Quinlan, T. (2004). Speech recognition technology and students with writing difficulties: Improving fluency . Journal of Educational Psychology , 96 (2), 337–346. https://doi.org/10.1037/0022-0663.96.2.337

A total of 41 children participated in this study. Participants were aged 11 to 14 and were recruited via flyers at public and private schools. Location was not specified. Children who responded to the flyers completed an initial screening session to gauge their overall writing and oral fluency. Participants met criteria for one of two groups: less fluent and fluent writers.

A total of 41 children participated in this study. Participants were aged 11 to 14 and were recruited via flyers at public and private schools. Location was not specified. Children who responded to the flyers completed an initial screening session to gauge their overall writing and oral fluency. Participants met criteria for one of two groups: less fluent and fluent writers.

Students participated in a writing class where they learned how to use SR and how to use a graphic organizer to plan their stories. Students then composed narratives in four conditions: via SR, via handwriting, with advance planning, and without advance planning. Participants had 10 minutes to write their narrative in each condition. Under advance planning conditions, participants were given five minutes to plan their narratives. Researchers assessed SR competency based on the percentage of words correctly recognized by the SR system. Narratives were analyzed to determine writing fluency, global text development, syntactic complexity, and surface errors.

For the group of participants who were less fluent writers, SR both increased the length of writing and decreased surface errors present in the narratives. SR, however, did not significantly improve the holistic quality of writing. Advance planning supported participants in composing more developed stories.