Teaching and Learning Forum 99 [ Contents ]

Evidence of the gap between students' learning approaches and instructors' teaching approaches in accounting education

Dennis W Taylor, Tungshan F Chou
Curtin University of Technology
James Fisher
Wilfrid Laurier University
There is a substantial body of literature on university students' approaches to learning. In this body of literature, a prominently featured instrument is Entwistle et al's (1979) approaches to study inventory (ASI) and its revised version, Tait and Entwistle's (1995) RASI. Approaches taken by students to learning have been found to be influenced by, and responsive to, the context in which learning takes place. However, the evidence is based on learning contexts that are commonly conceptualised and constituted in the minds of students, not in the minds of lecturers. To investigate the association between students' perceptions of learning contexts and their perceptions of their approaches to learning, verges on a tautology. This study seeks to avoid such a tautology by investigating the relationship between students' learning approaches, as perceived by students, and instructors' facilitation of students' learning approaches, as perceived by lecturers.

This is an exploratory study of accounting undergraduate students and their instructors. Our instrument has been designed to measure instructors' facilitation of students' learning, using the same constructs of deep, surface and strategic learning, as Tait and Entwistle's (1995) RASI instrument. This new learning facilitation instrument was administered to lecturers in a four-year undergraduate accounting program in Canada. During the same period, Tait and Entwistle's RASI instrument was administered to accounting students taught by these lecturers. Results reveal several significant differences between lecturers' self-reported approaches to facilitating learning and students' self-reported learning approaches. The gap between teaching and learning approaches becomes greater for fourth year cohorts than first year cohorts in the accounting program. When students are grouped according to gender, further significant differences are found. Implications of these findings for future change in accounting education are discussed.


The learning process in higher education represents a field of inquiry which should be of intrinsic interest and practical benefit to every instructor in higher education. However, according to Meyer and Muller (1990), the practical value of research findings and theory developments concerning the student learning process is "not always demonstrated in the classroom and it is, therefore, not surprising that (learning theories arising from research findings) evoke little enthusiasm on the part of the average teacher in higher education" (p.131).

There is a considerable body of literature on student learning approaches or study orientations which tells us, inter alia, that we can quantitatively measure differences between qualitatively different student approaches to learning or studying (see literature review section below). Theoretical and empirical constructs relating to studying/learning approaches should be of fundamental importance to our understanding of the student learning process, as well as the outcomes of the process. The problem that such research has not translated into direct practical value for the average instructor in higher education, as suggested by Meyer and Muller (1990), may be due to a fundamental question of what studying/learning approach instructors seek to facilitate. Do instructors seek to facilitate a diversity of student learning approaches, or do they expect all students to develop a common "ideal" approach? There is a lack of evidence about the connection between students' perceived studying/learning approaches and instructors' perceived role in the facilitation of these approaches.

Rather than investigate instructors' beliefs, prior empirical research has sought to relate students' studying/learning approaches to students' perceptions about the context in which studying/learning takes place. Two instruments have featured prominently in this line of research; both come from the same study by Entwistle and Ramsden (1983). They are the Approaches to Learning Inventory (ASI) and Course Perceptions Questionnaire (CPQ). In the CPQ, dimensions of the studying/learning context include students' perceptions of formal teaching methods and course goals and standards. According to Meyer (1988), if instructors alter the context, this does not necessarily alter students' perceptions of it. Therefore, it has been argued, "(students') perceptions of learning context represent potentially powerful variables that can ... influence the quality of student approaches to studying" (Meyer and Muller, 1990, p.132).

However, such a pursuit of student perceptions has been a line of inquiry which has made instructors' beliefs about the facilitation of students' learning/studying processes irrelevant. The fact that there has been a lack of translation of findings about student perceptions of their studying/learning approaches and contexts into practical classroom outcomes by instructors, therefore, is not surprising.

Our study differs from previous studies both conceptually and methodologically in one important respect. In relation to the context of students' studying/learning, we consider instructors' perceptions about their role in facilitating students' approaches. We develop a new instructor-oriented instrument to replace the student-oriented CPQ instrument. Our instrument covers those aspects of the studying/learning context identified in the Accounting Education Change Commission's (1993) Issues Statement on teaching practices -- viz., course materials, instructor presentation skills, pedagogical (including assessment) methods and instructor guidance and advising. At the same time, our instrument parallels the key dimensions of students' studying/learning approaches contained in the Revised Approaches to Studying Inventory (RASI) of Tait and Entwistle (1995). These are the dimensions of deep, surface and strategic studying/learning approaches.

The original motivation for research into studying/learning approaches was clearly stated in Entwistle and Ramsden's (1983) proposition that "... it might be possible to make improvements to the quality of student learning in higher education by alterations to the contexts in which it occurs" (p.192). This remains the motivation for our study. We seek to provide evidence on the relationship between students' perceptions of their studying/learning approaches and instructors' perceptions of their role as a major element of the context in which these approaches are carried out by students.

Literature review

How university students approach learning, that is, the relative emphasis they give to the seeking of meaning or the reproduction of facts, and how they organise their approach to study, has been a long-standing area of investigation. Three approaches to studying/learning have been consistently identified and are most commonly labelled deep, surface, and strategic.

The seminal work by Marton and Saljo (1976) distinguished between deep and surface approaches to reading an academic article. Essentially they characterised the deep approach as an active attempt by the student to understand the author's meaning, to explain the evidence in relation to the conclusion, and to relate the ideas contained in the article to the students' previous knowledge and experience. In contrast, they defined the surface approach as involving a tendency to memorise discrete facts or ideas, to be anxiously aware of the need to reproduce information at a later time, and to view a particular task in isolation both from the academic subject as a whole and from real life. Further work by Pask (1976) highlighted the fact that students used different studying strategies, which in normal academic environments were associated with characteristic methodologies. He referred to those who concentrated on step-by-step studying approaches and detailed arguments (and ignored inter-relationships between ideas), as operational learners who consistently adopt serialist strategies, To the other extreme, those who relied on analogies and anecdotes to build up their understanding of a topic and jumped to conclusions were referred to as comprehension learners who consistently adopted holist strategies.

The measurements of deep and surface approaches to learning, together with the strategic approach to studying, were captured by Entwistle et al. (1979) in their Approaches to Study Inventory (ASI). The ASI, derived from interview-based research, measures the three dimensions in terms of broader student orientations of meaning (deep approach), reproducing (surface approach) and achieving (strategic approach).

A revised version of the ASI (the RASI) was produced by Tait and Entwistle (1995). The RASI included several changes to the sub-scales within the deep and surface dimensions, and added scales labelled academic self-confidence, meta-cognitive awareness and lack of direction to reflect variables other researchers had found useful descriptors of learning processes. It became a 44 item, six dimensional questionnaire.

A comparable measurement instrument called the Study Process Questionnaire (SPQ) had been developed by Biggs (1979). Biggs' SPQ was derived from judgmental literature on the personal qualities needed for successful academic learning. It measured approaches to studying/learning in terms of sub-scales reflecting students' motives on each of the deep, surface and strategic dimensions.

Students' perceptions of their studying/learning context

In order to determine how studying/learning approaches are influenced by their context, Ramsden and Entwistle (1981) designed their CPQ and found that it produced two main factors. One described formal teaching methods, vocational relevance, and clear goals and standards, and the other represented a favourable departmental evaluation with the highest loadings on good teaching and openness to students. They then drew conclusions about relationships between students' perceptions of their main academic departments and their perceived approaches to studying/learning. A subsequent study by Meyer and Parsons (1989) found that there were no associations between any of the sub-scales of the CPQ and the two ASI dimensions of deep and surface approaches to studying.

The CPQ has been criticised by Meyer and Muller (1990) because, unlike the ASI, it does not distinguish between qualitatively different perceptions of any of its constructs. It determines, instead, the degree to which certain broad constructs (such as 'good teaching', 'workload', 'freedom in learning') are present or absent in a departmental learning context. Perhaps more fundamentally, a tautology seems to underlie attempts to establish a relationship between students' perceptions of their studying approaches and students' perceptions of their studying context. For example, items in the study approaches instrument include whether 'the volume of information strongly challenges your ability to cope', or whether 'ideas the come across are able to be related to those in other topics or other courses'. Of a parallel nature, students are asked, in the study context instrument, whether their department requires a high 'workload' of them, or provides them with a greater 'freedom of learning'.

Approaches to teaching

In accounting education, there has been a large-scale institutionalised endeavour to effect pedagogical change in higher education (particularly through the Accounting Education Change Commission in the US and the professional accounting bodies in Australia). A major element of this endeavour has been the promotion and evaluation of teaching effectiveness. In the US, the Teaching and Curriculum Committee of the American Accounting Association has identified five dimensions of effective teaching. As explained by Calderon et al. (1996) these are:
  1. Curriculum design and course development
  2. Use of well-conceived course materials
  3. Presentation skills
  4. Well-chosen pedagogical methods, assessment devices
  5. Guidance and advising of students
In our study, we seek to integrate dimensions of effective teaching (as defined above) with dimensions of students' learning approaches (as measured by Tait and Entwistle's, 1995, RASI) in order to create an instrument to measure instructors' perceptions of their approaches to facilitating their students' studying/learning. Because the task of curriculum design and course development usually involves the work of academic working parties and committees, and our new 'Approaches to Teaching' instrument is targeted at individuals, this dimension of effective teaching is excluded from our study. Our instrument has the purpose of identifying how lecturers perceive that dimensions of their teaching effectiveness provide the context for students' studying/learning approaches.



Student and academic staff participants were chosen from the accounting major of a BBA degree at a relatively small University in Canada which has a very high national ranking for its undergraduate business and economics programs. This University's School of Business and Economics states as its foremost goal that "the School ensures the relevance of its educational offerings and enhances the excellence of its teaching through continual review and the use of innovative teaching methods."

The first stage of data collection involved the administration of the RASI instrument to students in all four years of the BBA accounting program. There were 334 useable responses. A profile of respondents is given in Table 1. The demographic variable of nationality of student had no significant relationship to the three RASI factors of deep, surface, or strategic approaches to studying/learning. The variables of age and the year students started the program are shown, in Table 1, to be highly correlated to the students year-of-program. Therefore, year of the program is adopted as the test variable, along with gender.

Table 1: Statistical profile of student sample

1. Crosstabulation of GENDER by YEAR OF PROGRAM


Year 1Year 2 Year 3Year 4Total

GENDER: Number (%)
Male 64472728166 (49.7%)
Female 45543435168 (50.3%)

Total No. 1091016163334
Total % (32.6)(30.2)(18.3) (18.9)(100.0)
Number of missing observations: 7


Canadian27380.1 199712436.4 Under 21 years21462.8
Non-Canadian4112.0 19969828.7 21-25 years10530.8
Missing277.9 1995 5917.3 Over 25 years164.7

1994216.2 Missing61.8
Total341100.0 Other308.8
Missing92.6 Total341100.0


3. Bivariate Correlation of AGE, YEAR OF PROGRAM and YEAR STARTED

AGE 1.0000, P= .
YEARPROGRAM .7896, P= .0001.0000, P= .
YRSTARTED .6738, P= .000.8589, P= .0001.0000, P= .

The validity and reliability of data from the student sample was tested, and results are given in Table 2. The 29-item instrument loaded onto three factors, providing a cumulative percentage of explained variance of 57.1%. The Cronbach alpha was acceptably high for all three factors (surface, deep, and strategic approaches), as shown in Table 2.

Table 2: Validity and reliability tests for items in the students'
approaches to learning (RASI) instrument


Factors No.ofVarimax Rotated Factor MatrixCronb
itemsFactor 1Factor 2Factor 3Alpha

Surface approach 10.8198
Relying on memorising 2.6703.4462
Difficulty in making sense 2.7599.4387
Unrelatedness 2.7703.4675
Concern about coping 4.7991.7642
Strategic approach 9.7379
Determination to excel 2.4846.5100
Effort in studying 2.7824.3905
Organised studying 2.7243.2044
Time management 3.8158.7912
Deep approach 10.6242
Looking for meaning 2.5154.1455
Active interest/critical stance 2.6761.0292
Relating and organis ideas 3.7752.3248
Using evidence and logic 3.6946.6006
Pct of Variance
Cumulative Pct of Var.

The second stage of data collection involved the administration of the newly-developed Approaches to Teaching instrument to 34 academic staff who were lecturing on the accounting program to the students who were sampled. This instrument contained 42 items, 16 for each of the 3 studying/learning approaches of students. The validity and reliability tests for this instrument are given in Table 3. The factor loadings suggest that construct validity is questionable for the dimensions of surface and strategic approaches. The items used to measure both approaches fall into two factors, not one. The sample size (n = 34) relative to the items to be factor analysed (42 items) causes instability in the results of the factor analysis. Therefore, the three constructs of surface, deep, and strategic approaches will be disaggregated into their underlying items when we compare the data from the lecturers' sample with data from the students' sample. Table 3 further indicates a high reliability (Cronbach alpha) of the measures in the lecturers' sample.

Table 3: Validity and reliability tests for items in the instructors'
approaches to learning facilitation instrument


Factors No.ofVarimax Rotated Factor MatrixCronb
itemsFactor 1Factor 2Factor 3Alpha

Surface approach 16.8603
Relying on memorising 4-.8395.8973
Difficulty in making sense 4.6170.7955
Unrelatedness 4-.7527.8265
Concern about coping 4.8501.8459
Strategic approach 16
Determination to excel 4.5796.6295
Effort in studying 4.8038.7021
Organised studying 4.9309.9223
Time management 4.9230.8971
Deep approach 16.9050
Looking for meaning 4.9055.7545
Active interest/critical stance 4.8175.7726
Relating and organis ideas 4.7581.8452
Using evidence and logic 4.7502.6297
Pct of Variance
Cumulative Pct of Var.


Before testing for differences between students' studying/learning approaches and lecturers' perceptions of their facilitation of these approaches, the results for the student sample alone are presented. Two research questions are addressed in relation to students' perceived studying/learning approaches:
  1. Do students progress from a surface to a deep approach, or change aspects of their learning strategies, throughout their four-year undergraduate program?
  2. Do male and female students differ in the extent to which they adopt surface vs deep vs strategic studying/learning approaches?

Stage-of-program effect

Table 4 gives the comparative differences, based on one-way ANOVA, for students at different stages of their degree program. Firstly, the mean score (on a scale of 1 = agree to 5 = disagree) reveals that there is no significant difference between the four years for a deep approach (F-probability = .339). Nor is there a significant change from year 1 to year 4 in any item underlying a deep approach. That is, students do not perceive themselves as becoming any stronger or weaker in looking for meaning, taking a critical stance, relating ideas or using evidence and logic over stages of their accounting degree program.

Table 4: Degree program stages compared: Ranking
of dimensions and items in student sample

Tait & Entwistle (1995) rev.Stage of the Degree ProgramOne-way ANOVA
Approaches to Study Inventory Year 1Year 2Year 3Year 4between years
RkMeansRk MeansRkMeansRkMeansF-ratioF-prob
Dimension/item(n=113)(n=102)(n=62)( n=62)(df=3)

Deep approach 3.673.593.773.631.123.339
Looking for meaning 33.9343.7143.8533.72.445.721
Active interest/critical stance 43.8433.7624.0413.891.788.149
Relating and organis ideas 93.4393.3683.5283.36.823.482
Using evidence and logic 63.6273.6253.8053.68.967.409
Surface approach 3.373.503.313.142.659.048*
Relying on memorising 83.4763.64103.4073.57.852.467
Difficulty in making sense 103.36103.35113.27113.031.748.157
Unrelatedness 122.98122.93122.75122.701.297.275
Concern about coping 73.5123.7873.5793.214.447.004**
Strategic approach 3.853.603.813.642.349.076
Determination to excel 14.3014.0714.0743.718.101.000**
Effort in studying 53.7683.4563.7363.602.808.039*
Organised studying 24.0243.7133.9823.84.659.577
Time management 113.31113.0493.42103.191.917.126
Academic self-confidence 3.833.393.693.806.773.000**
Lack of direction 2.582.472.062.062.929.034*

Secondly, Table 4 shows that the mean score for a surface approach is significantly different (F-probability = .048) for stages of the program. In particular, year 4 students are less inclined to agree that they adopt a surface approach (mean = 3.14) than year 2 students (mean = 3.5). The single item which is driving this difference is 'concern about coping'. Concern for coping is greatest in year 2 (mean = 3.78) and least in year 4 (mean = 3.21). Interestingly, the item 'relying on memorising' has not significantly changed throughout the four years.

Thirdly, Table 4 shows that the mean score for strategic approaches to studying/learning is not significantly different from one year of the program to the next (F-probability = .076). However, the item 'determination to excel' has significantly improved from disagree in year 1 (mean = 4.30) to somewhat disagree in year 4 (mean = 3.71). Further, 'effort in studying' appears to have been significantly greater in year 2 than in other years.

A further interpretation of the results in Table 4 is found in the rankings of the items from agree to disagree. The highest ranked item (ie, the highest agree score) across all years is 'active interest/critical stance', whereas the lowest ranked item is 'relatedness' (ie, relating to the facts and details you study, more so than the bigger picture).

Gender effects

Table 5 presents the results for mean differences between male and female accounting students, based on an independent samples t-test. Significant differences are found for both the deep and surface approaches to studying/learning, but not the strategic approach. In relation to the deep approach, females are significantly higher than males in agreeing that they take a deep approach, especially in relating ideas to other topics or courses they study, and in using evidence and logic. On the other hand, males are significantly higher than females in agreeing that they take a surface approach, especially in relation to relying on memorising, relating less to a bigger picture and having concern about coping. Table 5 also reveals that males are significantly less self-confident about their approaches to studying than females.

Table 5: Comparison of means of gender for the RASI dimensions and items

Tait & Entwistle (1995) rev.GenderIndependent samples
Approaches to Study Inventory Males (n=166)Females (n=168)t-test (df=332)
Dimension/itemMeanStd devMeanStd dev t-valuesig

Deep approach 3.7103.5363.5621.573 2.44.015*
Looking for meaning 3.7410.8013.7246.828 0.18.854
Active interest/critical stance 3.9277.7693.7798.746 1.78.075
Relating and organis ideas 3.5171.6713.2996.668 2.97.003**
Using evidence and logic 3.7384.7463.5764.752 1.97.049*
Surface approach 3.1251.7623.5927.786 5.52.000**
Relying on memorising 3.31631.0613.75751.024 3.86.000**
Difficulty in making sense 3.2229.9533.34821.025 1.16.248
Unrelatedness 2.64461.0553.10421.052 3.99.000**
Concern about coping 3.21991.0013.8785.913 6.28.000**
Strategic approach 3.6805.8633.7628.684 0.97.334
Determination to excel 4.0723.7794.1138.738 0.50.618
Effort in studying 3.6114.8453.6257.871 0.15.879
Organised studying 3.88552.2163.8810.984 0.02.980
Time management 3.12651.1053.30751.088 1.51.132
Academic self-confidence 3.8298.7713.5074.781 3.78.000**
Lack of direction 2.37801.4412.30951.418 0.44.663

The gap between students and lecturers

The most central research question in this study is whether there is a gap between students' perceptions of their studying/learning approaches and lecturers' perceptions of the studying/learning approaches which they facilitate in their students. Because we had problems in achieving construct validity for the Approaches to Teaching instrument (see Table 3), our comparison of students' and lecturers' approaches was limited to an analysis of individual items only. Table 6 presents the results of a stepwise discriminant analysis between students and lecturers in which each of the 16 types of items common to the RASI and Approaches to Teaching instruments is entered as a separate variable.

Table 6: Stepwise discriminant analysis of learning approaches of students v instructors

StepVariable entered (variables in the analysis after step 4) ToleranceF to removeWilks' lambda

1Deep approach - relating and organising ideas .940110.7814.7447
2Strategic approach - time management .913522.9311.7685
3Surface approach - concern about coping .86194.0448.7314
4Surface approach - relying on memorising .826147.7086.8173
Wilks' lambda insufficient for further computation (variables not in the analysis after step 4)
StepVariable not entering ToleranceF to enterWilks' lambda

Deep approach - active interest and critical stance .88223.0936.7174

Deep approach - looking for meaning .90083.0043.7235

Deep approach - using evidence and logic .6349.0710.7233

Strategic approach - determination to excel .7860.4105.7227

Strategic approach - effort in studying .6339.0961.7233

Strategic approach - organised studying .7991.0112.7234

Surface approach - difficulty in making sense .6739.0823.7233

Surface approach - unrelatedness of topics .63752.3938.7188
F statistics and significances between pairs of groups after step 4. Each F statistic has 4 and 368 degrees of freedom.

Table 6 reveals that the greatest gaps between students' perceived studying/learning approaches and lecturers' facilitation of these approaches is found in a mixture of dimensions. The first item entering the discriminant function suggests that lecturers perceive themselves in a significantly stronger light at adopting teaching approaches which facilitate 'the relating together and organising of ideas', than do students at achieving this feature in their approaches to studying/learning. The other three items entering the discriminant function, given in Table 6, are associated with time management, concern about coping and relying on memorising.

Conclusions and implications

This study provides findings which extend the evidence about student approaches to studying/learning, using Tait and Entwistle's (1995) RASI. Using undergraduate accounting students, evidence is provided in relation to the nature and extent to which students change their studying/learning approaches over the stages of their degree program. Unexpectedly, no increase in the deep approach was found, and only a few aspects of the surface and strategic approaches changed. In terms of gender differences, the significant results favouring females on the deep approach and males on the surface approach, are contrary to prior research findings. Wilson et al (1996) found no significant gender differences on the equivalent deep, surface and strategic scales in a sample of psychology students.

Of more fundamental importance to the literature of studying/learning approaches is the development and testing of a new instrument to measure students' studying/learning context from the perspective of lecturers. It has been shown that a mix of specific items emerge from the deep, surface and strategic scales as significant discriminators between students' perceived studying/learning approaches and lecturers' perceptions of how they facilitates these student approaches. The identification of these discriminators has important practical implications. The potential means becomes available for lecturers to systematically identify the elements of teaching effectiveness which they are practicing in a way substantially at odds with their own students' studying/learning approaches or orientations. If this is achieved, then the issue of getting research findings and theory developments concerning the student learning process into the classroom, and into assessments of teaching effectiveness, can be progressed.


Biggs, J. B. (1979). Individual Differences in Study Processes and the Quality of Learning Outcomes. Higher Education, 8, 381-394.

Calderon, T. G., A. L. Gabbin and B. P. Green (1996). Summary of Promoting and Evaluating Effective Teaching. Journal of Accounting Education, 14:3, 367-383.

Entwistle, N. J., and P. Ramsden (1983). Understanding Student Learning. London: Croom Helm.

Entwistle, N. J., M. Hanley and D. J. Hounsell (1979). Identifying Distinctive Approaches to Studying. Higher Education, 8, 365-380.

Marton, F., and R. Saljo (1976). On Qualitative Differences in Learning: Outcome and Process. British Journal of Educational Psychology, 46, 4-11.

Meyer, J. H. F. (1988). Student Perceptions of Learning Context and Approaches to Studying, South African Journal of Higher Education, 2, 73-82.

Meyer, J. H. F. and M. W. Muller (1990). Evaluating the Quality of Student Learning: an Unfolding Analysis of the Association between Perceptions of Learning Context and Approaches to Studying at an Individual Level. Studies in Higher Education, 15:2, 131-153.

Meyer, J. H. F. and P. Parsons (1989). Approaches to Studying and Course Perceptions Using the Lancaster Inventory - a Comparative Study. Studies in Higher Education, 14:2, 137-153.

Pask, G.(1976). Styles and Strategies of Learning, British Journal of Educational Psychology, 46, 128-148.

Ramsden, P. and N. J. Entwistle (1981). Effects of Academic Departments on Students' Approaches to Studying, British Journal of Educational Psychology, 51, 368-383

Tait, H. and N. J. Entwistle (1995). The Revised Approaches to Studying Inventory. Edinburgh: Centre for Research on Learning and Instruction, University of Edinburgh

Wilson, K. L., R. M. Smart and R. J. Watson (1996). Gender Differences in Approaches to Learning in First Year Psychology Students. British Journal of Educational Psychology, 66, 59-71.

Please cite as: Taylor, D. W., Chou, T. F. and Fisher, J. (1999). Evidence of the gap between students' learning approaches and instructors' teaching approaches in accounting education In K. Martin, N. Stanley and N. Davison (Eds), Teaching in the Disciplines/ Learning in Context, 417-428. Proceedings of the 8th Annual Teaching Learning Forum, The University of Western Australia, February 1999. Perth: UWA. http://lsn.curtin.edu.au/tlf/tlf1999/taylor-d.html

[ TL Forum 1999 Proceedings Contents ] [ TL Forums Index ]
HTML: Roger Atkinson, Teaching and Learning Centre, Murdoch University [rjatkinson@bigpond.com]
This URL: http://lsn.curtin.edu.au/tlf/tlf1999/taylor-d.html
Last revision: 1 Mar 2002. The University of Western Australia
Previous URL 24 Jan 1999 to 1 Mar 2002 http://cleo.murdoch.edu.au/asu/pubs/tlf/tlf99/tz/taylor-d.html