Ference Marton: In The International Encyclopedia of Education. Secon edition , Volume 8. Eds. Torsten Husén & T. Neville Postlethwaite. Pergamon 1994, pp. 4424 - 4429.

Phenomenography [cont.]

Page II (I)

Hierarchy of capabilities
The different ways of experiencing a certain phenomenon, characterised by corresponding categories of description, represent different capabilities for dealing with (or understanding) that phenomenon. As some ways of experiencing the phenomenon are more efficient than others in relation to some given criterion we can establish a hierarchy of categories of description. It is better to have developed the idea of addition's commutativity and realise that 2+7=7+2, as in the above example, than not to have developed it. To see immediately that 2 and 7 are simply two parts of 9, their order being immaterial, is an even more efficient way of understanding numbers and number relations from the point of view of developing arithmetic skills. In this view, then, it is the way of understanding those phenomena which given skills have to handle and knowledge is about, which is the most critical aspect of skills and knowledge.

Awareness
A certain way of understanding something is a way of being aware of it. Awareness is seen as a person's total experience of the world at a given point in time. Following Gurwitsch (1964), awareness is not seen in terms of the dichotomy aware/unaware or conscious/subconscious, but as being characterised by an infinitely differentiated figure-ground structure. Certain things or aspects are in the foreground, they are explicit, thematized. Other things are in the background they are implicit, unthematized. There is, however, no dichotomy between two classes of things or aspects but rather a more or less continuous variation.

When we are dealing with a mathematical problem we are presumably aware of the quantities involved, the relations between them and the operations we may need to carry out. More vaguely, we are presumably aware of different parts of mathematics in general; it is through our previous mathematical experience that we make sense of the problem.

At the same time we are aware of things which are not immediately relevant to the problem but surround it in space and time. There is the experience of the situation of the world outside this situation, of what happened before we embarked upon the problem and of what is going to happen afterwards. The external horizon of the situation extends in space and time indefinitely. In this sense we are aware of everything all the time. But we are surely not aware of everything in the same way. Every situation has its own relevance structure. The world is seen from the point of view of that specific situation. At the same time the situation is seen through all of our experiences of the world. We are aware of everything all the time and we are aware of everything differently all the time. In a phenomenographic study we are exploring the different ways in which we can be aware of a certain phenomenon or situation. We want to find out the differences in the structure of awareness and the corresponding meaning of the phenomenon or situation.

Methods

Collecting data
It was already mentioned above, in the section on the origin of phenomenography, that the dominant method for collecting data has been the individual interview. How something is experienced can of course be expressed in many different ways. Not least, the way in which a person acts expresses how thing appear to them; in accordance with this, there are phenomenographic studies where group interviews, observations, drawings, written responses, and historical documents have been used as the main source of information. On the collective level, we could also examine artefacts - historically or comparatively, for example - from the point of view of the different ways of understanding the world around us that are embedded in those artefacts (Marton, 1984). A piece of equipment for programmed learning from the late 1960s, for instance, might tell us a great deal about the view of learning embedded in that equipment.

In spite of the variety of ways of collecting data, the preferred method is the individual interview. The reason for this has to do with what has been said about the object of research above, and especially about the structure of awareness. The more we can make things which are unthematized and implicit into objects of reflection, and hence thematized and explicit, the more fully do we explore awareness. There is interesting parallel here to the phenomenological method as described by Edmund Husserl.

Phenomenology too makes human experience its research object. It is however a philosophical method, an enterprise in the first person singular. It is the philosophers themselves who reflect on their way of experiencing the world, or rather specific phenomena in the world. It is not introspection, they are not trying to look into themselves, they are looking at the world, but they are trying to step out of "the natural attitude", in which one's way of experiencing the world is taken for granted. By "bending back" one's awareness - in a manner of speaking - its focus becomes one's way of experiencing something.

It is a similar shift that the phenomenographic interview is trying to bring about in the person who is the subject of the interview. As phenomenography is empirical research, the researcher (interviewer) is not studying his or her own awareness and reflection, but that of their subjects. The interview has to be carried out as a dialogue, it should facilitate the thematization of aspects of the subject's experience not previously thematized. The experiences, understandings, are jointly constituted by interviewer and interviewee. These experiences, understandings, are neither there prior to the interview, ready to be "read off", nor are they only situational social constructions. They are aspects of the subject's awareness that change from being unreflected to being reflected.
This type of interview should not have too many questions made up in advance, and nor should there be too many details determined in advance. Most questions follow from what the subject says. The point is to establish the phenomenon as experienced and to explore its different aspects jointly and as fully as possible. The starting question may aim directly at the general phenomenon such as, for instance, when asking the subject after some general discussion, "What do you mean by learning, by the way?" Alternatively, we could ask the subject to come up with instances of the general phenomenon, asking for example, "Can you tell me about something you have learned?" Most often, however, a concrete case makes up the point of departure: a text to be read, a well known situation to be discussed, or a problem to be solved. The experimenter then tries to encourage the subjects to reflect on the text, the situation or the problem, and often also on their way of dealing with it.

The interview thus aims at making that which has been unthematized into the object of focal awareness. This is often an irreversible process. This kind of research interview thus comes very close to a pedagogical situation.

Analysis
As was pointed out above, in the course of the interviews the participants in the research are invited to reflect on their experience of the phenomena dealt with. They are supposed to adopt an attitude which is similar to that of the philosophers who exercise the Husserlian method of phenomenological research. When the interviews have been transcribed verbatim and the analysis has begun it is the researcher who is supposed to bracket preconceived ideas: instead of judging to what extent the responses reflect an understanding of the phenomenon in question which is similar to their own they are supposed to focus on similarities and differences between the ways in which the phenomenon appears to the participants.

As the same participant may express more than one way of understanding the phenomenon, the individual is not the unit of analysis. The borders between the individuals are temporarily abandoned, as it were. The transcripts originating from the different individual interviews together make up undivided - and usually quite extensive - data to be analysed. The first way of reducing the data is to distinguish between what is immediately relevant from the point of view of expressing a way of experiencing the phenomenon in question and that which is not. (Such decisions may, of course, be reconsidered subsequently in the course of the continued course of analysis). It might sometimes be found that different topics or phenomena have been dealt with in the interviews. In that case the data have to be organised according to topic or phenomenon to begin with and the analysis has to be carried out for each topic or phenomenon, one at a time. The next step is to identify distinct ways of understanding (or experiencing) the phenomenon. There are two mechanisms through which a certain understanding appears. One is based on similarities: when we find that two expressions which are different at the word level reflect the same meaning, we may become aware of a certain way of understanding the phenomenon. When two expressions reflect two different meanings, two ways of understanding the phenomenon may become thematized due to the contrast effect. At this point the analysis boils done to identifying and grouping expressed ways of experiencing the phenomenon (literally or metaphorically making excerpts from the interviews and putting them into piles). In order to do this we have to aim at as deep an understanding as possible of what has been said, or rather, what has been meant. The various statements have to be seen in relation to two contexts. One of the contexts is "the pool of meanings" that derives from what all the participants have said about the same thing. The other context is - and here we have to reintroduce the individual boundaries again- what the same person has said about other things. We have thus to make sense of particular expressions in terms of the collective as well as of the individual context. This is the hermeneutic element of the phenomenographic analysis.

After the relevant quotes have been grouped, the focus of attention is shifted from the relations between the quotes (expressions) to the relations between the groups. We have to establish what are the critical attributes of each group and what are the distinguishing features between the groups. In this way we develop the set of categories of description in terms of which we can characterise the variation in how a certain phenomenon is experienced, conceptualised, understood. There are logical relations to be found between the categories of description and as they represent different capabilities for seeing the phenomenon in question, in relation to a given criterion, a hierarchy can be established. This ordered complex of categories of descriptions has been referred to above as the outcome space.

The different steps in the phenomenographic analysis have to be taken interactively. As each consecutive step has implications not only for the steps that follow but also for the steps that precede it, the analysis has to go through several runs in which the different steps are considered to some extent simultaneously.

The categories of description and the outcome space are the main results of a phenomenographic study. Once they are found they can be reapplied to the data from which they originate. There will thus be a judgement made in each individual case concerning what category - or categories - of description is (or are) applicable. We are then able to obtain the distribution of the frequencies of the categories of description.

Reliability
The question is often raised, would another researcher examining the same data come up with the same results? Such a question implies a view of the analysis as a kind of measurement procedure. And repeated measurements should yield similar results, of course. The analysis is, however, not a measurement but a discovery procedure. Finding out the different ways in which a phenomenon can be experienced is as much a discovery as the finding of some new plants on a distant island. The discovery does not have to be replicable, but once the outcome space of a phenomenon has been revealed, it should be communicated in such a way that other researchers could recognise instances of the different ways of experiencing the phenomenon in question. After having studied the description of the outcome space another researcher should be able to judge what categories of description apply to each individual case in the material in which the categories of description were found. As far as such a judgement is concerned there should be a reasonable degree of agreement between two independent and competent researchers. We let the expression "reasonable degree of agreement" refer, somewhat arbitrarily, to cases where the two researchers agree in at least 2/3 of the cases when comparing their judgements and where they reach agreement in 2/3 of the remaining cases after discussion.

Applications of phenomenography

The experience of learning
As was pointed out earlier, phenomenography developed from empirical studies of learning in higher education. Although the interrelated nature of the act and the outcome of learning was emphasised in these early studies, in quite a few investigation the experience of the act of learning, problem solving (Laurillard, 1984) and so on, and the understanding of the phenomenon of learning, understanding and so on (Marton et al, in press; Helmstad & Marton, 1992) have been held in focus.

Different ways of understanding the content learned
In other studies the major focus has been on finding critical differences in which central phenomena, concepts, principles in specific domains are understood (e.g. Linder, 1989; Renström et al., 1990). There is an idea that this may be the most powerful way of finding out how the development of knowledge and skills within these domains can be facilitated.

Describing conceptions of the world around us
There is a pure phenomenographic "knowledge interest" that transcends the educational context. By describing the different ways in which we can experience, or understand, the world around us, we are characterising the world as it appears to us, which is tantamount to characterising the collective mind, encompassing the different ways in which we are capable of making sense of the world (Marton, 1981).

Bowden, J., Dall'Alba, G., Martin, E., Masters, G., Laurillard, D., Marton, F., Ramsden, P. & Stephanou. A. Displacement, velocity, and frames of reference: phenomenographic studies of students' understanding and some implications for teaching and assessment. American Journal of Physics 60 (3), March 1992.

Gurwitsch, A. (1964) The field of consciousness. Pittsburgh: Duquesne University Press.

Helmstad, G. & Marton, F. (1992) Conceptions of understanding. Paper presented at the Annual Meeting of the American Educational Research Association, San Francisco, Ca, April 20-24.

Hodgson, V. (1984) Learning from lectures. In The experience of learning. Edinburgh: Scottish Academic Press.

Hounsell, D. (1984) Learning and essay-writing. In The experience of learning. Edinburgh: Scottish Academic Press.

Laurillard, D. (1984) Learning from problem-solving. In The experience of learning. Edinburgh: Scottish Academic Press.

Linder, C.L. (1989) A case study of university physics students' conceptualizations of sound. Unpublished PhD thesis, University of British Columbia, Canada.

Lybeck, L., Marton, F., Strömdahl, H. & Tullberg, A. (1988) The phenomenography of "the mole concept". In Chemistry. In P. Ramsden (Ed.) Improving learning - new perspectives. London: Kegan Paul, pp 81-108.

Marton, F. & Neuman, D. (1990) The perceptibility of numbers and the origin of arithmetic skills. Department of Education and Educational Research, University of Göteborg, no 5.

Marton, F. (1981) Phenomenography - describing conceptions of the world around us. Instructional Science, 10, 177-200.

Marton, F., Beaty, E. & Dall'Alba, G. Conceptions of learning. International Journal of Educational Research. (In press)

Marton, F., Fensham, P. & Chaiklin, S. (1992) A Nobel's eye view of scientific intuition: Discussions with the Nobel prize-winners in Physics, Chemistry, and Medicine (1970-1986). (Manuscript)

Marton, F., Hounsell, D. & Entwistle, N. J. (1984) The experience of learning. Edinburgh: Scottish Academic Press.

Neuman, D. (1987) The origin of arithmetic skills. A phenomenographic approach. Göteborg: Acta Universitatis Gothoburgensis.

Renström, L., Andersson, B. & Marton, F., (1990) Students' conceptions of matter. Journal of Educational Psychology, 82, 555-569.

Svensson, L. (1984) Människobilden i INOM-gruppens forskning: Den lärande människan. Rapporter från Pedagogiska institutionen, Göteborgs universitet. Nr 3. /The view of man in the research of the INOM-group: The learning man/

Säljö, R. (1982) Learning and understanding. A study of differences in constructing meaning from a text. Göteborg: Acta Universitatis Gothoburgensis.

Theman, J. (1983) Uppfattningar av politisk makt. Göteborg: Acta Universitatis Gothoburgensis.(Conceptions of political power)

Wenestam , C-G. (1984) Qualitative age-related differences in the meaning of the word "death" to children. Death Education, 8, 333-347.

Page I (II)