The influence of student experience with word processors on the quality of essays written for a direct writing assessment

Abstract

Although prior research has identified general procedural and qualitative differences between word-processed and pen and paper writing, little attention has been directed toward identifying how these differences relate to the prior word processing experiences of individual students. Additionally, few researchers have addressed the issue of fairness when discussing the use of word processors in writing assessment. This study investigates this relationship by comparing essays composed with pen and paper for a direct writing assessment to those composed with a word processor by students having different levels of experience with using word processors for writing. We observed differences between the two composition media similar to differences observed in previous studies of word-processed writing. That is, our results show that, overall, word-processed essays are neater and longer than were pen and paper essays. Word-processed essays also have a more formal tone and a weaker voice than their pen and paper counterparts. No composition medium differences were observed for the number of mechanical errors. In terms of how word processor experience interacts with writing quality, we found that word processors use neither improved nor worsen the quality of essays produced by students who have medium to high levels of experience using computers for writing. On the other hand, the word processor essays produced by students with a low level of experience writing with computers were scored, on a six-point scale, almost an entire point lower than those produced with pen and paper by these same students. Groups with high and medium levels of experience with word processors wrote slightly more words with a word processor than with pen and paper. On the other hand, the group with a low level of experience with using computers for writing wrote over 100 words fewer on word processors than with pen and paper. As for the number of simple sentences, groups with high and medium levels of comfort and experience with computers for writing wrote fewer simple sentences with a word processor than with pen and paper, while the group with a low level of comfort and experience with computers for writing wrote more simple sentences with word processors than with pen and paper. There were no group by media interactions for the number of mechanical errors.