Impact of Text-to-Speech Software on Access to Print: A Longitudinal Study

Introduction

In recent years there had been increased interest in the evaluation of the implementation and effectiveness of assistive technology (AT). Particularly, in these times of fiscal restraint educators are becoming acutely aware of the need to substantiate effective outcomes and wise allocation of resources. The 1997 and 2004 reauthorizations of Individuals with Disabilities Education Act (IDEA) mandated that AT devices and services be considered for each student with a disability when developing an Individualized Education Plan. The inclusion of special education student achievement within the accountability of the No Child Left Behind (NCLB) Act of 2001 raised the level of concern. Now educators are required to use evidence-based interventions with proven effectiveness. Despite these mandates, only limited research has been conducted on the effectiveness of AT to improve student achievement outcomes (Edyburn, 2003, 2007a) with contradictory results (Sorrell, Bell, & McCallum, 2007; Strangman & Dalton, 2005).

In a meta-analysis of 68 studies Alper and Raharinirina (2006) identified only 20 that evaluated the effectiveness of AT on the basic academic skills of reading, math, spelling, and writing. Twelve of those studies addressed the content areas of math, spelling, or writing. Eight of the studies investigated reading skills. The results supported improvement in skills such as comprehension, decoding, and fluency. Balajthy (2005) reviewed the impact of text-to-speech (TTS) software with struggling readers reporting mixed results across various populations from improvements in comprehension depending on student ability level to poorer results for better readers. Strangman and Hall (2003) identified 13 studies related to the effectiveness of TTS software. However, across all these studies it was difficult to draw firm conclusions due to the diversity of disabilities, age span, technology devices, and lack of replication. Also lacking were data on the effects of attitudes and preferences on the integrity of implementation of AT (Alper & Raharinirina, 2006; Smith, 2000).

In response to the need for more research-based interventions, there are an increasing number of studies of AT outcome measures being generated. In an action-based study, Dimmitt, Hodapp, Judas, Munn, and Rachow (2006) assessed the impact of the use of a TTS software program (Kurzweil 3000) on the reading skills of 73 middle school students on outcome measures of reading fluency and passage comprehension. The average reading rates improved by 16 words per minute in 23 weeks, which was 2.3 times faster than would be predicted by research on students with special needs (Fuchs, Fuchs, Walz, & Germann, 1993). The data also indicated a positive trend in the comprehension scores. The average comprehension score improved by 13 percent per student from 59 to 72 percent. Data showed it took 13 weeks for students’ comprehension scores on passages accessed by TTS software to exceed comprehension on print materials. The results demonstrate that the accommodation helps compensate for student reading deficits. Responses to online surveys demonstrated that students and teachers associated the use of the TTS software with improved academic performance, better on-task behavior, more engagement with the instructional material, and improved independent work completion. The study results relate to Parette, Peterson-Karlan, Wojcik, and Bardi's (2007) discussion of the compensatory versus remedial function AT.

Lance, McPhillips, Mulhern, and Wylie (2006) compared the performance of three groups (Read and Write Gold, Microsoft Word, and control groups) on literacy tests. After six training sessions of 45 minutes each, the AT group (Read and Write Gold) showed improvement on reading comprehension, homophone error detection, spelling error detection, and word meanings. The Microsoft group showed improvement on spelling error detection and word meanings with a poorer performance on homophone error detection. Meanwhile, the control group showed no improvement on any of these measures.