3(1) p. 43

Empirical Research in Word-Processing: Expectations vs. Experience

Deborah H. Holdstein and Tim Redman

Will any word processing software do for writing students? In determining the overall efficacy of the computer in writing instruction, we have learned several things: that some software used for word-processing can actually hinder the writing/rewriting processes, and that writing specialists' input is an essential component of decisions about hardware and software made on administrative levels.


Results of research on the use of computers in writing have (for the most part) emphasized the positive aspects of their implementation: one important study completed by Mimi Schwartz at Stockton State (1984) delineates her great success with several groups of writers who used a word-processor for composing. In another study, Collette Daiute's preliminary research at Columbia University suggested that word-processing improves the quantity of writing, the number of revisions, and the length of the manuscripts written by children. These results

p. 44

are, for the most part, typical. They substantiate our own expectations when we initiate research projects in hopes of substantiating our confidence in CAI and word-processing. They support our hopes of convincing recalcitrant administrators to loosen their budget lines.

Although Schwartz and others have begun to note the mechanical and practical difficulties students encounter when using the computer for writing, one researcher, emblematic of many, maintains that, "most obstacles can be overcome" (Hermann, 1983, p. 6). And in fact, one dismayed reviewer of this year's proposals for computer sessions at the Conference on College Composition and Communication noted a deluge of proposals singing only statistically significant praises for student writers' use of the computer. Although graduate researcher Andrea Hermann rightly warns that we "temper our enthusiasm with caution" (Hermann, 1983, p. 7), other literature thus far hasn't reflected such caution. In the experiment we describe in this report, we had hoped to be guilty of the same optimistic statements. This did not prove to be the case. However, we now realize that our mostly negative experience may be useful not only for research but also for making practical recommendations--particularly when we must decide what word-processing software to use, or more importantly, when we are faced with the choice of one somewhat undesirable package or nothing at all.


As a private university with a technological orientation, the Illinois Institute of Technology

p. 45

(IIT) has invested its limited funds in the best possible mainframe computer, a VAX that allows a great variety of applications. At the time of our study, there were a limited number of microcomputers available for general student use. (In the fall semester of 1984, the University opened an IBM-PC classroom.) We wanted to expose our beginning English composition students to the use of computers as an aid to revision, hoping that their technical background and technological orientation would make up for the less user-friendly aspects of the VAX system's formatting program, RUNOFF. ITT faculty members have regularly and successfully used RUNOFF to meet their own word-processing needs.

With terminals scattered all over campus, we hoped that our writing students would have no difficulty in adapting easily to the word-processing system and that their using the computer would encourage and aid them in revising papers. As you can imagine, students who view themselves as "technical" tend to buy the myth that their interests and the interests of good writing are incompatible. We hoped that putting the unfamiliar (writing) into a familiar context (computing) would help.

For the investigation, we used two classes of English 101: one with about 25 students, the other with 12. The latter were given computer accounts and told to use the computer for every aspect of one assignment. The control group were given no special instructions. So that students in the experimental section wouldn't have to learn writing and the VAX system at once, the unit was taught after the midpoint of the spring semester. In this way, we hoped to avoid the

p. 46

"double bind" effect, as described by Lillian Bridwell and Don Ross (1984, p. 112). The students were asked to do a division and classification paper about their fellow students. They were asked to bring five copies of the paper with them on the day it was due; each class was to be divided into peer editorial groups to comment upon the papers. Each paper was to be revised for a final grade and handed in one week later.

Our structure was simple and predictable: the experimental group was instructed for one period in the use of the VAX system's RUNOFF formatting program. This group was given a week before the papers were due. They were told to do all their work for the essay on the VAX system. The faculty member who instructed them was then available for consultation with students on an individual basis and, in fact, spent a good deal of time with students individually.

In order to record any changes in the students' writing process over the course of the semester, we gave them copies of a mimeographed process diary to fill out each time they did any writing. On the day the essay was due, the students went into groups of four or five each. They received a handout on editing for revision (designed by the instructor) and were told to read and discuss each paper for at least fifteen minutes. This review process continued during the next class day. The students were then asked to revise their own papers on the basis of their group's discussion and to turn them in after the weekend.

We were hoping that the students would find the benefits of easy revision outweighing the disadvantages of having to use the VAX. We did

p. 47

not "warn" them that we considered the VAX RUNOFF program somewhat cumbersome to use (one of us regularly uses RUNOFF, and that is our opinion). The faculty member teaching the students was, and continues to be, a committed and very enthusiastic champion of RUNOFF for effective word-processing and editing.

Because we weren't sure what were looking for, except perhaps to find that the VAX could reliably substitute for other word processors, our other procedures were open ended. We gave short-answer questionnaires to both groups at the opening and close of the unit. These were to assess students' attitudes towards word-processing, writing in general, and the assignments. Additionally, the experimental group was given a similarly formatted "midstream" questionnaire to fill out anonymously after the revision of the division and classification paper.

The questionnaire filled out by the group using the VAX yielded the more interesting results. At the beginning of the investigation, a number of the students showed positive expectations about the prospect of using the computer for their papers. This response may have been due to the technical orientation of IIT students; in any case, we had no reason to anticipate any problems with the unit, because we and the students approached the prospect of "computing and writing" with enthusiasm.

By the midstream questionnaire, that optimism had changed decidedly. The students' initial interest in the project had turned to dissatisfaction. One student was particularly eloquent:

p. 48

I tried to use the VAX as instructed, but I had a "massive mental block" because the terminal room was too noisy, and I was unfamiliar with the VAX. Now by midsemester I feel mechanically comfortable with the VAX, but it will take longer to be able to write original thoughts and structure a paper. The problem with the VAX is that you can't see the whole paper when writing. I like to go over my papers as I write, to keep from diverging, to continue the flow. I do not think it made my writing any better.

We wish this had been a isolated case; regrettably, the students had many difficulties with the VAX. Quite varied, these difficulties included problems with log on procedures, getting printouts, scheduling (especially for commuter students), instructions, inexplicable stoppages, system failure, and general unpredictability. (One student plaintively wrote, "It would just cut off the ends of sentences, and I don't know why.")

There were a lot of problems for the dozen students in the experimental group, and we are tempted to conclude that our new generation of computers may do nothing more effectively than produce a new generation of excuses--fare exceeding in ingenuity and complexity the Gutenberg-era litany of dog's appetites, grandmothers' health, and cars' mechanical failures. Just to keep up with the latest high-tech excuses, the average teacher needs to be familiar with microprocessor jargon.

Conversations with students in the experimental group and comments from their

p. 49

process diaries revealed similar problems. Although not all were negative, typical comments included the following:

These problems may explain, at least in part, why the experimental group, using the computer, did somewhat worse on the graded revisions than did the control group.

Surprisingly, the final questionnaire contained more of an even mixture of positive and negative comments about the use of the VAX for word processing than either of the previous two questionnaires. Several students clearly had become comfortable with the system; in fact, three voluntarily used it for another paper due at the end of the semester.

Most interesting to us were the frequent questionnaires that contained outright contradictions. To cite a few examples, students would say both "I found it easy to use and fun," and "I had problems with responsiveness, limits on what would be displayed, and access"; or "the VAX made things easier," and "I will not use it again"; or "revising was made easier by the use of the computer," but "I still use my typewriter for final drafts." These anomalies reveal how

p. 50

expectations in a computer age may override experience. They also suggest the need for caution in future work. As one student, who admitted that he never became comfortable with the VAX, put it: "the assignment is a good idea because everything is going computers anyway; why not get accustomed to them?"

One final result is worth noting. Contrary to our expectations, students did not increase the number of times they revised their papers over the course of the unit, or after. The number of revisions reported on the pre- and post-questionnaires for both control and experimental groups was almost the same.

One problem that we did anticipate when planning this investigation was the fact that the unit was too short and too late in the semester for real benefits to appear. One problem that we did not anticipate is that it takes a fair amount of time to become thoroughly familiar with the RUNOFF formatting program; the payoff for a one-semester course is simply not there. This fact suggests an obstacle for teachers who are committed to the idea of word-processing but who are limited to whatever system their university happens to have.

We recommend that those instructors either build in plenty of time for preparation, including individual attention at the terminal, or decide that the formatting aspects of the program are not that important and just teach students to write and revise without using commands such as ".papa; .center; s/ /," etc. (RUNOFF is not a "what you see is what you get" program.) Teachers should also encourage students to save or print their text after each section

p. 51

they write. Remember, however, that we had allowed for some of these things; RUNOFF was just far too cumbersome, at least in this writing context.

We believe that the use of computers should encourage in students the idea that writing is play. Getting students started on their first drafts, no matter how crude, will help them deal with the composition process in a less anxiety-producing situation. No matter what word-processing program the teacher is using, the key is to get the first draft flowing, to get students started so they do not worry that every mark they make on a paper is irrevocable and final.


What seems to be a first generation of advice, research, and interim reports is upon us--we've all learned the criteria for selecting software, for learning to love the computer, for integrating word-processing within the curriculum, for developing software ourselves. In fact, it almost seems outre to report yet another study of word-processing in the writing curriculum; but where are the other tales of failed attempts and frustrated instructors and students? Do we want to end up like the English instructor who heard from one student, "I liked your writing course because it taught me how to work an Apple"? What lessons can we learn when we forge ahead, blinded with our enthusiasm for word-processing, only to find that software selection may not be an option to those of us at real universities with real budget constraints?

p. 52

We have learned the following: that some accepted criteria for software selection don't always work; and that your own ease and your colleagues' ease with certain word-processing systems isn't necessarily an appropriate benchmark.

Yes, software selection is important--but not always for the reasons that we think. We can learn to live within the limits of any word-processing/composing/formatting program as long as we keep certain first principles in mind: one, that we do not mistake the easy changes possible on a word-processor for real revision of an idea or an essay; and two, that we encourage our students to treat these machines and their expensive accoutrements as carelessly and playfully as we hope they do a pen and a piece of paper. Most significantly, not just any word-processing software will do. We've been warned against such fatal flaws as no "word wrap," inappropriate word breaks, or hard-to-adjust margins. But we must learn to compare those software flaws that are external to the writing process with software flaws that actually interrupt the composing process--not necessarily ours, but our students'.

Of the various data we gathered, two trends emerged that particularly interested us. First, despite what we've heard about word-processing making it easier for students to revise, the students using the VAX reported almost no change whatsoever in their revising habits. The minuscule change that did occur was toward fewer revisions. Second, our students initially felt that using the VAX would be an important and positive experience; and despite

p. 53

the many problems they documented for us throughout their word-processing experiences, their final questionnaires still contained evidence of that initial enthusiasm even to the point of contradicting what they were saying in other parts of the questionnaire. This may mean that our students were merely trying to be polite towards our research. It may also indicate, as we believe it does, that the notion that computers are the wave of the future has been so drummed into students' heads that they continue to repeat platitudes about the value of their computer experience even while writing of the many problems and difficulties they encountered.

Our students' responses indicate a significant danger zone; we must build in ways of sorting out this form of computer piety from varieties of methods that actually work for an individual student--in both research and pedagogy. In short, whether dealing with administrators who force their word-processing investments upon us or with the preferences of other well-meaning colleagues, we must heed these rather significant warnings and channel them into practice.


Bridwell, L., & Ross, D (1984). Integrating computers into a writing curriculum: buying, begging, and building. In W. Wresch (Ed.), The computer in composition instruction (pp. 107-119). Urbana: NCTE.

p. 54

Schwartz, M. (1984, March). Conference on College Composition and Communication, Detroit.

Hermann, A. (1983, July). Using the computer as writing teacher: The heart of the great debates. Proceedings, University of Oregon.

Deborah H. Holdstein teaches at Illinois Institute of Technology. Tim Redman teaches at University of Wisconsin-Parkside.