The development of a handwriting screening assessment for academic accommodations at the University of Witwatersrand

Abstract

A small percentage of students at the university are academically compromised by their handwriting. Various components of handwriting and performance skills have been associated with dysgraphia and inefficient handwriting such as posture and the presentaion of handwriting in terms of corrections made, which are not include in handwriting assessment for students in higher education. The current study addressed the development of a new screening assessment to be used in awarding concessions for examinations to university students with dysgraphia or handwriting deficits which therefore evaluated observable motor and process performance skills related to handwriting in three sections, an Observation Checklist, a Writing Checklist and for Handwriting Outcomes (copying speed, legibility and automaticity of writing). The study was completed in three phases with the first phase addressing a pilot study on the development of the Handwriting Screening Assessment based on steps in instrument development and criteria for screening assessment development. Item validity was established using a review of the records of 287 students who had been referred for handwriting assessment. The Handwriting Screening Assessment was piloted for content validity and item and subtest validity as well as dimensionality using Rasch subtest analysis after adjustments to items on both checklists. Construct validity of the items on the three sections of the Handwriting Screening Assessment and the unidimensionality of the checklists were considered satisfactory for field testing with typical students and those referred for handwriting assessment in Phase 2. In the second phase the Handwriting Screening Assessment was tested for construct validity and reliability on a sample of 298 typical students and 61 students referred for assessment of handwriting or dysgraphia. Construct validity of the items and subtests were confirmed for this sample of students using Rasch analysis for the checklists. Differences for known group factors and between the two groups of students indicated construct validity and reliability were satisfactory although not all subtests differentiated between the typical students and the students referred for handwriting assessment. v The Rasch subtest analysis resulted in low person separation index scores which did not allow for students to be identified for different levels of risk for dysgraphia or handwriting deficits using the scores on the Observation and Writing Checklists. A similar result was found for the Handwriting Outcomes. This was due to individual differences and not all students presenting with deficits in all the subtests of the three sections of the Handwriting Screening Assessment. Therefore normative scoring cut-off points and “at risk quotients” (ARQS) were established for the each subtest so students’ level of risk for handwriting deficits or dysgraphia could be identified. Significant differences between the typical students and the students referred for handwriting assessment were found for the three sections of the Handwriting Screening Assessment confirming satisfactory construct validity based on the ARQs. The clinical accuracy of the Handwriting Screening Assessment assessed on the ARQs indicated adequate negative predictive values for all sections and adequate specificity for all sections except legibility. While the assessment eliminated those without handwriting deficits and dysgraphia the low sensitivity meant that some students with handwriting problems may be missed. The Handwriting Outcomes - copying speed and automaticity were convergent with reference assessments of handwriting speed and oculomotor dysfunction, Detailed Assessment of Handwriting Speed 17+ and the Developmental Eye Movement, confirming the validity of this subtest in the Handwriting Screening Assessment. All other subtests had divergent validity with the reference assessments indicating they assessed different components related to handwriting problems not usually assessed in students in higher education which were found to identify them at risk for handwriting deficts and dysgraphia. The usability and utility of the Handwriting Screening Assessment was established in Phase 3 of the study. A detailed analysis of the results for the students referred for assessment of handwriting dysfunction was completed to inform the usability in terms of interpretability of the screening assessment and guidelines for further assessments. The profile of the students referred for handwriting assessment and demographic factors and items on the Handwriting Screening Assessment that placed them at risk for dysgraphia or handwriting deficits were determined. These results indicated that the subtest for pen grasp should be discarded but that other subtests which did not differentiate the students referred for handwriting assessment from typical students should be retained as they were moderately or strongly correlated with the risk for dysgraphia. The utility of the Handwriting Screening Assessment in terms of the types of dysgraphia to guide concessions that should be awarded and the benefit of the assessment in terms of academic outcomes were analysed. The Handwriting Screening Assessment can be used to identify students in higher education at risk for dysgraphia handwriting deficts and to suggest further assessment and guide concessions required but the validity can be improved with further adjustment and revision of items and scoring