Abstract
Handwriting is an important ability for literacy (Falk et al., 2010); it is a means by which children can show their learning, creativity, and knowledge (Wallen et al., 2013). Recognized as an important motor skill acquired in school (Bisio et al., 2016), handwriting occupies a significant amount of school time (Feder & Majnemer, 2007); it allows children to register thoughts, emotions, and other information related to the acquisition of knowledge and integrates perceptual, cognitive, and sensorimotor functions (Calvo et al., 2014).
Studies have shown the importance of handwriting for brain development as well as for cognitive, motor coordination, and perceptual function (Gerth et al., 2016; James & Engelhardt, 2012; Maldarelli et al., 2015). There is evidence that the brain responds differently when one learns through handwriting compared with tracing letters or typing; handwriting is important for letter processing in the brain regions that underlie reading and can thus facilitate the acquisition of reading skills in young children (Berninger et al., 2009; James & Engelhardt, 2012). Handwriting is a very complex ability that subsidizes the integration of several skills and areas of the brain, because to copy letters and words it is necessary not only to control the fine movements and the strength of the fingers and hand but also to maintain a working memory to coordinate the visual and manual movements, integrating perceptual–motor information with linguistic rules about orthographic structure (Stevenson & Just, 2014).
The estimated prevalence of handwriting problems varies from 5% to 30% depending on the criteria and assessment instruments used (Karlsdottir & Stefansson, 2002; Rosenblum et al., 2004), and such problems are expressed by low legibility and slow handwriting speed (Lam et al., 2011). Children with handwriting problems tend to use their cognitive resources in the motor planning aspects of writing instead of in the composition and written expression of their ideas (Case-Smith et al., 2014). These children need more time to complete tasks in the classroom (Feder & Majnemer, 2007; Graham et al., 2006) or at home, and they may avoid writing, leading to restricted participation in this important daily occupation. This can further compromise school performance, emotional well-being, attitude, and behavior (Dinehart, 2014; Feder & Majnemer, 2007; Graham et al., 2000; Peverly, 2006).
Handwriting difficulty in school-age children is one of the main reasons for referral to occupational therapy (Hoy et al., 2011). Occupational therapy practitioners, in partnership with teachers, can offer individualized instruction and evidence-based interventions to attend to the children’s needs (Case-Smith et al., 2014); however, practitioners need valid and reliable instruments to assess and interpret handwriting problems and document the efficacy of the occupational therapy intervention (Feder et al., 2000). Moreover, the adoption of a client-centered perspective (American Occupational Therapy Association, 2014) requires the use of strategies to facilitate self-assessment and collaboration on the identification of treatment goals to involve the child in the assessment and intervention process (Amundson, 2005; Graham et al., 2013).
Considering that a self-assessment could promote learning and improve a child’s motivation to collaborate, occupational therapists from Israel and the United States created the Here’s How I Write: A Child’s Self-Assessment and Goal Setting Tool (HHIW; Goldstand et al., 2013), a handwriting self- and teacher assessment tool that elicits both the student’s and the teacher’s perspective. The instrument is easy to use and understand; it consists of pictorial cards illustrating abilities related to handwriting (e.g., “I stay on the line when I write,” “I don’t stay on the line when I write”). The child chooses the picture with which he or she best identifies. The picture-choosing process helps the child identify his or her strengths and difficulties and set intervention goals with a therapist or teacher to improve handwriting. The same questions shown on the cards are presented to the teachers in a questionnaire form (Goldstand et al., 2013).
Considering that handwriting difficulties constitute a frequent problem among school-age children and have important consequences for school life, the absence of reliable handwriting assessment tools translated to and validated in Portuguese limits the practice of Brazilian occupational therapists. Given the innovative character of adopting a client-centered perspective with application to handwriting intervention, the HHIW is considered a strategic instrument to translate and adapt for use with Brazilian children. The objectives of this study, therefore, were to translate and adapt the HHIW to Brazilian Portuguese and to investigate its measurement properties, namely, construct validity and aspects of reliability, thereby adding further evidence of the tool’s reliability and validity.
To examine the tool’s construct validity, we examined whether the HHIW differentiated between children described by teachers as having good versus poor handwriting. Children with poor handwriting should score lower than children with good handwriting, in both self- and teacher ratings. To verify reliability, we examined internal consistency, the degree of agreement between child and teacher ratings, and test–retest reliability. Finally, we examined aspects of the HHIW’s clinical feasibility, taking into consideration the time spent to administer the instrument, whether the children liked the HHIW, and whether teachers found it useful. The study was conducted in two stages: translation of the HHIW to Brazilian Portuguese and analysis of reliability and construct validity.
Translation and Transcultural Adaptation
The translation process followed Beaton et al.’s (2000) recommendations for transcultural adaptation of health-related questionnaires, consisting of the following six steps:
Two translations (Translation 1 [T1] and Translation 2 [T2]) were made independently by two Brazilian English teachers, one with expertise in the health area.
T1 and T2 were compared by two bilingual occupational therapists and one translator to create a unified translated version (T3).
Two professional translators from North America who currently lived in Brazil made two back-translations from T3 (Versions 1 and 2 [V1 and V2]).
An expert committee, composed of three bilingual occupational therapists and a translator, compared V1 and V2 with the original English version of the instrument to identify inconsistencies, preparing a prefinal version.
The prefinal version was pretested with 5 children and 11 primary school teachers, who were interviewed to verify the adequacy of each item. This process resulted in minor adjustments.
The final version was applied to the target sample.
During the translation and pretest, some items were modified to adjust for semantic equivalence with word substitutions so that the item would be consistent with the language used in Brazilian schools while retaining its original meaning.
Method
Participants
Sixty typically developing children ages 8–10 yr, 38 from four public schools and 22 from three private schools, were equally divided into two groups; Group 1 consisted of children with poor handwriting (irregular letter formation, inadequate size, disorganized or illegible handwriting, and skills under grade expectations), and Group 2 consisted of children with adequate handwriting (letters well formed, organized, and legible). The groups were created by identifying pairs of children (1 with good handwriting, 1 with poor handwriting) from the same classroom, matched on age and gender. The sample size was calculated with the intention to replicate a previous validity study of the English version of the HHIW (Cermak & Bissell, 2014); however, 10 children were added to each group to balance the age distribution from public and private schools.
Instruments
HHIW.
As mentioned earlier, the HHIW is a criterion-referenced handwriting assessment that embraces a client-centered approach by incorporating the view of both the teacher and the child who is involved in a self-assessment process. The instrument consists of a children’s questionnaire, presented in the format of 25 picture cards, one with an example and 24 with the test items. One side of each card contains a statement and a picture depicting a child with a well-developed ability to perform a specific handwriting component, and the other side includes a statement and a picture that depicts a child with difficulty in the same handwriting component.
For each card, the child first chooses which side better expresses his or her performance and then classifies it as a “usually” or an “always” condition, which are scored 2 and 1, respectively, for pictures depicting difficulties and 3 and 4, respectively, for pictures depicting good performance. The cards are divided into three categories: (1) 2 affective items involve feelings related to handwriting (e.g., “I like to write”); (2) 3 items relate to physical factors, including body posture and the stabilization of the page with the other hand (e.g., “I sit straight on my chair”), and (3) 19 items illustrate handwriting performance (e.g., “I write on the line”). The HHIW includes a teacher’s questionnaire (Teacher Assessment Form) with the same items, although not in pictorial format, scored similarly to allow comparison with the child’s perspective.
The original instrument in Hebrew was translated to American English (Goldstand et al., 2013) and culturally adapted to the United States (Cermak & Bissell, 2014). In both versions the student self-assessment and the teacher ratings discriminated between children with good and children with poor handwriting, supporting the instrument’s construct validity (Cermak & Bissell, 2014; Goldstand et al., 2013). The HHIW presented acceptable measurement properties with both Israeli and North American samples.
Brazilian Economic Classification Criteria.
The Brazilian Economic Classification Criteria (CCEB; Associação Brasileira de Empresas de Pesquisa, 2018) is an economic segmentation questionnaire that uses the number of appliances (e.g., television, computer, car) and access to services (e.g., a housekeeper) at home, along with the head of household’s educational level, to infer economic levels of the population. The CCEB was created to estimate the buying power of Brazilian urban families. Each economic stratum represents an estimate of the average monthly household income, from highest to lowest (values are given in current U.S. dollars): A (USD$6,106.19), B1 (USD$2,619.24), B2 (USD$1,332.54), C1 (USD$435.61), C2 (USD$725.60), and D/E (USD$192.70).
Semistructured Questionnaire.
We created a questionnaire to collect data on the children’s developmental history; the parents’ education; who assisted the child with school homework, including handwriting; and parents’ perception regarding their child’s handwriting performance. The questionnaire was designed to be very simple, with nearly all questions responded to by checking an alternative or choosing a yes–no option. Just one question, concerning parents’ opinion regarding the importance of handwriting for their child, was scored on a 4-point scale (4 = very important, 3 = important, 2 = little importance, 1 = not important).
Procedure
The study was approved by the research ethics committee of the Universidade Federal de Minas Gerais. Parents signed an informed consent form, and children provided assent.
To select the sample, we first asked 81 students, ages 8–10 yr, in regular classrooms of four public and three private primary schools to copy a text extracted from the McMaster Handwriting Assessment Protocol–Portuguese version (Pollock & Lockhart, 2009). The text written by each child was analyzed by his or her own teacher at school and by an external examiner (an experienced primary school teacher who had no contact with the children) who rated the handwriting quality as “good,” “quite good,” “poor,” or “very poor.” The children scored as having poor or very poor handwriting by both examiners were selected for the poor handwriting group (Group 1). The children classified as having good or quite good handwriting were selected for the good handwriting group (Group 2). Group members were paired by age, gender, classroom, and economic level. Children who presented signs, as reported by teachers and parents, of cognitive deficits, genetic or neurological diseases, autism, sensory deficits (blind or deaf), or orthopedic problems that could interfere directly with the capacity to write were excluded.
From the 81 children assessed, 21 were excluded; 3 did not meet the inclusion criteria, and 18 did not have matched pairs. This exclusion resulted in 60 participants. An occupational therapist and two research assistants, trained in the HHIW procedures, completed the HHIW assessment with each child individually, in a quiet area in the school. The time spent on the interview was recorded. The Teacher Assessment Form was answered by teachers (self-application), and after completing the HHIW, children and teachers indicated whether they had liked the experience. All 60 children and their teachers completed the HHIW twice, with a 15-day interval, to examine test–retest reliability.
Data Analysis
Descriptive analysis was used to characterize the participants according to economic level, gender, and age, with handwriting group membership taken into consideration. To verify whether children in the good and poor handwriting groups presented similar characteristics, which is necessary to compare their handwriting skills, we used the nonparametric Mann–Whitney U test, because the normality of the data was not confirmed (i.e., Kolmogorov–Smirnov test); χ2 was used for categorical data. We also used Mann–Whitney U tests to verify whether there were significant differences between the groups on the HHIW scores (i.e., construct validity). As for reliability, internal consistency was investigated with Cronbach’s α, and the intraclass correlation coefficient (ICC2,1) consistency was used to verify the test–retest reliability of the children’s and the teachers’ questionnaires. The agreement between the children and the teachers was also verified (ICC2,1 absolute agreement).
For all analyses, we considered only the scores obtained in the first administration of the HHIW; the second administration was used only for the purpose of determining test–retest reliability. The strength of the reliability values was interpreted according to a scale proposed by Tyson and Brown (2014): weak, <.04; moderate, .04–.06; good, .06–.08; and excellent, >.08. For all analyses, the significance level was set as α < .05.
Results
Data on sample characteristics obtained through the CCEB and the semistructured questionnaire are presented in Table 1. The sample included significantly more boys than girls in both groups, and the poor handwriting group had a higher percentage of children with problems at school, but we found no significant differences between the poor and good handwriting groups in children’s age (p = .69) or in fathers’ (p = .184) or mothers’ (p = .336) education level. Most parents had a high school or college education, and the groups also did not differ in regard to the distribution of economic class (p = .092); thus, this was a typical sample from public and private schools in big urban centers of Brazil, with a higher distribution in economic Levels B and C than in Level A. More children in the poor handwriting group used medications, and their parents reported more school-related problems. There was no difference between groups in the number of children in private therapy (i.e., speech therapy, occupational therapy, or psychology). Parents were generally aware of their child’s handwriting performance; the classifications of 73% of the parents of children in both the poor and good handwriting groups matched the categories of handwriting quality defined by the teachers. For both groups, 76.6%–80.0% of mothers helped with homework, and 100% of parents felt handwriting was important or very important.
Characteristics of the Good and Poor Handwriting Groups
Note. Dashes indicate that no children in the good handwriting group used medication, so it was not possible to calculate the difference. CCEB = Brazilian Economic Classification Criteria; M = mean; Mdn = median; SD = standard deviation.
Mann–Whitney U test. bPaired condition, no analysis required. cChi-square. dPrivate occupational therapy, speech therapy, or psychology. eAverage monthly income: A = USD$6,106.19; B = USD$2,619.24–1,332.54; C = USD$435.61–725.60.
Construct Validity
In regard to group comparisons, as shown in Table 2, there were significant differences between the handwriting groups on the Total score of the HHIW and its categories, both for the children’s self-report and the Teacher Assessment Form. In all instances, teachers of children with poor handwriting rated their handwriting more poorly than that of good writers.
Mean Total Score and Affect, Performance, and Physical Factor Scores for the Good and Poor Handwriting Groups
Note. M = mean; Mdn = median; SD = standard deviation.
Reliability
Internal consistency (Cronbach’s α) was excellent for both the children’s self-report (.915) and the Teacher Assessment Form (.953). Table 3 shows the agreement between children and teachers as well as test–retest reliability values for Total HHIW score, categories, and individual item scores. The degree of agreement was good for the Total score and categories; for Physical Factors, however, agreement was weak but still significant. Among the HHIW’s 24 items, agreement between the teacher’s and child’s responses yielded significant ICCs for 17 (70.83%) items, with good agreement for 9 items, moderate agreement for 6 items, and weak agreement for 9 items. Test–retest reliability was excellent for both child (ICC = .96, p = .0001) and teacher (ICC = .930, p = .0001) Total scores, as well as for the categories, and excellent to good for the majority of the individual items.
Total Score and Individual Item Reliability Indexes (ICCs) for Child–Teacher Agreement and Test–Retest Reliability for the Child and Teacher Questionnaires
Note. HHIW = Here’s How I Write: A Child’s Self-Assessment and Goal Setting Tool; ICC = intraclass correlation coefficient.
p < .05. **p < .001.
Clinical Feasibility
Completing the HHIW with the child took 14 (±2.9) min; the time spent by the teachers was not measured. The instrument was well accepted by the sample; 58 children (96.7%) answered, “I like it” when asked to describe their experience with the instrument (response options were “I liked it,” “I kind of liked it,” or “I did not like it”). The teachers responded to three questions about the importance of the instrument; they all indicated the instrument was important to evaluate handwriting.
Discussion
The HHIW was transculturally adapted and used in a sample of Brazilian students as the first step of its validation process in a new context. Participants were grouped into those with good and those with poor handwriting on the basis of ratings by both an external rater and the child’s teacher. No significant differences were found between groups in age, gender, economic status, or parental education. There was a higher proportion of boys with poor handwriting, which reflects the prevalence of handwriting problems reported in other studies (Case-Smith, 2002; Duiser et al., 2014; Overvelde & Hulstijn, 2011). Note that in the current study, even though some children demonstrated poor handwriting that was recognized by the teachers and most parents, they did not necessarily have clinical or academic problems, given that only 9 (30%) of children in this group were reported as presenting school difficulties (see Table 1).
The HHIW is useful to compare children’s and teachers’ perceptions of the children’s handwriting. Children rated by their teachers as having handwriting problems rated themselves more poorly than children who were good writers, indicating that Brazilian children ages 8–10 yr are capable of evaluating their own handwriting performance. This finding supports the construct validity of the HHIW as a handwriting assessment tool and is consistent with Cermak and Bissell's (2014) findings with a sample of American children. The test–retest reliability of the child’s self-rating and the Teacher Assessment Form was excellent for the Total score and categories and good to excellent for individual items, which suggests that the items are clearly written and that students and teachers are able to respond in a trustworthy way.
The good child–teacher agreement on the Total score as well as on the Affect and Performance categories of the HHIW indicate that children and teachers shared a perspective on various aspects of handwriting assessment. There was, however, less agreement concerning the Physical Factors category, which may be explained by the fact that details related to posture and the way the child supports the paper may not be the focus of the teacher’s attention; however, the picture cards provide the children with specific examples that might remind them of their own issues.
Lack of teacher–child agreement on some individual items might relate to the finding that children from both groups tended to rate themselves a little bit higher on all score categories than did their teachers (see Table 2). This discrepancy was also seen in the U.S. sample, particularly for poor writers (Cermak & Bissell, 2014), who scored themselves significantly higher than did their teachers. These children may be aware of their problems but may not know the level of impairment, or they may be reluctant to score themselves in the lowest portion of the rating scale. Considering other items that had weak teacher–child agreement, children might also have difficulty judging details such as whether they write enough on assignments (Item 3), start letters in the correct direction (Item 5), or keep the letter size consistent (Item 6), but they did agree on most items about legibility and spatial organization.
Limitations and Future Research
Even though the participants were recruited in public and private schools, they do not represent the whole Brazilian population. Considering that school-based occupational therapy services are not a reality in Brazil, children in the poor handwriting group did not constitute a clinical sample to be referred for intervention. Further studies should include children recruited from clinical settings or referred for therapy because they may present different ability to appraise their handwriting skills.
Our sample was recruited on the basis of a combination of the judgments of the child’s teacher and another examiner. The teachers could have been biased when completing the handwriting assessment, being stricter in their assessment of children with poor handwriting, which may have contributed to poor teacher–child agreement on some individual items. The rigor of the group selection process, however, relied on the assistance of the teachers.
In this study, the goal-setting part of the HHIW and the intervention procedures were not performed because the emphasis was on verifying the adequacy of the instrument for the Brazilian context. Future studies should not only examine the adequacy of the goals defined with the HHIW but also try to determine whether the self-assessment really contributes to higher engagement of the child in the intervention process. Further studies should investigate whether children’s and teachers’ perceptions, as expressed by scores on the HHIW, predict performance on standardized norm-referenced measures of handwriting.
Implications for Occupational Therapy Practice
The results of this study have several implications for occupational therapy practice:
The HHIW is a new assessment tool that combines the voices of the teacher and the student to evaluate handwriting ability. Considering the absence of specific handwriting assessment tools in Brazil, the HHIW–Brazil will enable occupational therapists to set client-centered goals and evaluate progress according to the child’s and teacher’s perceptions.
This study adds further evidence that the HHIW is clinically feasible; it differentiates children with good and poor handwriting and seems to facilitate children’s self-appraisal, which is important for engagement in intervention.
The HHIW provides information on a wide range of handwriting components, and it involves the child in the assessment process, encouraging him or her to become more aware of the quality of his or her own handwriting.
The instrument also includes teachers’ perceptions, which can lead to intervention goals that are consistent with the school context.
Conclusion
The HHIW showed excellent test–retest reliability, and there was good to moderate agreement between the child and teacher scores. The children liked the HHIW and were capable of reliable self-assessment. Significant differences between the scores of good and poor handwriters add further support to the validity of the instrument. The HHIW–Brazil can be a useful addition to therapists’ assessment tools, providing a means to include the child’s voice in the assessment process. The clinical utility of the goal-setting aspect of the instrument, as well as the instrument’s validity compared with other measures of handwriting performance, deserves further investigation.
Footnotes
Acknowledgments
We thank the Brazilian National Council of Research for supporting Lívia de Castro Magalhães’s research career. This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior–Brasil, Finance Code 001.