Profet, A New Generation of Word Prediction: An Evaluation Study

Writing Development in Struggling Learners, 2017

South African Journal of Education

Children with spelling difficulties are limited in their participation in all written school activities. We aimed to investigate the influence of word-prediction as a tool on spelling accuracy and typing speed. To this end, we selected 80 Grade 4 -6 children with spelling difficulties in a school for special needs to participate in a research project involving a cross-over within-subject design. The research task took the form of entering 30 words through an on-screen keyboard, with and without the use of word-prediction software. The Graded Word Spelling Test served to investigate whether there was a relationship between the children's current spelling knowledge and word-prediction efficacy. The results indicated an increase in spelling accuracy with the use of word-prediction, but at the cost of time and the tendency to use word approximations, and no significant relationship between spelling knowledge and word-prediction efficacy.

Routledge Handbook of the English Writing System., 2016

In previous chapters we saw how children learn to read in English, and the difficulties they may encounter on the way, thanks to the complex etymological origin of many of the English words. Because many visual word forms in English do not translate into their constituent sounds, in the way that individual letters do, learning to read in English normally takes longer than in other languages. When children struggle with learning how to read, we assess different aspects of their language abilities, in order to diagnose the cause of the problem, and to provide the appropriate remediation, or techniques to cope with the problem. The most common cause of reading difficulty in English--speaking children and adults is dyslexia.

The 24th International ACM SIGACCESS Conference on Computers and Accessibility

Figure 1: The LaMPost interface. The system augments a typical browser-based email editor with three AI-powered features: (left) users can generate an outline of the email's main ideas (a) with the option for a related subject line (b); (center) users can generate suggestions for possible changes to a selected passage, and (right) users can generate rewritten text for a selected passage based on a human-or machine-written instruction.

American Journal of Occupational Therapy, 2003

T he primary role of the school-aged child is that of the student. Within that role one of the key educational activities is the production of written work. Evidence suggests that, in elementary school classrooms, students spend an average of 43% of their day doing fine motor tasks and 85% of those tasks are paper and pencil activities (McHale & Cermak, 1992). Legible handwriting is necessary for children to carry out many academic activities (Amundson & Weil, 1996) and difficulties with handwriting can interfere with related writing processes such as planning and generating ideas (Graham, 1992). According to Benbow, Hanft, and Marsh (1992), "one of the most serious effects of poor handwriting occurs when the quality of handwriting detracts from the student's ability to convey information and ideas" (p. 33). Children with handwriting problems are frequently referred to occupational therapists (Benbow, 1995; Oliver, 1990). The focus for occupational therapy intervention with children who have handwriting difficulties is to assist the child in developing functional writing skills (Amundson & Weil, 1996). When the child's handwriting is so poor that the child is unable to meet the demands for written communication in his or her educational setting, strategies that are able to offset the functional delay need to be considered (

Learning Disabilities: A Multidisciplinary Journal, 2018

Participants in this study completed an on-line experiment in which they wrote essays by stylus or keyboard. Three translation measures (length of language burst, length of pauses, and rate of pausing) and four transcription measures (total words, total time, words/minute, and percent spelling errors) for composition were analyzed for two research aims. Research Aim 1 addressed whether upper elementary and middle school students with carefully diagnosed transcription disabilities (dysgraphia with impaired handwriting, n=18, or dyslexia with impaired spelling, n=20) showed significant differences from pretest to posttest, across modes of transcription (stylus or keyboard), and between diagnostic groups. Results showed significant (a) change after intervention (18 computerized lessons with learning activities in letter formation/selection, spelling, and composing) in length of pauses, total time, and words per minute; (b) mode effects (fewer words and less time by stylus; fewer pauses per minute by keyboard); and (c) interactions with diagnostic group in response to intervention on some measures. Research Aim 2 addressed whether following intervention each of the diagnostic groups performed comparably to a typical control group (n=15) in the same on-line experiment. Results showed (a) comparable performance of the dysgraphia and control groups on all keyboarding tasks but differences on two stylus measures; and (b) lack of comparable performance of the dyslexia and control groups on two stylus measures (total words and percent spelling errors) and the four keyboarding tasks related to transcription. Implications for assistive technology and writing instruction for dysgraphia and dyslexia are discussed.

Among the techniques involved in Augmentative and Alternative Communication, prediction or anticipation is a method that tries to anticipate what a user (normally a person with disability) is going to write next, in order to enhance his or her communication rate. There is a vast amount of possible factors that may influence the results that a predictive system may achieve. These factors can also condition the acceptation of the system by the users. In this paper, the interface-related features are reviewed and their influence in the performance and usability of the system is discussed.

Journal of Special Education Technology, 2010

A multiple-baseline design across subjects was used to investigate the effects of word prediction and text-to-speech alone and in combination on four narrative composition-writing skills (writing fluency, syntax, spelling accuracy, and overall organization) of six fifth-grade Hispanic boys with specific learning disabilities (SLD). Participants were divided randomly and equally into Cohorts A and B. During baseline, both cohorts wrote for 15-minute sessions using word processing only. During intervention, Cohort A participants used word prediction, and then word prediction with text-to-speech. Concurrently, Cohort B participants used text-to-speech followed by text-to-speech with word prediction. The results indicated that word prediction alone or in combination with text-to-speech had a positive effect on the narrative composition-writing skills of the targeted students. With text-to-speech alone, inconsequential results were observed.

Cervical spinal cord injury, text input speed, word prediction software, frequency of use, computer 3 Abstract

Proceedings of the fifth …, 2002

In the absence of a full understanding of the nature of dyslexia, the researchers' approach was to identify some of the most commonly noted problems which dyslexics encounter when reading and producing text. On the basis of these common difficulties, they identified ways in which each individual might be able to minimise the consequences of their own particular problems by manipulating the appearance of their word processing environment and text presented within it. This stage of the development process involved the start of an iterative cycle of user-centred prototype development and evaluation with dyslexic computer users. The work in progress ultimately led to a software system which provides a highly (and easily) configurable environment for dyslexic people to use for reading and producing text. The researchers' approach has been to examine the parameters of the situation, including cost, cost benefit, existing software and user demand, with a view to finding an optimal path to the production of a software system which is of real use in practical situations. The general approach has led not only to a useful product but also to some promising avenues for future development, a few surprises and has identified those ideas which proved to be less promising than at first thought.