An Internet Journal devoted to the Psychology of Personal Constructs

PDF version

Journal Main
Contents Vol 5





Susan J. Bell*, Richard C. Bell**

* Private Practice, Melbourne, Australia,
**Department of Psychology, University of Melbourne, Australia



Although Self-characterization is a reasonably well known tool for PCP diagnosis, the rules by which information is drawn from the writing are less well known. Here we show how these rules can be applied to a child's drawing, and how attention to these rules highlights the importance of the process in making inferences about constructs in drawings.
Keywords: self characterization; construct elicitation; children’s drawings


The nature of personal construct theory seems especially suited to research with children, since it is essentially a forward-looking theory that emphasizes change, and all children change as they grow up (Hayhow, Lansdown, Maddick & Ravenette, 1988, p.199). Yet there is relatively little research with children in Personal Construct Psychology. (The bibliography of Chiari (1990) lists only 45 references to childhood in a total of around 1700. Why this is so is hard to say, however it may well be partly that the traditional methods of enquiry in personal construct psychology are not well suited to work with children.
In fact a variety of PCP techniques have been used in both research and clinical applications with children. This has included: a form of self characterization (Jackson, 1988; Jackson and Bannister, 1985); responses to set questions (Bannister & Agnew, 1976); a board game variant of the grid (Jones, 1997); a range of drawing techniques (Ravenette, 1977, 1999, 2003); and of course the ubiquitous repertory grid (Applebee, 1976; Baxter, Jack, & Schröder, 1998; Edwards, 1988; and Salmon,1976). This range of techniques is surprising, considering the general picture of dominance of grid methods (see for example Figure 1 of Neimeyer, 1985 p. 146) in personal construct research. In this paper we wish to focus on the issue of process in personal construct inquiry. In doing so we hope to show some previous initiatives; i.e., Jackson and Bannister's (1985) work with self characterization, and a drawing approach, somewhat different to that of Ravenette (1977, 1999, 2003), can be combined to provide a method of personal construct inquiry appropriate for children. In doing so, we propose to apply Kelly's formal rules of self-characterization to a drawing and show how the process of constructing the drawing contains useful information in itself.
The Repertory Grid
Kelly (1955) proposed two major techniques for identifying a person's constructs, the Repertory Test and the Analysis of Self-Characterization. The former has gone on to fame if not fortune, although it is a much generalized approach and has been renamed the Repertory Grid Technique in order to avoid the normative connotations of the word 'test'. As a technique, it is very much a process; elements are defined one after another, then constructs are elicited, again one after another, and finally, elements are related to con­ structs (through ratings, rankings or some such). The latter two stages may be completed alternately [as in the original Kelly approach and in computer approaches such as FOCUS (Shaw & Thomas, 1978)], or sequentially, where all constructs are elicited prior to ratings being made. Keen and Bell (1981) have proposed a further amalgamation where elements, constructs and ratings are all collected in turn.
However there have been few considerations of the actual process of collecting personal constructs. Some computer based approaches [e.g. Shaw & Thomas (1978), Bell and Keen (1981)], have attempted to assess the differentiation among constructs or elements during the process of grid elicitation, but this has aimed simply at increasing the variation in the grid, rather than describing the process.
Salmon (1976) reviewed the use of grids with children. She reported that a child needed to be about eight years old in order "to categorize elements presented in verbal terms" (p.24). In general she suggested, that the younger the respondent, the more concretely defined would the elements need to be. Allison (1972), was reported as working with children as young as four years using names of people as elements. However, no checks were made on the meaningfulness of the childrens' usage of these names.
The use of triadic elicitation of constructs seems even more equivocal. Such procedures have tended to produce more concrete, physicalistic constructs rather than abstracted psychological ones (Brierly, 1967; Jones, 1997; and Ravenette, 1964), leading to either supplied construct approaches (eg Applebee, 1976; Hicks & Nixon, 1989), or dyadic rather than triadic elicitation (e.g. Allison, 1972; Baxter, Jack, & Schröder, 1998). However dyadic elicitation may increase the possibility of problems of the subsequent use of constructs in that Yorke's (1983) 'bent' scales may emerge.
Supplied constructs depend on how relevant one person's constructs are for another, and correspond in a sense to the direct questioning of Bannister and Agnew (1976), or the supplied prompts in Ravenette's (1977) elaborated complaints approach. Edwards (1988) in a study of four year-old children, used situation grids with supplied elements and constructs, drawing these from extensive pilot work and representing individuals and objects with simple line drawings, though noting that "they were able to deal with familiar situations as elements with visual signifiers" (p. 235)
Self-characterization is more often seen as a product, given that it is an essay produced in response to specific instructions:
"I want you to write a character sketch of Harry Brown, just as if he were the principal character in a play. Write it as it might be written by a friend who knew him very intimately and very sympathetically, perhaps better than anyone could ever really know him. Be sure to write it in the third person. For example, start out by saying 'Harry Brown is ..''   (Kelly,1955, p.323)
Recitation of these instructions generally seems to be the extent of acquaintance with the technique in work subsequent to Kelly (see e.g. Fransella, Bell, & Bannister, 2004) in literature on personal construct psychology tech­niques. However Kelly (1955) himself went much further, giving quite specific directions for analysis, using four techniques; observation of sequence and transition, observation of organization, reflection against content, and collation of terms. The analysis also proceeded through three bases: contextual areas, thematic analysis and dimensional analysis. The inclusion of observations of sequence and transition, pointed to an awareness by Kelly of the process that underlay the product.
Jackson (Jackson, 1988; Jackson & Bannister, 1985) analysed children's self characterizations via six scores; view of others, personal history and future, psychological cause and effect, psychological statements, contradictions, and insights. These were related to corollaries of the theory but could also be seen as corresponding to Kelly's bases of r interpreting self-characterization: view of others and personal history and future as contextual areas, psychological cause and effect as thematic analysis, and the rest as dimensional analysis. Kelly indicated that this last basis was the most directly relevant to personal construct theory since it was here that the constructs could be identified. Also, in one of the few classical developmental aspects to his work he linked preverbal constructs with the construal of "those elements of which an infant could be aware" (p.461) and saw them as representing "a kind of core of the client's construct system".
This suggests two conclusions. One is that we can expect that the younger the child, the more important will be the preverbal constructs in his or her repertoire. This will not be due simply to the child's restricted verbal skills, but also to the closeness to that period of life (infancy) where preverbal constructs were pre-eminent. We can also conclude that it is important, since these constructs are at the core of the child's construal system, to explore these in some detail with the child. A consequence of these conclusions is that in order to inquire into such construct systems, we may need a technique that is both non-verbal and relatively unstructured. Drawings provide such a possibility.

Children's drawings
The use of drawings in a personal construct context has been pioneered by Ravenette (1977). One of his techniques is reported as requiring the child to
"to draw pictures to show five occasions when he would be troubled or upset. While the invitation is being made a sheet of paper is folded into six rectangles and a mark is put into five of them. It is pointed out to him that these marks are merely to help him get started on his drawings and he does not have to use them. He is given the pencil and invited to carry on. ... In the sixth space the child is invited to draw a situation in which everything would be fine, he would feel good and people would seem good." (Ravenette, 1977, p.277)
Ravenette's example is design for use with older more articulate children, since the child is asked to verbally elaborate on this. The technique can also be seen to be an elaboration of a superordinate construct of 'troubled occasions - good occasion'.
Another of his techniques was a variant of the ‘squiggle’ approach of Winnicott.
“It involves the drawing, in the centre of a sheet of paper, of a three-inch-long line bent over for a further half inch. The child is then invited to turn the line into a picture (not just an object). When that is finished, and only then, the child is asked to draw a picture which is an opposite.” (Ravenette, 2003, p.290)                                                                                                                
The child is then invited to give an account of the two drawings.
Here we wish to consider the information in a single drawing, and in a more general sense. In the preceding section we concluded that core preverbal constructs were of particular importance in working with children. If we wish the child to present us with a representation of these core constructs then we know that these will be largely symbolic, may be threatened by the child's candour, and may not be accessible to verbal labelling. Self characterization is a way of doing this with the verbally competent (although we sus­pect it is much less used [in comparison with the grid] than it should be). With young chil­dren we are left to find other means. The traditional ones have been oral discourse, draw­ings and play. Of these three, we would suggest that drawing provides the construct thera­pist with the most appropriate context for looking at the child's world. Drawings, like the rep grid and self characterization, are a record of what happened. Like self characterization (and unlike the grid) the process can be mapped over time, in that the order in which a child introduces things to the drawing can be recorded. Unlike the self-characterization however, the child is not constrained to produce material in a serial order, he or she can switch from object to object from outline to detail and back again.
Kelly (1955, p 325), in talking about the structuring of the self-characterization task, re­ferred to the need to minimize threat - “The threat that the client's own sense of candour imposes on the situation, rather than the threat which he identifies with the clinician”. Although Kelly does not elaborate on the nature of this threat, he suggested that “an effort is made to place the client initially in a protected spot within a loosely construed system”. Drawings, we suggest, can provide an excellent 'protected spot'. It is hypothesized that this will allow the child to produce symbols for covert and even threatening constructs, in that the potential space enables such constructs to be produced in such a way as not to be threatening. The flexibility of drawings also brings with it a price - how are such data to be analysed? The suggestion proposed here, is through the application of Kelly's formal guidelines for the analysis of self-characterization data. This is elaborated through the following illustration.

The data
Tim is seven years old and the youngest of four children, the oldest being 20. He lives in an outer Melbourne suburb with his mother and the other children. His father left about three years ago although he has continued to see the children every other weekend. Tim suffers from asthma and a severe allergy to bee-stings.
The instructions for the drawings that Tim was asked to do were as follows:  
I am going to ask you to do some drawings. As you draw you can talk out loud about your drawings if you like. Later after we've finished the drawing, I'll talk to you about what you've drawn.
I'd like you to think of these sheets of paper as a space for you to be, however you'd like. It's sort of like a part of your world, a place for you to be while you draw.

Here I'd be happy if you drew anything that you would like. It's your space to draw something that you'd like.
Tim's drawing is shown in Figure 1.

Figure 1. Tim's Drawing
An outline of how Tim put together this drawing is as follows:
Tim has difficulty in deciding what to do. He turns the paper on its side. Begins a line at the top (what turns out to be part of the giraffe), then a lower line, shading it in rather than drawing a solid line. Stops and begins the fence line on the left hand side, putting in windows or bricks. Adds second line of bricks.
Works on giraffe again - draws head, horns/feelers, eye - square markings. Extends neck downwards first on one side for a little, then on the other and continues alternating in this fashion.
Begins a tree on the right hand side. Works from the bottom up - from the wall line. Adds a curve in the centre of the picture and a curly bit on the top.
On the right hand side bricks are added from the top of the wall downwards. The bottom of what turns out to be the zoo-keeper's booth is drawn next.
At this stage Tim is drawing very intently and privately. He has his arm up against his head and works very close to the paper. Adds "The Zoo" to the entrance arch.
More bricks are added to the right hand side of the drawing. Then in turn, more work on the tree trunk, the right hand side branches, the left hand side branches, and an animal is put in the tree.
Further bricks are added, line by line. Tim stops and gets an eraser to rub out a mark under the giraffe's eye.
Next the side boundaries of the wall are defined. More, very precisely drawn bricks are put in. Tim blocks out the view of his drawing again. Left hand margin bricks are added, becom­ing larger as more are added. This continues from left to right.
Tim smiles and looks self conscious.
A zoo sign is drawn in and then more bricks on the right hand side. He begins what may be a person, but blocks any view with his arm again. It is revealed as a person paying. The animal in the tree is elaborated. A second person, head, body, arms and legs are added. A third smaller person, a child is added. The smile drawn on this figure is erased and redrawn in an exaggerated fashion.
More bricks are added near the gatekeeper. Bricks are added in several rows to the left hand side, then to the right hand side. All this time he rests his head on one hand.
"I've finished"
Later, during discussion Tim adds the sun as a line across the top left hand corner.
Following this, Tim was questioned about his drawing.

Tell me about your drawing...
That's the ticket seller. That's the queue. That's the giraffe, the koala, and the tree. Thats all the bricks. That's the (bridge?) opening. That's me and Mum and Dad.
What's it like being there?

Good. I like animals. I like looking at the monkeys and I like the butterflies. I meet friends there and I go there for school excursions. I like it when it's not too crowded and I like the giraffes.

Because they're very tall and I like their colours. Red and Orange. No, yellow and brown, I like, that's why I like giraffes. They've got funny ears. And I like the lions, they're big and they roar.
What would be good about being a giraffe?

Good. [Query] Because I'd get fed every morning and tea and lunch. I'd eat the leaves.
How are the koala and the giraffe different?

The koala climbs trees. Eats gum leaves. The giraffe can't climb trees and eats different sorts of leaves. And I like the monkeys, they swing from the branches and I like the apes. They're big. The elephant I like because it's big and it's got a trunk. I like tigers, their stripes. Tigers have got black stripes and I like black stripes. I like deer, reindeer, because they've got things on their heads. I like birds, because they have beautiful, nice colours and can fly. And I like pelicans. Ducks, I mean pelicans, have big beaks and float on water. I like kangaroos because they sometimes box and the mother has the baby joey and the fathers box.

Do you think you're more like a giraffe or a koala bear?

I'm more like a koala because the branches come over me and I can have shade and I can drink water and I like little baby koalas. They're cute.
What would your father like?

Dad. He's small. I have to do ...(?) Dad likes monkeys or lions or tigers. He likes tall ani­mals. I think he'd like the giraffe.
What would your mother like?

Mum would like the giraffe. She likes yellow and brown animals.
I like pandas. I saw them when they came last year. I like black and white animals. They're lucky because they travel around the world. I like them because they're furry and warm and they eat lots and lots and lots of bamboo shoots.
When you go to the zoo, what will the day be like?

It will be sunny [adds sun to drawing]. Lots of kids will be there.

The analysis
The analysis of the drawing via Kelly's self characterization procedure was as follows:

(1) Process
1(a) Observations of sequence and transition

Simply looking at the finished product of this drawing obscures an important quality of Tim's drawing, the sequence in which the drawing was put together. Kelly suggests we should consider breaks in continuity not as discontinuities, but unexpected elaborations of subjectively similar content or contrast elaborations. Thus we saw Tim move from the gi­raffe to the wall, back to the giraffe, over to the tree, joining the two parts of the wall, elaboration of the wall, back to the tree. These elements stand very much in contrast to one another in animate and inanimate senses, but also share similarities in texture or in shape. The people are added towards the end of the process in the middle of an elaborated enclo­sure, and even after they are added, further bricks are drawn on th wall.
1(b) Organization
A notable feature of this drawing is its symmetry. The top of the wall splits the vertical aspect of the picture in two, the wall itself is divided, and there is a tall object on either side above the wall.

Kelly suggests attention should be paid to the opening structure, and the way this is elabo­rated through the narrative. Here, the opening structure is the drawing of the giraffe's head, and this is done with the paper on its side and relatively close to one edge, as though there is some security in this. Tim also constructs, later in the drawing, a boundary within the picture by putting outer edges on the wall, though he doesn't do this for the inner edges.
1(c) Reflection against content
In a similar fashion, Kelly suggests that we should look for the core statement, and consider how this is elaborated in the context of the protocol as a whole. If we take (as a working hypothesis or construct) the giraffe as the core statement, then we can see this elaborated through the wall where the texture picks up the pattern on the giraffe's neck and in the tree which contains an animal looking out.
1(d) Collation of terms
Kelly suggests that we should be alert to terms which are repeated, and to linkages between the terms. The icon that is strongest in this drawing is the square block patterning of both the wall and the giraffes neck. It contrasts with Tim's use of curves, which are used in isola­ tion on the animals (the koala and the giraffe's head) and again, this time in combination, in the half circles used to decorate the entrance and to enclose the people, and in the rounded heads trunks, and arms of these people. Some elements are not repeated. The giraffe's horns (possibly mistaken by Tim for ears) are carefully drawn circles but the tree branches are straight and either end in points or are squared off. As the wall are linked by the arch, the people are linked too, by their hands.
1(e) Shifting emphasis
The predominant emphasis in this drawing are the building shapes, blocks or curves. Kelly suggests that terms repeated point to constructs with a wide range of convenience. [In retrospect, we should have taken this up with Tim in subsequent questioning and asked him to elaborate on this construct.]
1(f) Restatement of the argument
Attention to the process of this drawing showed a neat restatement by Tim - i.e., after the people had been added as the last figures in the picture, a few more bricks were added - just to be on the safe side perhaps.

2. Contextual areas

The topical area for this drawing is a zoo. In it everything is contained in some way by the wall, but the animals can look out. There is also a sense of isolation - one giraffe, one koala, one tree. The people are small and bounded by the wall, suggesting a lack of security. We are reminded of Kelly's views about security in this context:

"Some clients write their entire self characterization on comparatively safe ground. Others, more venturesome, take their chances in areas in which they are not so certain where they stand. But we always know this much: the areas chosen are those in which the client sees enough uncertainty to make exploration interesting and enough structure to make it mean­ingful."  (1955, p.334)
Kelly also suggests that in this area we pay attention to mechanics, where mistakes indicate an attempting to work with deeply rooted constructs. Here Tim was worried about the giraffe's eye, which was redrawn, and the little boy's smile which was redrawn to increase the smile.

3. Thematic analysis
Here we draw principally on Tim's talking about his drawing. He begins by talking about animals, monkeys and butterflies (neither one of which featured in his drawing), at the zoo where he likes to meet with his friends from school, but doesn't like it if it's crowded. Tim liked the giraffes (it wasn't crowded in his drawing and there was only one giraffe). Tim had some initial confusion about the colour of the giraffe and moved on from this element in two ways, one on to lions with their big roar, the other onto how giraffes are fed and can feed themselves (though no food such as leaves on the tree appeared anywhere in Tim's drawing). When asked how koalas and giraffes were different, Tim saw differences in climbing and in the kinds of leaves they ate. From the koala he moved on to monkeys (who were free to swing and climb) to apes that were big. He then began a catalogue of animals, emphasizing their particular qualities. This was focussed principally on their appearances (tigers with black stripes, deer with things on their heads, birds with nice colours etc). Kangaroos were noted as having mothers to look after the baby and fathers to box. Tim himself thought he was more like a koala, in being protected while his parents would prefer the giraffe.

4. Dimensional analysis
If we consider all the above material we see a number of important constructs being used by Tim. The world was divided into the secure and the not secure, where things had to be either constrained or free, and needs (particularly in relation to food and protection) were met or not. Figures were attractive and powerful or not, and either aggressive or caring. Figures could be alone, and in some sense, independent and self reliant, or dependent. Tim focussed his attention on the giraffe (which he identified with his parents) but saw himself as a koala, protected in theory, but as his drawing showed, rather exposed, unattractive, and without sustenance.

This paper has demonstrated that a drawing for a child may be seen as a 'protected spot' that enables preverbal and submerged core constructs to be dealt with through their symbolic representations, the interpretation of which involves the interviewer. This issue of the role of the interviewer was raised by a reviewer, and rightly so. The interpretations depend on the construing processes and skills of the therapist. It is an issue for this approach - but it is also an issue for self-characterization more generally since the therapist makes sense of the written protocol. For example, Neimeyer (1993, p.86) shows how the therapist infers the construct poles of constructs identified (by the therapist) in the narrative. And it is also true of all other enquiry methods in personal construct psychology, although rarely recognized. Neimeyer (1993) shows how the grid user may make inferences from construct labels in the repertory grid, and even in the quantitative analysis of such data, the grid analyst determines the picture obtained from the grid data (see for example, Fransella, Bell, & Bannister; 2004, pp 87-88,112). These problems have consequences for the use of such data as a research tool, but in a therapeutic setting such therapist construals can be tested with the external world of the client in the same way that Kelly suggested all constructs are validated.
Not only does the use of drawings involve process, they have an advantage over verbal self-characterizations in that they are not bound by the sequential process of recounting a story. Various behaviours, such as changing some things, deleting others, and adding emphasis makes the behaviour of drawing as important as the picture itself.


Allison, B. (1969) The development of Personal Construct Systems: A Preliminary Study Unpublished manuscript, Memorial University, St Johns, Newfoundland, Canada.
Applebee, A. N. (1976) The development of children's responses to repertory grids. British Journal of Social and Clinical Psychology, 15, 101-102
Bannister, D., and Agnew, J. (1977) The child's construing of self. In A. W. Landfield (ed) Nebraska Symposium on Motivation 1976. Lincoln, Nebraska: University of Nebraska Press. (pp. 91-125)
Baxter, I.A., Jack, F.R., & Schröder, M.J.A (1998) The use of the repertory grid method to elicit perceptual data from primary school children. Food Quality and Preference, 9, 73-80.
Bell, R. C., & Keen, T. R. (1981) A statistical aid for the grid administrator. In M.L.G. Shaw (ed) Recent developments in Personal Construct technology London: Academic Press. (pp. 209-216).
Brierly, D. W. (1967) The use of personality constructs by children of three different ages. Unpublished Ph.D. thesis, University of London.
Chiari, G. (1990) References in Personal Construct Psychology & Psychotherapy. Rome: Centro do psicologia e psicoterapia costruttivista.
Edwards, A. (1988) A child of four could tell you: a study of identity in the nursery school using situation grids. In F. Fransella and L. Thomas (Eds.) Experimenting with Personal Construct Psychology London: Routledge & Keegan Paul (pp. 232-244).
Fransella, F., Bell, R., & Bannister, D. (2004) A manual for the repertory grid technique – 2nd edition. Chichester: Wiley.
Hayhow, R., Lansdown, R., Maddick, J., & Ravenette, T. (1988) PCP and children. In F. Fransella and L. Thomas (Eds.) Experimenting with Personal Construct Psychology. London: Routledge & Keegan Paul (pp. 199-209)
Jackson, S. (1988) Self-characterization: dimensions of meaning. In F. Fransella and L. Thomas (Eds.) Experimenting with Personal Construct Psychology London: Routledge & Keegan Paul (pp. 223-231)
Jackson, S. R., and Bannister, D. (1985) Growing into self. In D. Bannister (Ed.) Issues and Approaches in Personal Construct Theory. London: Academic Press. (pp. 67-82).
Jones, R.A. (1997) The school behaviour game: Making ‘human sense’ of the repertory grid in research with children. British Journal of Guidance & Counselling, 25, 399-407
Keen, T. R., & Bell, R. C. (1981) One thing leads to another: A new approach to elicitation in the repertory grid technique. In M. L. G. Shaw (ed) Recent Advances in Personal Construct Technology London: Academic Press.
Kelly, G. A. (1955) The Psychology of Personal Constructs. New York: Norton.
Neimeyer, R. A. (1985) Problems and prospects in personal construct psychology. In D. Bannister (Ed.) Issues and Approaches in Personal Construct Theory. London: Academic Press, (pp.143-171).
Neimeyer, R.A. (1993) Constructivist approaches to the measurement of meaning. In G.J. Neimeyer (Ed.) Constructivist Assessment: A casebook. Newbury Park, CA: Sage (pp. 58-103).
Ravenette, A. T. (1977) Personal construct theory: An approach to the psychological investigation of children and young people. In D. Bannister (ed) New perspectives in Personal Construct Theory London: Academic Press (pp.251-280).
Ravenette, A. T. (1999) Personal construct theory in Educational Psychology: A practitioner’s view. London: Whurr Publications.
Ravenette, A. T. (2003) Constructivist intervention when children are presented as problems. Ch 28 in F. Fransella (Ed.) International handbook of personal construct psychology. Chichester: Wiley (pp. 283-293).
Salmon, P. (1976) Grid measures with child subjects. In P. Slater (ed) The measurement of intrapersonal space by grid technique, Vol. 1. London: Wiley. (pp. 15-46).
Shaw, M. L. G., & Thomas, L. F. (1978) FOCUS on education: An interactive computer system for the developmental analysis of repertory grids. International Journal of man-Machine Studies, 10, 139-173.
Yorke, M. (1983) Straight or bent? An inquiry into rating scales in repertory grids. British Educational Research Journal, 9, 141-151.

Susan Bell trained as a Child Clinical Psychologist and worked as such for a number of years at the Royal Children's Hospital in Melbourne. She currently works in private practice.

Richard Bell is an Associate Professor in the School of Behavioural Science at the University of Melbourne. He has a slight interest in analysing the data of repertory grids.

E-mail: rcbell@unimelb.edu.au 
Homepage: http://www.psych.unimelb.edu.au/people/staff/BellR.html



Bell, S. J., Bell, R. C. (2008). An illustration of self-characterization in a child's drawing: The importance of process. Personal Construct Theory & Practice, 5, 1-9.
(Retrieved from http://www.pcp-net.org/journal/pctp08/bell08.html)


Received: 30 October 2007 – Accepted: 9 January 2008 Published: 18 March 2008


ISSN 1613-5091

Last update 18 March 2008