Concept-mapping as a Research Tool: A Study of Primary Children’s Representations of Information and Communication Technologies (ICT)
Matthew Pearson, University of Huddersfield
Bridget Somekh, Manchester Metropolitan University
Paper presented at the British Educational Research Association Conference, Cardiff University, 7-10 September 2000.
Concept-mapping as a Research Tool: A Study of Primary Children’s Representations of Information and Communication Technologies (ICT)
Matthew Pearson, University of Huddersfield: Bridget Somekh, Manchester Metropolitan University
This paper presents an approach to concept mapping which was used in Project REPRESENTATION[1] to study ten year old pupils’ representations of Information and Communication Technologies (ICT). Project REPRESENTATION was funded by the EU Multimedia Task Force, 1998-2000 (MM1045). It aimed:
(1) to benchmark and track the developments in children’s representations of computers and computer systems
(2) to develop a computer-based concept-mapping tool which could be used to explore these new modes of representation.
(Representation 1998, p. 2, ‘Goal and Rationale’)
As part of its work, Project REPRESENTATION carried out case study work in schools in six countries: Catalonia, England, Denmark, France, Greece and the Netherlands. The present paper refers mainly to the work carried out in schools in Calderdale LEA, led by researchers at the University of Huddersfield.
The form of concept mapping used by the Project involved drawing images, as opposed to producing text labels in linked boxes, and the computer-based concept-mapping CBCM tool developed by REPRESENTATION incorporated libraries of images, as did Inspiration[2], the commercially-produced CBCM tool which was used by project schools during 1999-2000. Although this computer-based concept mapping work was of considerable interest, it is not the subject of this paper. Computer-based concept mapping and paper-based concept mapping proved to be rather different tasks, suited to different kinds of uses. The concept mapping tasks referred to in this paper were all carried out with sheets of A3 paper and pens. The same task was administered twice, to one group of around 30, year 5, pupils in each of six schools, first in October 1999 to collect Benchmark Data (BM1) and again the following June to collect comparative data (BM2). During the intervening period many of the schools took delivery of new PCs and were connected to the Internet through the UK government’s National Grid for Learning (NGfL) initiative.
The pupils were first instructed in how to produce a concept map and given an opportunity to practice on a topic such as ‘holidays’ or ‘our school’. The task was then administered using an agreed ‘script’ (translated into the home language in other countries) to ensure that as far as possible all children carried out the same task. They were asked to ‘communicate their ideas through drawing’ and reminded that this would be particularly useful because the teachers and pupils in the other countries spoke different languages. The topic of their concept maps was Computers in Today’s World and the introductory script included the following words:
This is an investigation, not a test of either your knowledge of your drawing ability. We want to know what you think, not what you have learnt at school, or what the person sitting next to you thinks. Please draw what is in your mind. You can draw things that are imaginative and fun as well as things you know. You will need to draw quickly.
(…)
We want you to think about your world and all the types of computers within it. What would you say was the most simple computer system, and the most complex computer system you can think of? Where are computers placed at home … at school … in the outside world … or in the work place? Are they connected in any way? Think of all the people who use them. And why they use them. Take a minute of two to think before you start drawing. (Suggest that children might close their eyes and put their heads down on their folded arms for this thinking time.)
To guard against any unfairness to those better able to communicate through words, the children were also given a written task to ‘describe a computer system to an extra-terrestrial being … who has no knowledge of such things.’
In using concept mapping to identify and track the development of children’s mental representations of ICT, Project REPRESENTATION was building on a psychological theory of mind ‘both constituted by and realized in the use of human culture’ (Bruner 1996, p. 1), which has been developed in recent years by a number of researchers, who have been influenced by the work of Vygotsky. Rather than learning being seen as a process of assimilating concepts and processing these in the memory, learning is seen as an integral part of engaging in human activity, necessarily ‘situated’ in a context that involves interaction with others (Brown, Collins et al. 1989). As Lave and Wenger put it:
Learning is not merely situated in practice – as if it were some independently reifiable process that just happened to be located somewhere; learning is an integral part of generative social practice in the lived-in world.
(Lave and Wenger 1991, p. 35)
Children, according to this theory, would be learning about ICT as part of their daily experience in their homes and the community, as well as through planned curriculum activities (or other unplanned experiences) at school. These would be constantly organised and re-organised in a dynamic process of meaning-making that exists in the mind as ‘representations’. In developing their representations of ICT, children would be learning to use a new set of tools capable of ‘mediating’ human action. According to Wertsch (1998, p 25), this process of ‘mediated action’ consists of two inseparable components held in tension: agent and mediational means. The child, in this case, has already mastered a large number of tools such as knives and spoons, bags, bicycles, pens, pencils and scissors, and has made considerable progress in mastering cognitive tools such as language and number; and through using them is able to exercise agency and accomplish tasks. ICT is now added to these potential tools and, as language and the printing press extended human capabilities for their distant ancestors, has the potential to extend their own capabilities considerably during their lifetime. This tool use depends, however, not merely on having access to ICT tools, but on the children’s representations of those tools. Cole (1999, p. 91, develops the ideas of Wartofsky (1979) to explain the essential nature of two kinds of tool artifacts in mediating human activity. Primary artifacts are the tools themselves, such as computers, printers and the Internet; secondary artifacts include, in addition to cognitive tools such as language and number, the representations of the primary artifacts which enable the agent (child, in this case) to imagine and plan how to use them. According to this theory of the relationship between human agency and mediating tools, access to ICT is not enough, and learning skills in using specific pieces of software and hardware is not enough either; what is necessary in order to use ICT tools to mediate action is access to ICT plus a complex mental representation of ICT that makes its possibilities apparent and attractive. With both of these in place, the acquisition of skills of use becomes merely a technical matter that is likely to be unproblematic. This, in fact, fits the pattern of what we know about the way in which most of those with computer expertise, including children, acquired their skills.
Concept mapping appeared to be a good method for investigating children’s representations of ICT. As originally developed by Novak and Gowin (1984) concept mapping was a means of externalising internal processes. Initially, it was developed as a tool in which the learner could produce ‘maps’ of concepts ‘linked by words in a semantic unit’ to put forward propositions (p. 15). These can be used to structure learning and to ‘negotiate meaning’ in educational settings, for example when children are learning scientific concepts (Anderson-Inman and Ditson 1999). Buzan (1993) extended this idea into something much more exploratory, seeing Mind Maps as a means of understanding one’s own internal mental processes, what he calls ‘these shimmering and incessant embraces’ within which ‘the infinite patterns, the infinite Maps of the Mind, are created, nurtured and grown.’ (Buzan quoted in McAleese 1998, p.1). The concept mapping tasks in Project REPRESENTATION aimed at capturing these patterns, recognising that children’s representations of ICT would not be fixed and rigid but fluid and dynamic (Jonassen, 1998); yet confident that they would give a sufficiently clear indication of the nature of the children’s understanding and its development over a year to be able to inform the development of the REPRESENTATION tool. What proved just as interesting, ultimately, for the English team, was the insights that the children’s concept maps gave us into the appropriateness of the current National Curriculum for ICT.
The concept maps we collected are immediately pleasing as aesthetic objects which provide a window into the world of children; however, to be effective as research evidence we needed to develop a method of analysing them both quantitatively and qualitatively. Altogether, in England, we collected around 360 maps from 180 children (one class of around 30 in each of six schools). The process can, perhaps, be seen as akin to brainstorming, since the children were encouraged to draw freely and make their own links and connections. There was no intention on the part of the researchers of looking for ‘right answers’ or ‘better’ or ‘worse’ maps. We simply wanted the know how the children conceived of computers in today’s world. Two approaches to analysis were adopted, one for quantitative analysis of the entire set of maps, the other as a preliminary means of analysing some maps qualitatively. The overall approach was phenomenographic. In each case, maps were scored as follows:
- the number of nodes
- the number of links
- the type of map[3]
- the depth[4]
Phenomenographic methods (Marton 1994) usually involve open-ended, in-depth interviewing but the analytical methods can also be applied to other forms of data (Hogg, Johnston et al. July 1997). Analysis of the transcribed interviews, or in this case concept maps, is carried out by two researchers who establish a ‘hierarchically ordered set of categories’ to describe each individual’s process of acquiring conceptual understanding. The aim is to establish general trends across a cohort rather than absolute accuracy in relation to any one individual. Marton suggests that sufficient accuracy of analysis will be achieved if the two researchers agree in at least 2/3 of the cases when comparing their judgments and where they reach agreement in 2/3 of the remaining cases after discussion (Marton 1994, p. 5).
It was clear that there would be some advantage in using drawings rather than text to communicate in a European project involving children with six different national languages, in addition to those with other mother tongues; but we also believed that there would be particular advantages in investigating children’s representations of ICT through images rather than words. We live in a world where the written word has been traditionally privileged and revered as the means of communication ideas, but we also live in a world where the power of the visual image dominates advertising; where ‘brands’ and ‘logos’ comprised of (labels and) images sell products; where television, combining moving images and spoken language, is arguably still the most powerful medium of mass communication; where photographs sell newspapers as much as headlines; and where the Internet, that comprises a mix of still and moving images with text, is capturing the imagination of young and old and merging with other technologies to challenge many existing economic and social structures. To gain an insight into the richly iconic and visually complex world which children routinely inhabit, one only needs watch a children’s television programme[5]. These programmes often split the TV screen into a number of areas and different kinds of information is projected into these different spaces. Likewise contemporary computer games overlay information about the game on top of the main action screen, and the screen often contains a subtle counterpoint of different types of information. Children growing up with media which deploy these forms of information presentation soon become expert users, and are able to “read” a complex representational space with a multiplicity of messages presented in different channels. Adults, more used to processing a single-stream of information, may struggle with the same tasks.
There is no reason to suppose that children growing up today should feel more confident communicating in words than in drawings. In fact it is quite possible that the opposite may be true since the acquisition of skills in written language is an arduous process. ICT, including the whole range of available software and the Internet is heavily icon-dependent, hence there is a certain logic in inviting children to draw ‘icon-like’ images in producing a map of their mental representations of ICT. Images are also, according to Cohen (1998, p. 158), ‘polysemic’ and ‘their meaning is always provisional, being decided by their anchorage in specific texts and contexts’. This may allow children to externalise the meanings they ascribe to ICT more easily, creating their own contexts for the images by exercising free association rather than fully formed thoughts. Their resulting concept maps, which draw upon their socio-cultural experiences, are then open to semiotic interpretation of the kind described by Warburton, (1998, p. 252), although these images are private rather than public and do not draw on the kind of established system of signs and genres that shape communication in the cartoons that are the subject of his research. In discussing individual concept maps later in this paper, we have, therefore, used a similar method to Warburton in first ‘describing what one sees’ and making judgements about what counts as signifiers, and then ‘attempting to “confirm” strands of meaning carried by signifiers’ in this case by referring to what we know of children’s experiences in current day society and their own homes. This provides a basic tool for qualitative analysis that could form the basis for further in-depth phenomenographic interviews with some children.
The empirical work undertaken during project REPRÉSENTATION has allowed us to build and rich and complex picture of the ways children understand computers, the Internet and digital technology. The task given to the children was open-ended and required them simply to draw on a sheet of paper, their own ideas about ‘Computers in Today’s World’. Some guidance on concept mapping techniques and methods was given to the children prior to the completion of the task, but emphasis was placed on mediating the subject’s productions as little as possible, and allowing each child to develop their own ideas. The adoption of an open-ended research task saved the project from concentrating purely on the elicitation of instrumental knowledge, and allowed the children to show us that their emerging thinking about digital technologies transcends factual knowledge about components, protocols, processes and systems. As we shall demonstrate later, children of this age tend to represent technology as part of a complex web of social and cultural influences. They do not just focus on what the technology “is”, but also what it can do, and in particular what it can do for them. The children also prove themselves to be accomplished futurologists and produced some fascinating and challenging representations of the way they see digital technology developing. Many of the children represented technology in ways which made them active participants in digital communities of practice, and often revealed complex identity negotiations taking place in relation to their usage of technology. Those pupils who had access to the Internet outside school, and so had the freedom to explore this medium without the limitations placed on them by teachers or the curriculum, represented themselves as confident consumers of digital products, downloading screensavers, wallpaper and music. Vaguely couched concerns from the older generation that new technologies may be encroaching on human autonomy appear not to trouble 10 year olds, who see instead the possibilities of enhanced and expanded identities created by interaction with the digital world.
The first important finding from the concept mapping task relates to the research task in its entirety. Whilst it might be expected that some pupils would struggle with the task, perhaps as a result of limited experience of computers, or a reluctance to engage in a task which they would not normally do in school, this was not the case. Less than 3% of the children produced outputs which could not be categorised as concept maps, and even the maps which showed a paucity of detail or a misunderstanding of the task in question still have things to teach the alert and sympathetic researcher. The majority of children relished the task and set about mapping their concepts with purpose, inventiveness and imagination. It soon became clear that although types of maps could be identified, and we could begin to build a taxonomy of 10 year old’s representations of digital technology, the productions themselves appeared to confound simple categorisation techniques and a formulaic approach to analysis which relied on reducing maps to a set number of types would only tell half of the story. The scale of the project, and the need for a system of analysis which could be used to coordinate the dataset across the 6 countries taking part in the research did force the research teams to adopt a standardised practice when interpreting and classifying maps, but the maps themselves appeared to offer resistance to this approach and this created a different interpretive practice amongst researchers. This alternative approach sought not to reduce differences amongst maps, but to celebrate these differences: classification gave way to case-study and large amounts of effort were directed at teasing out the significance of individual maps. Most maps could tell us something unique about children’s perceptions of digital technologies and an interpretative methodology which allowed us to work with the concept maps in such a way that individual differences were not prematurely subsumed into categories was needed. Phenomenography provided the needed theoretical framework, and the preliminary stage was a Phenomenographical analysis of some of all of the maps. It is important to stress that phenomenographic interpretations of research data do seek to produce a finite and logically ordered set of final categories into which all of the representations collected as part of research task can be assembled. Ultimately researchers have to try and interpret and classify human experience. For the second stage of more detailed analysis of individual maps we used the process of semiotic interpretation described earlier . Where phenomenography has the edge over other research methods, particularly when applied to data of this kind, is the capacity of the method to generate broad categorisations that derive from the complexity of mental representations. Phenomenography’s starting point lies in capturing the subtly different but inter-relating ways in which people perceive the world around them. Marton (1994) sums up this point in the following way: