General Comments

Comment from Received
Ken Johnson 17 January 99

Ken Johnson, Australian National Univiersity


These comments are made in the light of experience developed in teaching a course which has as its primary goal to provide an environment in which students can learn about the creation and communication of information from data.

The vehicle for the course is the analysis and interpretation of longer term climate series and the classification of multi-attribute spatial data. One aim is to teach modern nonlinear, nonparametric techniques of analysis and explore graphical representations. The S-Plus software system forms the basis of the analytical system which has a simple easy to understand menu to provide access to and extend the S-Plus algorithms and graphic representations.

The analytical system developed from a fascination with the variation apparent in limited views of the variability of climate. Research developed to better understand the variation in longer term series. The results of this work are in final drafting after presentation at a range of conferences. The time series work focuses on variation of the seasonal cycle.

To develop the understanding of recently developed analyses, a different approach is taken from the traditional one of statistics. This change is essential if we are to be able to deal with many contemporary questions which demand answers from large data bases.

To develop an approach to the issues involved extensive research has been undertaken into ideas in philosophy, psychology and cognitive science, as well as teaching pedagogy. An action research approach has been adopted in consultation with the university's academic teaching research and advisory group.

The course is supported with materials both on the Net for students at ANU and available in printed form. Materials include: a text (in preparation), course guide, 'maps' of concept and model structures, case studies and CAL tutorials. This wide range of materials facilitate a degree of flexibility in learning.

Comment for Conference

Linking teaching, learning and research is essential if we are to provide a valuable learning experience for students. The problem is that the culture of academic work and its setting in the community directs attention away from its own practice towards other problems.

There are three reasons why research and its application in teaching and learning in higher education is essential:

  1. The great challenge posed by the rapid change of knowledge and techniques in the discipline and other studies.

  2. Recent developments in our understanding of human intelligence.

  3. The need to be able to use information technology (IT) effectively and efficiently as the technology has developed and been applied across a wide range of activities.

As a vehicle for the argument it is worth considering one of the greatest challenges facing geography: the development of massive data bases. Deriving information from these data bases faces several problems: outcomes are remote from direct field observation experience; their spatial and temporal coverage is wide; and the development of understandings from the data faces major problems of computation and representation. How to deal with this in the traditional setting of the discipline is problematic.

We have to question whether established approaches are appropriate to our needs. As an example I would take the traditional approach of training in statistics which places so much emphasis on statistical inference and hypothesis testing, often in experimental settings. In a data taking environment data mining and exploratory data analysis are more effective. But to train in these techniques what abilities and steps are essential?

IT systems can help, but we have to give time to training to develop the ability of students, and staff too, to effectively and efficiently use the systems. This is a problem in programs crowded with course offerings and limited time in daily lives.

IT can be involved in several roles in teaching and learning:

All three are important and demonstrate the range of issues involved. One great and positive feature of our age is the rapid development of new ideas, often stimulated or forced by the development of the new ways and means created by technology.

Some of the most fascinating work is in the realm of human intelligence; the target and essential medium of teaching and learning. Even if you do not think that teaching and learning can develop intelligence, it is important that teaching and learning does understand and address human intelligence.

It seems to me that we can only work with the pressing problems outlined above if we develop both the intelligence of human beings and the things best done on IT systems. Teaching and learning has to be aware of the attributes of intelligence, create an awareness of the nature of intelligence among students. Approaches should be linked to particular abilities of human intelligence. One that is frequently referred to in the literature of the last decade or so, and is supported by modern graphics systems, is what we might call visualisation. This is thinking which involves a multidimensionality, not just thought about individual items in serial order.

Visual perception and thought about the imagery in the mind is an essential in field observation. In this thought both direct perception of landscape and different representations of field areas are important (maps, aerial photographs and satellite imagery to name the key general ones). How our mind deals this material is little understood and perhaps we have to take more notice of what mathematicians say about how they conceptualise and solve problems.

Ken Johnson
Department of Geography, Australian National Univiersity, Canberra, ACT 0200, Australia
Phone: 02 62494267 / +61 2 62494267; Fax: 02 62493770 / +61 2 62493770; Email:

Return to Internet Discussion Comments

Return to Internet Discussion Papers

Page last updated 18 March 99
GDN pages maintained by Phil Gravestock