Handwriting development in grade 2 and grade 3 primary school children with normal, at risk, or dysgraphic characteristics By: Anneloes Overvelde and Wouter Hulstijn

Fig.1. Mean BHK quality scores in children with normal writing (NW), at risk writing (RW) and dysgraphic writing (DW) in grade 2 and grade 3 at T1 and T2.

Fig.2. Mean BHK speed scores in children with normal writing (NW), at risk writing (RW) and dysgraphic writing (DW) in grade 2 and grade 3 at T1 and T2.

(test session*grade: F (2,166) =14.60, p<.001, partial h2=.150). Mean score in the DW group improved to within the range of at risk (BHK score from 33.58 to 26.16), while the RW group reached the normal writing level at T2 (BHK score from 24.83 to 20.40). Children in the NW group only showed minor improvement (BHK score from 16.29 to 15.47).

Correlations between the BHK quality scores at T1 and T2 were significant (r=.596, p<.001). In grade 3, a different pattern of change in BHK scores was observed (test session*grade: F(2,67) = 6.66,p=.002, partial h2=.166).The10 children in the RW group showed the most improvement (BHK score from 24.10 to 18.30), while all four children in the DW group had reached the boundary level of dysgraphia (BHK score from 32.25 to 29.00). Performance of the large group with normal writing (n=56) was poorer at T2 (BHK from 13.71 to 15.07), but remained well within the range of normal.

Thus, during grade 3, only small shifts occurred and 71% of the children (almost exclusively with normal writing) remained in the same subgroup. The correlation between BHK quality scores at T1 and T2 was significant (r=.583, p<.001). In grades 2 and 3, as expected, the overall performance of the girls was better than that of the boys (grade 2 BHK scores at T1 in girls and boys are 22.32 and 27.97, respectively; at T2: 18.23 and 23.38, respectively; in grade 3 BHK scores at T1 in girls and boys are 13.33 and 18.86, respectively; at T2: 14.85 and 17.65, respectively).

At T1 in grade 2, 49% of the girls (40 out of the 81) were categorized as NW, while 49% of the boys (43 out of the 88) had dysgraphic scores. In grade 3, 94% of the girls (31 out of the 33) had reached the normal writing level compared to 68% of the boys (25 out of the 37). The DW group comprised only (four) boys. Developments in handwriting quality were significant only in grade 2 (p<.001). No interactions in T1 and T2 scores were found between boys and girls.

Table 2 shows the differences in development between grade 2 and grade 3 on an item level. In grade 2, the improvement in handwriting quality was caused by improvements in three clusters of items.

The effect size of improvements in letter size and height (items 8 and 9) was higher than the decline incorrectly written letter forms (items 10 and 11). Fine motor ability aspects (items 5 and 13) improved, as well as the organization of written work.

In grade 3, improvements in two items (consistency of letter size and steadiness of writing trace) between T1 and T2 could be interpreted as progress in the consistency of the handwriting. Only one (low-scoring) item, collision, had deteriorated between grades 2 and 3.

Comparisons between the two grades, however, showed a different picture: in grades 2 and 3, all the items in the clusters Letter formation and Fine-motor ability differentiated between the three handwriting subgroups, while four out of the five items in the cluster Organization of written work differentiated significantly between the subgroups in grade 2, but only one of the latter items was significant in grade 3.

Details of handwriting speed are shown in Table 1 and Fig. 2. As expected, the total number of written letters was higher in grade 3 than in grade 2 (F(1,237) = 40.57, p<.001). Children in grade 2 showed a significant increase in the number of letters (n=20.51; F(1,166) = 64.66,p<.001, partial h2=.280). The number of written letters in grade 2 differed significantly between the children with normal, at risk and dysgraphic writing (F(2,166) = 6.307, p=.002, partial h2=.071), and the interaction with group and test session was significant (F(2,166) = 3.40, p=.036, partial h2=.039).

Post hoc analysis revealed that this interaction was due to a smaller increase in the number of letters in the dysgraphic children (DN letters =1 2.06) than in the at risk (DN letters = 22.7, p=.010) and normal writing groups (DN letters = 27.97, p=.006). In grade 3, no significant effects of test session (T1–T2) were found in handwriting speed (F(1,67)=.058, p=.811).

Although there was a considerable increase (of 20.73) in the number of written letters in the large normal writing group (55 out of the 80 children) (t(56) = 3.38, p=.001), in contrast to a decrease (of 15.60) in the number of letters in the at risk subgroup and an equal number of letters in the four dysgraphic children, the interaction between test session and the handwriting subgroups was not significant (speed T1–T2*Grade: F(2,67)=2.48, p=.091).

In grades 2 and 3, the correlations between T1 and T2 BHK speed scores were significant (grade2: r= .597, p<.001; grade 3: r=.359, p=.002). With respect to gender, results of handwriting speed were comparable with the results of handwriting quality in grade 2. Overall, the girls out performed the boys in speed of handwriting (grade2: F(1,167) = 12.08, p=.001, partial h2=.067), but not in grade 3 (p=.091). No interaction between gender and test session was found.

BHK quality		Grade 2 n=169				Grade 3 n=70
Item descriptions	It	T1	T2	Partial n^2 T1–T2	Partial n^2gr	T1	T2	Partial n^2 T1–T2	Partial n^2gr
Organization of written work
Overall letter size	It1	1.93	1.22	0.171*	0.263*	0.70	1.00	0.000	0.315*
Left margin widening	It2	1.17	0.78	0.051*	0.048*	0.51	0.44	0.030	0.016
Alignment	It3	3.57	2.62	0.199*	0.317*	1.54	2.04	0.001	0.058
Word spacing	It4	2.73	2.78	0.001	0.055*	2.71	2.47	0.003	0.049
Collision	It7	0.08	0.41	0.165*1	0.001	0.20	1.03	0.074*1	0.012
Letter formation
Inconsistent letter size	It8	4.15	2.62	0.384*	0.190*	2.63	1.26	0.174*	0.155*
Incorrect letter height	It9	1.72	1.23	0.136*	0.288*	0.81	0.66	0.008	0.209*
Letter distortion	It10	0.40	0.66	0.048*1	0.183*	0.49	0.93	0.017	0.115*
Ambiguous letter forms	It11	2.36	2.89	0.097*1	0.324*	1.90	2.50	0.021	0.352*
Fine-motor ability
Acute turns	It5	2.44	1.86	0.092*	0.226*	1.27	1.00	0.017	0.212*
Irregularities	It6	2.48	2.20	0.023	0.360*	1.64	1.63	0.019	0.123*
Correction letter forms	It12	1.02	1.11	0.005	0.240*	0.73	0.94	0.000	0.095*
Unsteady writing trace	It13	1.22	0.54	0.130*	0.170*	1.13	0.43	0.272*	0.307*

*p<.05; significant decline in item scores are coded with 1

 

 

 

3.4. Visual-motor integration

Mean VMI standard scores and correlations between BHK quality and VMI-SS are given inTable1.All mean standard VMI scores were close to the average standard score (= 100) in grades 2 and 3 at T1 and T2. There was a small, but significant increase at T2 in grades 2 and 3 (grade 2: 4.68; p<.001; grade 3: 3.46; p=.004), due to a learning effect. Significant differences between handwriting subgroups were only found in grade 2 (F(2,165) =4.80, p=.009; grade 3: F(2,67) = 2.847, p=.065).

Post hoc analysis showed that this difference (of 5.20 SS) was caused by the NW group versus the DW group (p=.008). Significantly low correlations between the BHK quality scores and the standard VMI scores were found in grade 2 and at T1 in grade 3. Clustering on an item level (in conformity with the BHK) showed that in grade 2, the clusters Letter Formation and Fine Motor Ability had low correlations (r=.136 and r=.133, respectively). In grade 3, only the cluster Organization of Written Work had a low correlation (r=.256). There were no significant differences in the numbers of children with poor, normal, or good VMI-SS between the handwriting subgroups (grade 2: X2(4) =.283; grade 3: X2(4) =.180). Only 26% of the DW subgroup in grade 2 had poor VMI-SS, while only 11% of the NW group in grade 3 had good VMI-SS.

4. Discussion

This study tested the widely held belief that the proper time to assess the quality of children's handwriting is in the second half of grade 2. By that age, handwriting is assumed to have been mastered (i.e. during grade 1 and the first half of grade 2 and it is expected to remain stable from that time onwards). Contrary to this widely proposed hypothesis, the present study demonstrated considerable improvements in the quality of handwriting throughout grade 2 and even into the first half of grade 3, with stability being reached in the second half of grade 3.

More importantly, the percentage of children who were classified as dysgraphic decreased strongly from 37% to 17% in grade 2 and diminished further to a low and stable rate of 6% in grade 3. It is also worth noting that stability in handwriting quality only occurred in the children with normal writing.

The subgroups who were classified as at risk or dysgraphic, continued to show significant and substantial improvement during grades 2 and 3. In this study, data were gathered from two large samples of 169 children in grade 2 and 70 children in grade 3 who were attending regular mainstream primary schools.

The demographic variables of these children, as well as the prevalences of minor learning, behavioural or motor problems and the number of children who were receiving therapeutic assistance, were all within the normal range, which suggests that these samples were representative of the general population at mainstream schools.

4.1. At risk and dysgraphic writing

The percentages of children with normal writing increased from 33% (T1) to 54% (T2) in grade 2, reached 80% in the first half of grade 3 and then remained stable during the rest of that grade.

For the Dutch operationalization of the qualitative criteria of Developmental Coordination Disorder (DCD; DSM-IV-TR, APA, 2000) with respect to poor handwriting, a cut-off point of BHK 21 has been recommended, which includes the ranges of at risk and dysgraphic writing (see the Dutch version of the Leeds Consensus Statement [LCS], 2006).

According to this norm, as many as 46% of our children would have fulfilled this qualitative criterion for handwriting-DCD at the proposed time of testing, i.e. in the second half of grade 2. The large improvement in quality observed during grade 2 and the first half of grade 3 indicates that it is not justified to label or classify handwriting as at risk or dysgraphic until grade 3.

By this age, poor and illegible handwriting was less common and quality itself remained more stable, which resulted in no more than 14–16% of the children being classified as at risk and 6% as dysgraphic, based on their BHK quality score. None of the studies that used the BHK to assess handwriting focused on children in grades 2 and 3 and observed the rapid improvement in handwriting found in the current study.

Mostly, children in the second half of grade 2 and in grade 3 were combined into one group, as occurred in the study by Smits-Engelsman et al. (2001) on 125 children. These authors had 72% of their children in grade 2 and found that 6% of their participants fell within the dysgraphic range, while another 27% were at risk. In our study, if we combined the scores obtained in the second half of grade 2 (71%) with the scores of grade 3, then14% of our children would have been classified as dysgraphic and 25% as at risk.

Thus, compared to the study by Smits-Engelsman et al. (2001), we had more than twice as many dysgraphic children. In a study on children who were about two years older, Van Hoorn, Maathuis, Peters, and Hadders-Algra (2010) found lower percentages of children with poor handwriting (2% with dysgraphia and 19% at risk). These percentages were similar to those found in studies that used other instruments. Rubin and Henderson (1982) obtained data from 82 grade-4 teachers (representing approximately 2500 children) and found a lower percentage of children with serious handwriting problems (12.3%).

Maeland (1992) reported that 9.6% of their 10-year-old children had handwriting problems, while Karlsdottir and Stefansson (2002) classified 13% of their grade-5 children as having dysfunctional handwriting. These findings indicate clearly that handwriting should not be assessed during grade 2, but postponed until grade 3 or later.

 

 

  4.2. Development of handwriting

In agreement with the studies on handwriting development by Hamstra-Bletz and Blote (1993) and Karlsdottir and Stefansson (2002) we observed considerable improvement in the quality of handwriting during grades 2 and 3, but only in the children with at risk or dysgraphic writing. Hamstra-Bletz and Blote (1993) reported that the children who had dysgraphia in grade 2 (10% of the sample) still had writing problems in the subsequent grades.

More informative results were presented in the study by Karlsdottir and Stefansson (2002) on the development of handwriting from grade 1 to grade 5. These authors reported that handwriting quality increased rapidly during grade 1 and remained stable from grade 2 to grade 5, which was largely confirmed by our observations. However, these results only applied to the children with normal writing. Their dysfunctional writers made more gradual and weaker improvement from grade 1 to grade 5. 

 

4.3. Analysis on a BHK item level

In contrast to most handwriting evaluation scales, the Concise Assessment Scale for Childrens Handwriting (BHK) enables detailed scoring on 13 items (characteristics) divided into 3 clusters (see Table 2). Our BHK results showed that poor handwriting was not confined to a specific set of these 13 items. In grade 2, all of the items except for one contributed to differentiating between the three subgroups (normal, at risk and dysgraphic). In grade 3, this applied to nine out of the 13 items, eight of which belong to the clusters Letter formation and Fine-motor ability.

In addition, the improvements in handwriting in grade 2 involved almost all the characteristics. Please note that the names given to the three clusters only specify the visible product of the handwriting and its legibility. Therefore, they cannot be used as indications of the underlying deficits. Rubin and Henderson (1982), Maeland (1992) and Karlsdottir and Stefansson (2002) stressed that weak letter formation was the cause of reduced legibility in poor handwriting. Smits-Engelsman and Van Galen (1997) and Smits-Engelsman et al. (2001) presented additional kinematic data on children with handwriting problems that suggested deficits in fine motor control as the basis for poor handwriting.

4.4. Handwriting speed

It is generally assumed that handwriting speed increases as the grades progress. Speed increased in all three subgroups in grade 2 in the current study, but only in the NW group in grade 3. The dysgraphic subgroup in grade 2 had a significantly lower writing speed that discriminated them from the other subgroups.

This might have been because these children needed to pay extra attention to their handwriting in order to produce their best effort. In grade 3, handwriting speeds did not differ between the three subgroups, which was in line with the results reported by Blote and Hamstra-Bletz(1991) and Karlsdottir and Stefansson (2002).

4.5. Beery VMI as a screening instrument for handwriting quality

Although Beery and Beery(2004), one of the designers of the Beery Developmental Test of Visuomotor Integration (VMI), pointed out that the VMI was not intended to evaluate handwriting abilities, this test is widely considered to measure an integral component of skilled handwriting and has often been used in handwriting assessment.

Our data showed that the VMI was unsuitable to screen children for dysgraphia. Although we found significant differences in VMI standard scores between the handwriting subgroups, all the VMI standard scores fell within the normal range (mean SS from 96 to 101). Significant but low correlations were only found in grade 2 and at T1 in grade 3. Then a non-significant correlation of r = 0.11 was seen at T2, which was the best time to assess the children for handwriting problems.

In a study on mainstream school children (two years older than our participants) Van Hoorn et al. (2010) found comparable VMI scores and the same lack of correlation with dysgraphic handwriting. Equally low correlations between handwriting quality and VMI performance were also reported by Karlsdottir and Stefansson (2003), Maki, Voeten, Vauras, and Poskiparta (2001) and Goyen and Duff (2005). This leads to the conclusion that the VMI is inappropriate as a screening instrument for handwriting difficulties.

4.6. The quality of handwriting in the past 20 years

The first mean BHK scores were reported by Hamstra-Bletz and Blote (1993) from a study on 121 children in grade 2. Their mean BHK score was 15.0 (Hamstra-Bletz & Blote, 1993), which was much better than our mean BHK score of 20.9 in a comparable group of 169 children. Their mean score in the poorest 10% of the children was 29, whereas in the present study, about 20 years later, the poorest 10% had a mean score of 32.

In the publication discussed above, Smits-Engelsman et al. (2001) found only half as many dysgraphic children compared to the present study in their 125 children in grade 2 and grade 3. These findings all suggest that the quality of handwriting in grade 2 has declined over the past two decades. Given that our findings on handwriting quality in grade 3 did not seem to be at odds with earlier research, it might be suggested that factors related to the amount of time spent on handwriting or the quality of the lessons in grade 2 have changed over the past 10–20 years.

 

5. Conclusions and implications

The best time to screen for dysgraphic handwriting seems to be of crucial importance. In this study, the children had not reached their final handwriting quality level until the first half of grade 3. Thus, a diagnosis of dysgraphia and decisions about referral to a physiotherapist or occupational therapist (for remedial treatment of dysgraphia) must not be made too quickly and definitely not in the second part of grade 2, which is now common practice.

As handwriting quality is an important criterion for the diagnosis of Developmental Coordination Disorder (criterion A; DSM-IV-TR, APA, 2000; see Leeds Consensus Statement [LCS], 2006) the course of dysgraphich and writing development should be taken into account. Children with normal handwriting reached a stable level in grade 2, but the children with handwriting problems (those at risk, or with scores in the dysgraphic range) continued to develop strongly during grade 2 and the first half of grade 3.

Our results showed that the VMI was in appropriate as a supplemental screening instrument for handwriting problems in young children. In line with the task-specific approach taken in assessment and therapy, handwriting should be measured with a handwriting instrument.

 

 

Acknowledgements

We would like to thank all the children and their parents and teachers for their enthusiastic contributions to the study, Merel van Nimwegen and Linda Sluijs for helping to score the many handwriting products from the children and Judith Abma-Hill for reviewing the English language.

References

Abbott, R. D., & Berninger, V. W. (1993). Structural equation modeling of relationships among developmental skills in primary-grade and intermediate-grade writers. Journal of Educational Psychology, 85(3), 478–508. APA. (2000). Diagnostic and statistic manual of mental disorders (4th rev. ed.). Washington, DC: American Psychiatric Association.

Beery, K. E., & Beery, N. A. (2004). Beery VMI: The Beery-Buktenica developmental test of visual-motor integration (5th ed.). Minneapolis: Pearson.

Blote, A. W., & Hamstra-Bletz, L. (1991). A longitudinal study on the structure of handwriting. Perceptual and Motor Skills, 72, 983–994.

Bosga-Stork, I., Overvelde, A., Van Bommel-Rutgers, I., Van Cauteren, M., Halfwerk, B., Nijhuis-van der Sanden, R., et al. (2009). Inventarisatie van verwijzingspatroon, onderzoek en behandeling van kinderen met schrijfproblemen. Een digitale enque ˆte. Nederlands Tijdschrift voor Kinderfysiotherapie, 2, 14–18.

De Ajuriaguerra, J., Auzias, M., Coumes, F., Denner, A., Lavondes-Monod, V., Perron, R., et al. (1964). Le´criture de lenfant I: le´volution de le´criture et ses difficulte´s. Paris: Delachaux et Niestle´.

De Ajuriaguerra, J., Auzias, M., Coumes, F., Denner, A., Lavondes-Monod, V., Perron, R., et al. (1979). Le´criture de lenfant I: le´volution de le´criture et ses difficulte´s (3rd ed.). Paris: Delachaux et Niestle´.

Feder, K. P., & Majnemer, A. (2007). Handwriting development, competency, and intervention. Developmental Medicine and Child Neurology, 49(4), 312–317.

Flapper, B. C. T., Houwen, S.,&Schoemaker,M.M.(2006). Fine motor skills and effects of methylphenidate in children with attention-deficit hyperactivity disorder and developmental coordination disorder. Developmental Medicine and Child Neurology, 48(3), 165–169.

Goyen, T.-A., & Duff, S. (2005). Discriminant validity of the developmental test of visual-motor integration in relation to children with handwriting dysfunction. Australian Occupational Therapy Journal, 52, 109–115.

Graham, S., & Weintraub, N. (1996). A review of handwriting research: progress and prospects from 1980 to 1994. Educational Psychology Review, 8(1), 7–87.

Hammerschmidt, S. L., & Sudsawad, P. (2004). Teachers survey on problems with handwriting: referral, evaluation, and outcomes. American Journal of Occupational Therapy, 58(2), 185–192.

Hamstra-Bletz, L. (1993). Het Kinderhandschrift. Ontwikkeling en Beoordeling [Childrens Handwriting. Development and Assessment]. Thesis. Leiden: University of Leiden.

Hamstra-Bletz, L., & Blote, A. (1993). A longitudinal study on dysgraphic handwriting in primary school. Journal of Learning Disability, 26, 689–699.

Hamstra-Bletz, L., De Bie, J., & Den Brinker, B. P. L. M. (1987). Beknopte beoordelings methode voor kinderhandschriften: Experimentele versie [Concise evaluation scale for childrens handwriting: Experimental version]. Lisse: Swets&Zeitlinger.

Kaiser, M.-L., Albaret, J.-M., & Doudin, P.-A. (2009). Relationship between visual-motor integration, eye-hand coordination, and quality of handwriting. Journal of Occupational Therapy, Schools, and Early Intervention, 2, 87–95.

Karlsdottir, R., & Stefansson, T. (2002). Problems in developing functional handwriting. Perceptual and Motor Skills, 94, 623–662.

Karlsdottir, R., & Stefansson, T. (2003). Predicting performance in primary school subjects. Perceptual and Motor Skills, 97, 1058–1060.

Maeland, A. F. (1992). Handwriting and perceptual-motor skills in clumsy, dysgraphic, and normal children. Perceptual and Motor Skills, 75(3 Pt 2), 1207–1217.

Maki, H.S., Voeten, M.J. M.,Vauras, M. M.S., &Poskiparta, E.H.(2001). Predicting writing skill development with word recognition and preschool readiness skills. Reading and Writing, 14, 643–672.

Olive, T., & Kellogg, R. T. (2002). Concurrent activation of high- and low-level production processes in written composition. Memory andCognition, 30(4),594–600.

Rubin,N., &Henderson,S.E.(1982). Two sides of the same coin: Variation in teaching methods and failure to learn to write. Special Education: Forward Trends, 9(4), 17–24.

Smits-Engelsman, B. C., Niemeijer, A. S., & Van Galen, G. P. (2001). Fine motor deficiencies in children diagnosed as DCD on poor grapho-motor ability. Human Movement Science, 20(1–2), 161–182.

Smits-Engelsman, B. C. M., & Van Galen, G. P. (1997). Dysgraphia in children: Lasting psychomotor deficiency or transient developmental delay? Journal of Experimental Child Psychology, 67(2), 164–184.

Sugden, D. A. (Ed.). (2006). Leeds consensus statement. developmental coordination disorder as a specific learning difficulty. Economic and social research council research seminar. Brochure. Retrieved from the World Wide Web: http://www.dcd-uk.org.

Van Galen, G.P. (1991). Handwriting: Issues for a psychomotor theory. Human Movement Science, 10, 165–191.

Van Hoorn, J. F., Maathuis, C. G. B., Peters, L. H. J., & Hadders-Algra, M. (2010). Handwriting, visuomotor integration, and neurological condition at school age. Developmental Medicine and Child Neurology, 52(10), 941–947.

WHO. (1992). The ICD-10 classification of mental and behavioural disorders: Clinical descriptions and diagnostic guidelines. Geneva: World Health Organization.