeneous socioeconomic backgrounds. Two classes were then randomly selected from each grade in each school. The age and gender distributions of the children are summarized in Table 2. Materials Children were provided with A4-sized writing paper (21.0 cm 29.7 cm) with a pre-printed grid. The size of the grid s boxes used by Grades 2 and 3 was 1.7 cm 1.7 cm, with a total of 120 boxes per sheet. The size of the boxes used by Grades 4, 5 and 6 was 1.2 cm 1.2 cm, making a total of 231 per sheet. The box sizes were in accordance with those of the workbooks used in school.

298 Tseng and Hsueh TABLE 2: Age ( in years) and gender distribution of sample Boys (n = 825) Girls (n = 700) Totals (n = 1525) Grade n Mean age SD n Mean age SD n Mean age SD 2 153 7.64 0.39 141 7.57 0.40 294 7.61 0.39 3 169 8.55 0.32 116 8.56 0.35 285 8.55 0.33 4 156 9.52 0.28 121 9.55 0.31 277 9.53 0.30 5 172 10.57 0.36 149 10.48 0.34 321 10.55 0.35 6 175 11.51 0.31 173 11.54 0.34 348 11.53 0.32 In Taiwan, the teaching materials, including Chinese textbooks for elementary education, are unified across the country. Children were asked to copy texts from a Chinese textbook that they had studied in the second semester of the first grade. This was done so that children would not encounter any unfamiliar characters that might impede their handwriting speed. The texts, which contained 475 characters, were re-typed on a sheet of A4-sized paper in order to remove phonetic and punctuation symbols, leaving only Chinese characters to be copied. The number of strokes of each Chinese character varied from one to 22, with a mean of 9.09 (SD = 4.14). The font size was 0.7 cm 0.7 cm. The font type was Kai-Su, which is the most common style and is the same as that used in children s textbooks (see Appendix).[AQ] Procedures The handwriting speed test was administered by the researcher to the whole class simultaneously. Subjects were tested in a classroom situation: they were asked to copy the text in pencil and were reminded to write as quickly and legibly as possible and not to stop to correct mistakes. The test lasted for five minutes, after which the researcher collected the writing samples. During the test, the researcher recorded the writing hand used by each subject. The same test was re-administered to a group of 25 subjects, five from each grade, one week later to gauge test retest reliability. Data analysis Handwriting speed was defined as the number of characters written per minute. Characters that were malformed or had one or two additional or missing strokes were counted; however, incompletely (defined as those in which three or more strokes had been omitted) written characters were not counted. Descriptive (means and standard deviations) and inferential statistics were

Chinese handwriting speed test 299 produced using SPSS for windows version 6.0 (Norusis, 1993). A two-way ANOVA was used to analyse the influence of gender and grade on handwriting speed. RESULTS Distribution of writing hand Ninety-seven per cent (n = 1481) of the subjects were right-handed, and 2.9% (n = 44) were left-handed. Of girls, 98.4% were right-handed, and 1.6% were left-handed. Of boys, 96.1% were right-handed, and 3.9% were lefthanded. Writing speed The mean handwriting speeds are summarized in Table 3. A two-way ANOVA showed significant main effects of grade (F(4, 1515) = 329.49, p <0.001) and gender (F(1, 1515) = 405.77, p <0.001). Handwriting speed increased with age, with children in Grade 6 (18.10 characters/minute) being 2.5 times faster than children in Grade 2 (7.27 characters/minute). The greatest increase in handwriting speed was seen between Grades 2, 3 and 4. Furthermore, there was significant interaction between grade and gender (F (4, 1515) = 2.83, p <0.05). Post hoc analysis indicated that girls wrote significantly faster than boys in Grades 3, 4, and 5 (F (1, 283) = 9.457, p = 0.0023; F (1, 275) = 6.537, p = 0.0111; and F (1, 319) = 8.108, p = 0.0047, respectively). However, the handwriting speed of girls and boys in Grades 2 and 6 was not significantly different (F (1, 292) = 2.165, p = 0.1423; and F (1, 346) = 3.721, p = 0.0545). Retest reliability was computed through intraclass correlation. The intraclass correlation coefficient was 0.98 with a one-week interval. TABLE 3: Mean handwriting speed (characters/minute) for each grade Boys (n = 825) Girls (n = 700) Totals (n = 1525) Grade n Mean SD n Mean SD p n Mean SD 2 153 7.50 2.94 141 7.01 2.82 0.142 294 7.27 2.89 3 169 10.79 3.71 116 12.17 3.74 0.002 285 11.35 3.77 4 156 15.23 4.01 121 16.36 3.76 0.011 277 15.77 4.03 5 172 16.18 3.29 149 17.25 3.17 0.005 321 16.69 3.27 6 175 17.59 5.41 173 18.61 4.34 0.055 348 18.10 4.92

300 Tseng and Hsueh DISCUSSION Results showed that writing speed increased with age, which supports earlier studies (Gates, 1924; Freeman, 1954; Groff, 1961; Lamme, 1979; Ziviani, 1984; Weil & Amundson, 1994). This is in accordance with empirical evidence that indicates continual progress in coordinated handwriting movements by age and schooling (Meulenbroek & Van Galen, 1986, 1989; Søvik, 1993). These improvements in coordination can largely explain why older children are able to write significantly faster than younger ones. The nominal writing speeds in this study were much slower than those reported from Western countries. However, for the English writing system, handwriting speed is defined as the number of letters written per minute, whereas for the Chinese writing system it is defined as the number of characters written per minute. Since a Chinese character contains far more strokes than a letter (an average of 9.09 strokes in the present study); then other things being equal, it would naturally take more time to write a character than to write a letter. Moreover, written practice and repeated copying of the same sentence, which may facilitate writing speed, were adopted in several of the previous studies (Ayres, 1912; Freeman, 1954; Ziviani, 1984). In the present study, children were asked to copy texts that they had already studied but without pretest practice and repetitive copying; although this might have led to slower handwriting speeds, we consider this to be a more realistic reflection of natural writing behaviour. As to overall handwriting performance, teachers pay much attention to both the quantitative (speed) and the qualitative (legibility) aspects of children s writing when testing and observing children s progress in this skill (Søvik, 1993). Even so, speed is probably the simplest overall measure of proficiency in writing. The rate of increase was found to be greatest through Grades 2, 3, and 4 where the mean ages were 7.6, 8.6, and 9.5 years, respectively. This is consistent with the results of Søvik (1975) and Ziviani (1984), who found that the rate of increase is greatest between 7 and 9 years, after which it tapers off until about the age of 14, when there is little further increase. However, in the present study, we did not investigate the average handwriting speed of children beyond Grade 6, where the mean age was 11.5 years. Gender differences The finding that girls wrote faster than boys in Grades 3, 4, and 5 was in accordance with earlier studies (Groff, 1963; Rarick and Harris, 1963; Strickling, 1974; Anderson, 1976; Goetz, 1980; Ziviani, 1984). There are known to be gender differences in maturational status in childhood, with girls having accelerated rate of maturation (Eaton & Yu, 1989), and these differences may account for the observed gender differences in speed. Furthermore, the process of gender-role identification has been recognized as an important potential source of environmental influence on gender differ-

Chinese handwriting speed test 301 ences in motor performance (Greendorfer, 1980; Thomas & French, 1985). Parents and teachers have different expectations for boys than for girls (Fagot, 1978; Thomas & French, 1985). Girls are usually expected to participate more in fine motor tasks as opposed to gross motor activities. Girls would thus have more practice in fine motor skills than boys; this could lead to girls having greater dexterity for handwriting than boys. Handedness The percentage of left-handedness (2.9%), as defined by the writing hand, is much lower than that in Ziviani s (1984) study. In her study, 56 children (9.7% of the sample) were left-handed. This may possibly be due to differences in cultural practice. Chinese tend to depreciate those who are lefthanded. Thus, parents of children who are predisposed to be left-handed are inclined to force their child to use their right hand especially when writing and using chopsticks. Limitations of the study The preliminary norms for handwriting speed were mainly based on the school-aged children drawn from a specific geographic location, ie., only limited to the Taipei area, which is the metropolis of Taiwan. Thus, the norms may not be applicable to school-aged children in other areas of Taiwan, especially the rural area. Future studies may need to expand the sample size to incorporate participants from different geographic locations in Taiwan in order to be reflective of the entire population of school-aged children. CONCLUSION Preliminary norms for handwriting speed of Chinese school-aged children were established in this study. Taking these norms as an objective guide, occupational therapists will be able to screen school-aged children with slow handwriting. Differences in writing speed have been revealed over an agerange of 7.6 11.5 years, and also between the sexes, in favour of girls. Future studies using subjects aged from 11.6 years to 14.5 years are needed to investigate at what age the writing speed reaches the plateau, as well as at what age the sex differences cease to exist. ACKNOWLEDGMENT The authors gratefully express their appreciation to Yu Chang, Susanna Mei- Kum Chow, and Dr Grace Kaiping Yao for their valuable inputs. This study was supported through funding received from the National Science Council, NSC 86-2314-B-002-236.

302 Tseng and Hsueh REFERENCES Amundson SJ, Weil M (1996). Prewriting and handwriting skills. In J Case-Smith, AS Allen, PN Pratt (eds) Occupational Therapy for Children. St Louis, MO: Mosby, pp. 524 41. Anderson EM (1976). Handwriting and spina bifida. Education Forward Trends, 3, 17 20. Ayres LP (1912). A Scale for Measuring the Quality of Handwriting of School Children. New York: Russell Sage Foundation. Cermak S (1991). Somatosensory dyspraxia. In A Fisher, EA Murray, AC Bundy (eds) Sensory Integration: Theory and Practice. Philadelphia: F.A. Davis, pp. 138 70. Eaton WO, YU AP (1989). Are sex differences in child motor activity level a function of sex differences in maturational status? Child Development, 60, 1005 11. Fagot BI (1978). The influence of sex of child on parental reaction to toddler children. Child Development, 49, 459 65. Freeman FN (1954). Teaching handwriting. What Research Says to the Teacher, No. 4. NEA, 33pp.[AQ] Gates AI (1924). Relation of quality and speed of performance: A formula for combining the two in the case of handwriting. Journal of Educational Psychology, 15, 124 44. Goetz E (1980). Early Writing: A Developmental Approach. (Report No. PS 012 045). (ERIC Document Reproduction Service No. ED 198 939). Greendorfer SL (1980). Gender differences in physical activity. Motor Skills: Theory into Practice, 4, 83 90. Groff PJ (1961). New speeds of handwriting. Elementary English, 38, 564 5. Groff PJ (1963). Who writes faster? Education, 83, 367 9. Kao HSR, Van Galen GP, Hoosain R (1986). Graphonomics: Contemporary Research in Handwriting. North-Holland: Elsevier. Lamme LL (1979). Handwriting in an early childhood curriculum. Young Children, 35, 20 27. Larsen SC, Hammill DD (1989). Test of legible handwriting. Austin, TX: Pro-Ed. McHale K, Cermak S (1992). Fine motor activities in elementary school: Preliminary findings and provisional implications for children with fine motor problems. American Journal of Occupational Therapy, 46, 898 903. Meulenbroek RGJ, Van Galen GP (1986). Movement analysis of repetitive writing behaviour of first, second, and third grade primary school children. In HSR Kao, GP Van Galen, R Hoosain (eds) Graphonomics: Contemporary Research in Handwriting. Amsterdam: North-Holland, pp. 71 92. Meulenbroek RGJ, Van Galen GP (1989). The production of connecting strokes in cursive script: Developing co-articulation in 8 to 12 year-old children. In R Plamondon, CY Suen, ML Simner (eds) Computer Recognition and Human Production of Handwriting. Singapore: World Scientific, pp. 273 86. Norusis MJ (1993). SPSS for Windows: Base System User s Guide, Release 6.0. Chicago, IL: SPSS Inc. Oliver CE (1990). A sensorimotor program for improving writing readiness skills in elementaryage children. American Journal of Occupational Therapy, 44, 111 24. Phelps J, Stempel L. (1987). The Children s Handwriting Evaluation Scale for Manuscript Writing. Dallas: Texas Scottish Rite Hospital for Crippled Children. Phelps J, Stempel L, Speck G (1985). The children s handwriting scale: A new diagnostic tool. Journal of Educational Research, 79, 46 50. Rarick GL, Harris TL (1963). Physiological and motor correlates of handwriting legibility. In VE Herrick (ed.) New Horizons for Research in Handwriting. Madison: University of Wisconsin Press, pp. 55 94. Søvik N (1975). Developmental Cybernetics of Handwriting and Graphic Behaviour. Oslo: Universitetsforlaget.

Chinese handwriting speed test 303 Søvik N (1993). Development of children s writing performance: Some educational implications. In AF Kalverboer, B Hopkins, R Geuze (eds) Motor Development in Early and Later Childhood: Longitudinal Approaches, New York: Cambridge University Press. Strickling CA (1974). The effect of handwriting and related skills upon the spelling score of above average and below average readers in the fifth grade. Dissertation Abstracts, 34(7), 3717-A. Suen CY (1983). Handwriting generation, perception and recognition. Acta Psychologica, 54, 295 312. Thomas JR, French KE (1985). Gender differences across age in motor performance: A metaanalysis. Psychological Bulletin, 98, 260 82. Tseng MH, Cermak S (1993). The influence of ergonomic factors and perceptual-motor abilities on handwriting performance. American Journal of Occupational Therapy, 47, 919 26. Weil MJ, Amundson SJC (1994). Relationship between visuomotor and handwriting skills of children in kindergarten. American Journal of Occupational Therapy, 48, 982 8. Ziviani J (1984). Some elaborations on handwriting speed in 7- to 14-year-olds. Perceptual and Motor Skills, 58, 535 9. Address correspondence to Mei Hui Tseng, Associate Professor, School of Occupational Therapy, National Taiwan University, 7 Chung-shan South Rd, Taipei, Taiwan. E-mail: mhtseng@ha.mc.ntu.edu.tw APPENDIX