This is the second part of my dissertation for the CALL option component of my MA in Applied Linguistics at Essex University in May 1989, an experimental study of using a basic French word processor in a secondary school.

First Part

The aim of the study was to observe the reaction of English mother tongue school learners of French to performing a text editing exercise in French using a simple word processing package. Moreover it was an attempt to identify any differences that might arise between the performance of the activity on the word processor and the same activity with pen and paper.

Following the example of Windeatt (in Leech and Candlin 1986) both a pre-task and a post-task questionnaire was completed by the students (see appendix) in order to determine in what ways, if any, the experience of working with the word processor in this context affected the students' attitudes to various aspects of their learning of French at school in particular to writing in French.

THE SUBJECTS

The subjects involved were a group of six third-year learners of French taken at random from a top stream class of a comprehensive school (Thomas Lord Audley School, Colchester). All six were girls, but it was not a deliberate decision to have an all female group. Practical constraints of time and internal administration (timetables, availability of hardware, etc.) under which the study was carried out meant that it was not feasible to undertake a more selective group composition or to work with more than this one group.

Thus, although the composition of the group remained stable throughout the study, this was considered consistent with the limited scale of the study. Although it could be argued that the findings may have been compromised by the failure to make more deliberate efforts at obtaining a more 'representative' set of subjects, the reply must be that any limitations to this study could be taken account of in future attempts at larger scale, more longitudinal studies in this area. Although all subjects had used a computer before, either at home or at school, this was their first experience of working with a word processor in a language learning context.

THE SITUATION

Throughout the course of the study, every possible assistance was proffered by an enthusiastic and sympathetic, yet hard pressed teaching staff coping with the day-to-day pressures of working in a large comprehensive school. The Language Department of the school has no computer hardware or software resources of its own despite the campaigning efforts of the staff. The hardware was borrowed from the Special Needs and Maths departments and the Language Department's seminar room was temporarily given over to the needs of this study, although this sometimes clashed with the various other demands of a busy schedule, in particular those of examinations which were being conducted at the same time as well as those of a parallel study being conducted by a colleague. The software, "European Folio" (Tedimen Software), was loaned by the Centre for the Teaching of Reading and Language Development, Clacton, who also provided valuable technical support for the project.

The study took place over a three week period, which, however, involved actual contact with the subjects on only four occasions. One session, in fact, had to be cancelled because there was an electrical problem in the part of the building where the classroom housing the computer that day was located, resulting in a power failure in all electrical sockets; no other room was available at the time. The time that the subjects spent on the tasks was further limited by constraints relating to the availability of the computer for the task itself since some of the time that the computer was available was spent in preparing the questionnaires and reviewing and printing out of texts. The situation was further complicated by what Ahmad et al. refer to as the problem of 'portability' of software, whereby certain programs will only run on certain types of computer. The particular problem encountered in this study was that, although the BBC computer is quite common and a few machines were available in different locations, it was only the particular MASTER version used by the school which operated the FOLIO work disc containing the texts created by the subjects of the study. Timetable constraints meant that sessions relating to this study immediately followed or preceded those relating to the project work of a colleague. Thus some time was also spent in the handing over of the hardware between the two projects.

HARDWARE AND SOFTWARE

The computer used was a BBC Master Series for which the Folio word processor has been specially designed. The word processor used was the French version of European Folio which is based on the original Folio (UK) word processor which, the manufacturer claims, is used in over 20,000 schools. The particular advantages of the French version are that the menu screens display options in French and accents are keyed in independently of the letters to which they apply. This resembles how accents are added in handwriting more faithfully than other programs which often use a single key for the one-touch typing of the letter and its accent. This feature is also valuable for the creation of exercises which involve the deletion of accents from a text which then have to be inserted by the learner. A further advantage is that characters are always displayed on the screen as they will appear on the printout. This means that if a user prefers to work with a particular typeface, of which there are three, then his/her final product will be printed in that same typeface. This was found to be a motivating characteristic of the program, since the subjects seemed to like to see their work, both on screen and on print out, in a particular cursive typeface ("Lettre") which seemed the most 'human' of the print formats and in which most of the texts were printed (see appendix). The favoured print size was MEDIUM, which was chosen from a choice of four. It is also quite difficult to delete written work, and almost impossible to do so by accident since all work is automatically saved on the work disk.

The most appealing feature of this particular word processor, however, is what the manufacturer calls its "elegant simplicity". Although, for the purposes of this study, the subjects were not required to manipulate the more complicated operations of the program, they learned the basic writing and text editing operations quickly and were not daunted by the technical demands of the task. Those used to using the more powerful 'professional' word processors (WORDSTAR, WORDPERFECT) may point to the limitations of Folio, for example its relative slowness, the problem of portability, the small document size (2000 words), or its annoying tendency to display a 'CATALOGUE FULL' error even when the disc itself is not full. Yet these limitations are of little consequence to the foreign language learner, while they may exasperate the teacher who uses the program to enter texts or exercises for the learner to work on.

THE TASK

The subjects were first asked to fill in a pre-task questionnaire (see appendix) and then to write a paragraph or two describing themselves. The reason for choosing this topic was that this was the first contact with the subjects and, in the absence of close familiarity with their knowledge of vocabulary and grammar, it seemed a safe assumption that this was a subject on which they were likely to be able to produce, without assistance from the researcher or from each other, a reasonable amount of text on which to base the subsequent editing exercise. It was also assumed that this was a topic which would motivate learners to write, since they were being asked to give information to somebody who did not know them and who actually wanted information from them, about them. No other guidance or instruction was given at this stage since these texts were to form the basis of the editing task on the word processor, so the subjects were assured that they were not being tested on their writing ability and would not be criticised for their mistakes. This part of the exercise they completed willingly and silently, without conferring or asking questions, perhaps because they were focusing on writing information rather than writing correct French for no particular purpose (the message rather than the medium).

This initial exercise was done with pen and paper, since time did not permit each one of the six to enter their first draft onto the computer, even using a simple word processor. Although the idea that three of the group should do this, with the other three as a control subgroup seemed a feasible one, this would have introduced the notion of testing word processing skill, which was not the aim of the project and again there may not have been enough time to pursue it. It would also have seemed a little unfair to deny students partaking in a study of word processing the chance to use the computer. The scripts were collected and analysed for errors, but not marked in red ink in the conventional way. They were stored as they were on the work disk as six separate files, each file bearing the name of the author, as far as the six character would limit would permit (ALISON, HANNAH, SUZANNe, GILLian, etc.).

The group was divided into two sub groups of three, one spending a session working together at the computer, the other away from the computer but in the same room using pen and paper. Both groups were asked to proof read three of the original texts and produce a corrected version. The texts were selected in no particular order. Each member of the computer group took a turn at the keyboard for each text. Since the texts had been saved on the work disk the subjects found it easy, after brief instruction, to summon up a text by selecting CHARGER (LOAD) from the initial menu and entering a name at the screen command WHAT IS THE WORK CALLED? (QUEL EST LE NOM DU TEXTE?). They also found it easy to manipulate the cursor and delete keys since, although this was their first encounter with this kind of exercise, they appeared to have had previous experience of using a keyboard, either on computer or typewriter. The two sitting either side of the keyboard operator appeared willing to give assistance, either verbally or by pressing the appropriate key themselves; which method depended on how confident the keyboard operator appeared. They sometimes expressed slight frustration with each other and with themselves when keyboard mistakes were made, although this took the form of a good humoured eagerness to get on with the task rather than genuine annoyance. There was much discussion among the three concerning errors in the texts and how to correct them. When questions concerning language were asked of the researcher by either group, the guidance given was framed in such a way that the subjects might work out the answer themselves. Such clues, were given to both groups, irrespective of who had asked the original question.

The group working with pen and paper were given photocopies of three of the original scripts and, as with the computer group, asked to arrange themselves so that they could work together to check them for errors. A different member of this group was to write a corrected version for each of the three scripts. Thus, as far as possible, the task was identical for both groups apart from the fact that one was working with the word processor and the other with pen and paper.

At the end of the session, the hand-written 'fair' copies were collected from the non-computer group; the edited versions of the original texts on the work disk were saved on the same disc and printed out as they were completed by the computer group. Again, after brief instruction, they found no problems in choosing the relevant commands at the menus. These texts were given appropriate filenames, e.g. (Hanna2, Suzan2, Clare2).

Each group worked on only three texts for various reasons. Firstly, the time taken was consistent with the duration of one session (about thirty minutes); secondly it permitted each subject to take a turn at having the responsibility for producing a corrected text, whether this took the form of writing a final copy away from the computer or of saving/printing out of a final copy at the computer; thirdly, it allowed each group to spend a session working on a set of scripts different from those of the previous session. For the next session, the activity was repeated, but this time with the original computer group working with pen and paper and vice versa.

OBSERVATIONS

As might have been expected, the group working on the word processor took less time to complete the exercise. This appeared to be for two reasons. They were saved the laborious task of writing out a corrected fair copy of the original text since the act of finding and correcting errors was itself the process of creating the corrected copy. Moreover, since the texts were displayed on a vertical screen, each member of the group had a clear view of it and was more involved in the activity; the responsibility for finding and correcting errors was not assumed to be that of the keyboard operator. The result was, as Chaika (1982) terms it, a "collaborative learning" atmosphere. It was observed that all three of the group asked questions of the researcher. These questions were of three types:

1. questions relating to points of grammar, spelling, vocabulary and punctuation,

2. queries about computer operations,

3. suggestions about style and the appearance of the text on the screen.

Questions put by the non-computer group were asked exclusively by the subject who had the responsibility for writing out the corrected version of the text. These questions were only of the first type mentioned above. This was accounted for by the fact that the work done by this group was carried out in a less collaborative atmosphere than that of the computer group. This may have been because it was more difficult for all three of the group to have as clear a view of the text on paper as the computer group had of the text on screen. The only subject with a perfect view was the one who was to write the fair copy. The result was that the other two were less inclined to become involved in the activity and were more easily distracted.

This deeper involvement of the computer group in the activity and the distraction of the non-computer group may be attributed to what Piper (1986) refers to as the 'fascination' which the computer holds for learners. This may have applied initially, and was certainly the case when the attention of members of the non-computer group wandered towards the other group, yet it was observed that this fascination soon transformed into an involvement in the task itself. Moreover, it is not certain that the computer holds quite the 'fascination' for these young (13/14 years old) students that is assumed by those of the previous generation. It may be the case that the views held by CALL practitioners and theorists of how young people today regard computers are related more to their own experience, which might not reflect the reality of learners', particularly young learners', conception of the computer. Thus the poor quality of some CALL materials may to a certain extent be attributed to this misconception. At the very least, it suggests that more learner- centred research should be undertaken in this area.

It was also observed that the second non-computer group were unwilling to perform the task on a script which they had already worked on using the word processor in the previous session. This was not the case with the second computer group; they were quite willing to correct and print out a text which they had worked on away from the computer in the previous session. This can be seen as a manifestation of what some observers have identified as the 'powerful motivation to write' exercised by the word processor.

In the final session, all six of the subjects were allowed to work on the second version of their own texts, which had been edited and then saved by the computer groups over the two previous sessions. This time, all six were grouped around the computer, each taking a turn at the computer to summon their text from the work disk and use the keyboard, assisted by the other five. Having studied a printout of this first edited version of their text on which any remaining errors had been highlighted (but not corrected) by the researcher, they edited it on the screen. This final version of the texts was saved on the work disk (as HANNA3, SUZAN3, CLARE3, etc.). The intention was to provide the subjects both with more time at the computer and a professional-looking final copy of their work rather than to gain textual data for analysis since .

RESULTS

It was hypothesised that an indication of the relative efficiency of editing text on the word processor and on paper might be gained by comparing the number of errors that were corrected by the two groups. Data was obtained from an analysis of the hand-written fair copies produced by the non-computer groups and the first and second corrected versions produced by the computer groups. Because, as mentioned earlier, the subjects were disinclined to edit texts on paper that they had worked on at the computer, not all six of the subjects' texts were represented at all stages of editing. The four subjects' texts which were represented at all stages of editing yielded the following results:

Analysis of the editing stages of the remaining two subjects' texts yielded the following results:

With the exception of subject D's text, more errors were identified and corrected by students working with the word processor than those working with pen and paper. For reasons mentioned earlier, no hand-written fair copy was attempted for the text of subject E. Furthermore, it can be seen that the second editing exercise at the word processor produced three edited texts in which all errors had been corrected and three texts in which all but one of the errors had been corrected. No corresponding pen and paper results were obtained for this stage because the subjects were understandably unwilling to undertake such an exercise.

Although it is recommended that a larger-scale experiment be conducted, to include an analysis of the type as well as the number of errors corrected, the data would appear to indicate that using the word processor as an aid in the foreign language writing class increases the willingness of learners to become involved co-operatively in the task. In this way, they develop a more acute awareness of errors made in writing, and are therefore more likely to produce correct written work since they are more willing to review and edit their work at the word processor than when using pen and paper. Further research is suggested into whether editing skills acquired at the computer are transferred to the learner's writing in general and also into the implications that an increased awareness of errors might have for the learning of grammar and vocabulary, or other areas of language learning.

THE QUESTIONNAIRES

The subjects completed a questionnaire at the beginning of the project and another at the end (see appendix). The most interesting result of the analysis of their responses would seem to relate to question 3 of the pre-task questionnaire and question 1 of the post-task questionnaire. From the former , we see that five of the subjects predicted that they would enjoy doing an exercise on the computer A LITTLE, (although the type of exercise had not been specified, they were aware that it would involve word processing) and only one predicted that she would enjoy doing the exercise VERY MUCH. From the latter , we see that four of the subjects reported having enjoyed working on the computer VERY MUCH, and two as having enjoyed it A LITTLE. From an analysis of individual responses, we see that this involved a positive shift in attitude in three of the subjects, while the attitude of the other three remained unchanged.

Although only six respondents were involved, there does appear to be an implication that the experience of working with the word processor brought about a more favourable attitude to this type of work. All six indicated that they would like to work again on the computer but they would generally prefer to do a different kind of exercise to repeating the one described in this experiment. Only five, however, of the six respondents thought that working on the computer could help them improve their French.

Data obtained from the responses relating to the subjects' attitude to the skills of reading, writing, speaking and listening in French provided little upon which to base predictions or conclusions about the influence of their experience with the word processor in the shaping of these attitudes. Clearly, more extensive empirical research is required into how using the word processor can affect learners' attitudes to the teaching and learning of these skills, in particular that of writing.

CONCLUSIONS

The scale of the study was admittedly limited, in terms of the number of subjects and types of activity involved as well as the time scale involved. It can therefore be argued that this compromises the validity of the results. Yet, notwithstanding the practical limitations of the experiment, observations made during the course of the activities undertaken by the subjects, both at the word processor and on paper, provide a compelling argument for the adoption of the word processor as an aid to the teaching of foreign language writing skills, and perhaps beyond.

Yet the suggestion is not that all teaching should revolve around the word processor. Until longitudinal studies of its 'lasting appeal' have been carried out, as has been done for commercial games software, it would be rash to advocate anything more than frequent, carefully managed sessions. Data obtained from the analysis of the students' texts at various stages of editing and from the questionnaire support what have been the intuitions of teachers for some time about the potential of the computer in mother tongue and foreign language learning and teaching. These are that the word processor provides powerful motivation to write; writing can be seen by the learner as a creative process since ideas and perceptions can be immediately realised in the text. Learners become less inhibited when they can come to realise that their errors are not 'mistakes' in the traditional, negative sense but a necessary part of the creative process. They are more likely, therefore, to see proof reading of their work as an important and constructive activity because it leads them towards a tangible goal: a final text which can be printed out in an elegant, professional-looking format and in which they can take pride.

Kevin Wilkinson May 1989

 Part I of the dissertation

Homepage