This is a description of some action research on the use of electronic mail (e-mail) to foster student engagement in syllabus topics. Two cycles of action and reflection are described. Both used a semester-long "assignment" of e-mail messages between student and lecturer regarding the syllabus topics. The changes introduced between the first and second cycles, and the reasons for them, are described. The author concludes that the approach seems useful. Further work on this approach will aim at giving students more assistance in developing their own learning approaches, and in reducing the lecturer's time commitment.

1. The problem and its setting

A degree in information technology appeals to certain types of student, and most would be to some extent technophilic. As a generalisation one could say of these students that they like there to be one "right" answer to any question; given a subject requiring reading and discussion to arrive at the range of factors which might properly influence a decision or an outcome, they may be inclined to dismiss the argument in favour of some superficial answer; and in this author's experience when under pressure they are more likely to put aside preparatory reading than they are to do more technical preparatory work.

The subject "Control and Evaluation" in our Bachelor of Computing and Bachelor of Business(IT) degrees involves a range of topics, and is supported by a booklet of readings from the periodical literature, rather than a textbook. Most students are in their third year, but a few are in their second. The subject requires mature reflection and deep understanding, rather than the acquisition of technical skills. It seemed a prime candidate for some sort of action to increase the likelihood that these students would engage in it fully rather than treat it superficially.

2. Goals of the project

The lecturer was new to this particular subject this year (the "regular" lecturer was away), but familiar with the genre.

The e-mail assignment was introduced in an attempt to:

• encourage students to do the pre-reading;
• encourage them to acquire some background on each topic before the class, to increase their understanding and to increase the chance that they would have a questioning approach to opinions offered there.

There were also a number of ancillary goals:

• to ensure that students were capable of using e-mail as a communication medium;
• to encourage students to develop improved critical reading skills.

All of these could have been achieved by having the students post messages to a list-server. Our Outdoor Education course uses this, and we recognised that it would have the advantages of fostering discussion, and a reduced lecturer workload compared with e-mail. We rejected it in the expectation that many students would offer basically the same comment from their reading, and that a list would leave them wondering what they could contribute if "their" comment had already been made. It would also have been difficult to tell if a subsequent message from someone else was a genuine attempt to phrase the same original point differently, or simply a paraphrase of the comment.

3. The concept of engagement

Ramsden (1992) says "...the quality of our students' understanding is intimately related to the quality of their engagement with learning tasks". His index term is "engagement with learning", and he seems to equate high engagement to a deep understanding of meaning, with a holistic approach to the structure of the material studied (Ramsden, 1992, pp.39, 42-43).

The terms "cognitive engagement" and "student engagement" are ERIC's descriptors relevant to this study. We have concentrated on the cognitive aspects, but have simply used the term "engagement" as all-encompassing. Its intuitive use seems to imply an element of excitement or charm missing from the more academic interpretations.

E-mail as a vehicle for promoting engagement seemed to fill Bixler and Askov's (1994) criteria for effective instructional technology, even though they were more concerned with the evaluation of instruction materials per se.

4. The steps in the project to date

4.1 Cycle 1: Semester 1 1996

The e-mail project described here was part of an assessable component of the subject ("Assignment 1"), worth 25% of the subject's marks. The objectives were stated to the students as "To assist you to engage yourself in the content of this subject. This assignment gives you some incentive for reading, research and participation". 15 of the marks were assigned to the e-mail component, and 10 to tutorial participation.

Students were required to communicate with their lecturer/tutor by e-mail. Five types of message were encouraged:

a) a message at least a week before the scheduled lecture topic, indicating what they hoped would be covered;

b) a message by the day before the tutorial, with a question arising from the reading;

c) notes on some additional source they had found useful on the topic;

d) comparisons (by the day before the tutorial) between opinions stated in the readings, or between them, the lectures, or other sources;

e) a message, by the end of the day after the tutorial, with follow-up comments on the discussion.

The class of forty-five students was divided into three tutorial groups, of which the author took one. Students from other groups sent me mail only in category (a) above.

The recipient of the mail (lecturer or tutor) acknowledged it, sometimes with simply "Thanks, XXX ... Regards, YYY", sometimes with more, and kept a printed copy for later reference.

At the end of semester, each comment was graded out of ten, and the final mark was divided by ten - this allowed us to differentiate between the good and the mediocre messages. In the final tutorial of the semester we went through this printed record of each person's mail with them. This allowed them to be sure that all messages had been included, and to discuss the grade for both the e-mail and the tut.

4.2 Cycle 1 reflection

From the staff point of view, the tutor faced more problems than the lecturer. With thirty students, the volume of e-mail was too much to cope with, so that sometimes it was not read until after the tutorial to which it referred: "I could not keep up with all the e-mail questions. I asked students to repeat them in tutes: few ever did". Another problem was the difficulty for the tutor in detecting whether the students' ideas were their own or the lecturer's. For the lecturer, with half the number of students (and of course a knowledge of any asides in the lecture), neither of these problems existed.

Whether this made any impact on the students we could not tell: the end-of-semester student evaluations were transcribed by the Educational Services Unit to maintain anonymity, and in the process answers to each question were grouped together, so we could not sort them into tutorial groups. Thus we could not tell if a student's comments corresponded to their tutor's impression of the exercise, or vice versa!

The spectrum of student comments, from most positive to most negative, was as follows:

• I really enjoyed participating in the e-mail. It was nice to be less formal and it encouraged thought about the topics on an almost weekly basis.
• Good idea though hard to get motivated to do it.
• It sucked, I really mean this. I always read the articles but rarely had good questions, so didn't e-mail often. I don't think it is a good indication of what you know. However, it did make sure I kept doing it. Maybe allocate less marks next time!

The decision to continue with the approach was made before these transcribed evaluations had been returned, but would possibly have been made anyway. The lecturer would be doing her own tutoring, and the volume of messages was to be set at a maximum of one per week, even if it contained comments on several topics, so both of the main problems identified by the separate tutor were addressed.

New directions for the next semester were sparked by serendipitous reading during the lecture break of Peter November's (1996) paper on journal keeping.

4.3 Cycle 2: Semester 2 1996

The assignment was modified. Its loading and stated objectives remained the same, but all the marks were given for the e-mail and none for tutorial participation. Students were asked to head each message or section of message with the date written and the message type.

The first message (inspired by November (1996) and Dodd (1995)) was to be a personal description: "who are you? what is your background, what are your interests, and what are your ambitions? What is your educational background, and what course are you doing? What do you hope to get out of this subject?".

Apart from that, message types were as before, but with the following additions:

f) notes relating content to the student's own experiences;

g) reflections on the personal learning process.

Suggestions here were:

• let me know what you've been finding too hard, or too easy;
• if you've been falling behind with any part of this subject (attendance, reading, group work, keeping up this journal) say why this is so, and say what you're going to do about it.
• general reflections of the subject, your learning, your journal, etc.

h) The following not-for-marks category was also added, "Factual/courtesy message, e.g. questions about due dates, telling me you'll be missing a day because of an interview, etc".

One aspect we reviewed was whether to continue with e-mail or move to a list server. The e-mail workload had been recognised, and our fears of plagiarism had been allayed - friends often sent messages at approximately the same time, but they always seemed genuinely original. However, e-mail was obviously going to be more appropriate for a journal. There was another factor too: one outstanding student in semester 1 would have left any other student reading her messages feeling inferior!

The previous semester's students had been uncertain of what was expected. In cycle 2, students were shown two messages at the outset: an excellent (10/10) one, and a mediocre (4/10) one (Appendix A), and the differences were discussed.

Another introduction was a tutorial on critical reading, where a paper each on the same topic (Internet censorship) from a newspaper, from the Web and from a refereed journal were discussed . Axelrod and Cooper's (1993) techniques were suggested to aid the pre-reading - but the students seemed to have some difficulty with them, and another time we would prefer to use Turabian's (1969, pp.32-35) evaluation criteria.

This semester saw a somewhat different group of students. The majority in this case were Business students doing an Information Technology major, and many more of them than in the previous semester's group had never used e-mail before, nor the World-Wide Web (specific tuition on these had to be introduced, although that is outside the scope of this paper). While most were in their final year, a higher proportion were in their second year of study.

4.4 Cycle 2 reflection

Asking for a personal description had seemed a risky, potentially invasive act - but nobody seemed to interpret it that way. All students sent this message, although two never sent another, and a few sent very few subsequent messages. It seemed a useful way to establish personal contact, although it did mean one had to go back and re-read the introductions a few weeks into the semester, to make sure one remembered!

The standard of comment received this semester was far below that of the previous semester. There were no suggestions on proposed lecture content (except when it came to the final revision session), and very few notifications of other sources. Where other sources were suggested, they were almost always from preceding or concurrent text books, whereas in Semester 1 there had been many, but mostly Web sites.

On the positive side, there were (obviously, since the assignment required it) more comments on students' own learning, and these seemed to indicate a greater general satisfaction with the assignment than in the previous semester.

The formal evaluations took a different form from the preceding semester, to be consistent with the School "quality" guidelines. Only two questions were open - the "best" and "worst" aspects of the subject - and this assignment had very few mentions in the "worst" category (whereas it dominated it in semester 1).

The identifiable comments, i.e. those received by e-mail, were mostly positive:

• [final message] I will now sign off. I have generally found the journal of benefit and interest, however it was hard to maintain entries in busy periods. I thank you for your replies.

5. Perceived impact of the approach on:

5.1 Lectures:

In semester 1, when more students suggested lecture topics, the exercise was useful. Messages as shown in Appendix A were obviously useful in gauging the state of students' knowledge (the calibre of comment was taken into account!).

5.2 Tutorials:

The ideal of this approach was the questions should be raised by the students, and this did not meet with their approval - to them, it lacked structure. In semester 1 a typical student evaluation comment was "Even though discussions about articles were interesting, we needed more of a guide about what would be discussed in each tute before they occurred".

We would have liked to believe that evaluation comments like the following were deeply perceptive: "Didn't feel like we learnt much in tutorials, although maybe we did without really knowing", and to have continued with the "unstructured" approach in the hope that it would benefit the students, despite being unpopular. However, in semester 2 several factors intervened. One was that pressure of e-mail meant (as the tutor had found in semester 1) that one sometimes faced a tutorial without having read all the relevant e-mail. Another was that the quality of the comments and questions posed by the students this semester was not adequate as a basis for tutorial discussion. Taking these together, it seemed advisable on pedagogical grounds, not just as a bid for popularity, for the lecturer to provide a set of tutorial questions for discussion.

5.3 Staff-student personal relationships

This type of assignment seems to forge a much stronger bond than the average essay-type assignment. A surprisingly useful aspect was that it seemed to make it easier for students to explain absences from class (as well as making it easier to remember who had said what), so one could follow up on interview successes, etc.. Students seemed ready to make these positive explanations, but in the one case where a student had serious personal problems these were not explained via e-mail - the student visited the lecturer personally instead.

5.4 Staff workload

The workload in dealing with the mail is heavy. In Semester 1, with only fifteen students (one tutorial group) plus some messages on expected lecture content, it was manageable, but with double that in Semester 2 it was really on the threshold of unacceptable (as the tutor in Semester 1 had found). The payoff, of course, comes around week 10 of the semester, when the end is in sight for the e-mail assignment, but all of one's colleagues are groaning under piles of assignments to be marked.

The journal style, being more personal than that required in semester 1, seemed to demand longer replies. A personal impression was that this "courtesy" aspect was achieved, but that the content discussion goals were not, and it seems that if the approach is to be useful its goals should be met.

5.5 "Quiet" students and "outgoing" students

From these experiences, outgoing students seem quite at ease with any sort of informal communication! The advantages of the e-mail were with the quiet and mature-age students. Some people who never volunteered anything in tutorials were quite articulate and showed considerable understanding in their e-mail messages. These messages, when in advance of the tutorial, allowed one to know that a student had a worth-while opinion on a topic and to ask them to expand on it then. An advantage not anticipated was that the more sensitive type of mature-age student may feel hesitant about relating their personal experiences on the topic at every tutorial, but can write about it regularly with a clear conscience, and they, too, can then be asked to expand in tutorials.

It must be admitted that this active use of the message content in tutorials was more thoroughly done in semester 1 than semester 2, when simply dealing with the e-mail seemed onerous enough.

6. Pointers to successful use of this approach

The following suggestions seem obvious in retrospect, but we had to "discover" them:

• Give students preliminary instruction in the use of e-mail. This may not be necessary if such instruction has been formally given before, but we found that most were self-instructed in this, and their exemplars were mostly from postings to newsgroups. Points of etiquette like repeating the relevant parts of a message to which a reply is being sent, of putting one's own message where the receiver will see it on the first screen, and of avoiding over-long .signatures seemed as novel to the "hackers" as to the novices.
• Ask students to identify the type of comment and the date in the message header, and to include a signature identifying themselves.
• Consider the full range of message types that will be acceptable. Our omission of a category for messages relating to personal experiences was unacceptable.

7. Some unresolved quandaries

There is an urge to reply to an interesting message rather than postpone it to a tutorial. Which would be better for the student, and/or for the rest of the class?

There is the question of whether or not to correct spelling and grammar in e-mail messages. Nightingale (1996, p.119) says "I know of no one who thinks corrections of any kind should be made, including corrections to spelling and grammar ... the last thing I want to do is read stilted English that has no fluency and flow". This advice was followed, although it then had to be stressed that for their more formal second assignment the presentation requirements were strict.

And there is the general question of assessment and the need for feedback. In both cycles, marks were not given until the final session, although a student who came and asked would be given an indication of the degree of satisfaction with their message type and a count of the messages received. Nightingale (1996, p.118) counsels against marking journals as they are written as he finds "students often take between five and seven weeks to become comfortable with the method and use it for its intended purpose of promoting reflective thinking rather than merely reporting events. I call this point in time the 'take-off'" and "Reflections on how much, how, and why they have changed can only be made late in a course". One's satisfaction with this approach would be greater if one were more satisfied that a number of students really had reached take-off!

8. Conclusion

In the two cycles described, no ideal combination of lecturer and student satisfaction with the approach was reached, but still the exercise has been most rewarding and the approach seems worth pursuing.

The greatest limitation is seen as the fact that in its current form it is really only suitable for small classes. To combat this, the next cycle would be likely to utilise a World Wide Web-based discussion list rather than e-mail.

That would, of course, have some disadvantages too, and one of the purposes of the attempt would be to see if these outweighed the advantages as far as this type of class was concerned. One disadvantage anticipated might be that students would see their development of skill in using a Web discussion group as less relevant than the skill of using e-mail. Another could be that the loss of personal lecturer-student contact would reduce the value of the experience.

Further exploration is likely to be in some different context, as the subject used for this study will be reclaimed by its regular lecturer next year.

Acknowledgements

Thanks to John Olle, who tutored in the subject in Semester 1 and contributed valuable suggestions for Semester 2. He is not responsible for any errors, omissions or misinterpretations in this report.

References

Axelrod, R.B. and Cooper, C.R. (1993) Reading Critically, Writing Well: A Reader and Guide. 3rd ed. New York: St Martin's Press.

Bixler, Brett and Askov, Eunice N. (1994) "Characteristics of effective instructional technology" Mosaic: Research Notes on Literacy, v.4 no.2 (Dec.) pp.1,7.

Dodds, A.W. (1995) "Engaging students: what I learned along the way" Educational Leadership v.53 no.1 (Sept.) pp.65-67.

November, Peter (1996) "Journals for the journey into deep learning: a framework" Higher Education Research and Development v.15 no.1 pp.115-127.

Ramsden, Paul (1992) Learning to Teach in Higher Education. London: Routledge.

Turabian, Kate (1969) Student's Guide for Writing College Papers. 2nd ed., rev. Chicago: University of Chicago Pr.

Appendix A: Sample messages

1. A good message, Semester 1 (10/10)

TOPIC: Testing

The following is a list detailing issues that have arisen as a result of reading the articles, that I would like to see be covered in the lecture on Testing:

* A definition of Testing. The article titled "Testing Object­Oriented Programs", states what testing is and isn't. But from further readings, it appears testing is much more than just an 'activity of trying to find bugs'.

* More information on Black Box and White Box testing. Both articles mention these testing techniques, but basically just go as far as defining them. Also do different software packages/programming languages force the tester to use a particular testing method? For example, it appears that in testing object­oriented programs, the tester is pushed toward black box testing techniques.

* Could random testing go hand in hand with coverage testing? The article titled "A Checklist Approach to Writing Test Plans", describes the effectiveness of random testing in detecting defects, not detected by coverage testing. Therefore wouldn't a joint approach result in more comprehensive testing?

* Development of Test plan. The above article includes a checklist for writing test plans. Is using a checklist the best approach to developing a test plan? Is the checklist generic in nature? Along a similar line, which is the best way to approach testing ­ top­down, bottom­up, or sandwich? Or does it depend on what is being tested?

* The article mentioned in point one, outlines a small number of commercial tools used to test object­oriented programs. Do such tools exist for testing other languages or software packages?

Thanking you ...

2. A mediocre message, Semester 1 (4/10)

I was reading the article "Internet test for RSA IP security" and it mentioned a single security standard for firewalls. What are firewalls? It assumes the reader should know what a firewall is. I have heard this term before but can't remember what it means.

Yours sincerely ...

Jennifer Goddard ©1996.

The author assigns to ASCILITE and educational and non-profit institutions a non-exclusive licence to use this document for personal use and in courses of instruction provided that the article is used in full and this copyright statement is reproduced. The author also grants a non-exclusive licence to ASCILITE to publish this document in full on the World Wide Web and on CD-ROM and in printed form with the ASCILITE 96 conference papers, and for the documents to be published on mirrors on the World Wide Web. Any other usage is prohibited without the express permission of the author.