9.1.2 Identifying Personal Change: The Messy Change Grid

К оглавлению
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 
119 120 121 122 123 124 125 126 127 128 129 130 131 

Rather than working with two grids, each with the same elements and

constructs, as above, you’re working with different ones: Messing About in

Problems, to quote the title of a fascinating book on the ways managers reflect

on their thinking about practice (Eden et al., 1983).

It’s probably struck you by now that situations in which the simple change

grid applies are rather special. Clearly, if you ask a person to work with their

original elements and constructs the second time round, you may not be

giving them full scope to indicate the extent to which their views have

changed. Left to themselves, they may well have chosen to provide new

constructs if not new elements. Preventing this may be justifiable in a

controlled study or experiment, but less so if you were investigating change in

a counselling or advisory setting.

That’s right. If all that changes betweentwo gridsis the ratings on existing constructs,

it’s arguable that the interviewee hasn’t changed in any fundamental way. S/he is

telling you s/he feels differently about the topic, but the terms which apply haven’t

changed, and s/he isn’t thinking of theworld in different terms.

Your intervieweemayeven changetheirmind about anelement completely, changing

a rating of ‘1’ to a rating of ‘5’, and thereby going fromone extreme of the construct to

the other, so that the opposite end of the pole applies. It is rather like amarble rolling

along a slot from one end to the other, which is why this particular form of change is

called‘slot-rattling’.But, intheliterature oncounselling, therapy, andpersonalgrowth,

it is felt that this represents a relatively shallowlevel of change, since the old construct

remains, and a person who changed their mind so as to provide an opposite rating

may just as readily change their mind back again! They’re still stuck in the same old


In contrast, change which involves new constructs (and indeed new elements) is

generally seen as a matter of more fundamental personal alteration. Some alternatives

to slot-rattling are as follows:

. Drawing on another construct easilyavailable in the personal repertoire.

. Putting pre-verbal constructs into words and using themin the grid (you will recall

the short discussion on pre-verbal constructs in Section 4.3.3).

. Considering the whole construct hierarchy, and the extent to which alternative

constructs more compatible with superordinate personal values, which have not

initially occurred to the interviewee, might bemade available.

. Elaborating (furtherdeveloping) therepertoire of constructs, andpossibly reconsidering

the hierarchical relationships between constructs.

. One way of extending the repertoire of elements and constructs during grid elicitation

is through a process called elaboration. Note when two constructs are being

usedvery similarly (inthe sense that theratings ofallelementsonthese constructs

are identical, or nearly so), and ask the interviewee if they can think of a new

element which would receive maximally different ratings on each of the two

constructs. Likewise, towards the end of a grid, if it is obvious that two of the

elements have received identical, or very similar, ratings on all of the constructs,

the interviewee might be asked to provide a new construct, on which the two

elements would receive maximally different ratings. (See Jankowicz & Cooper,

1982, for further particulars of these procedures.)

. Experimenting with new forms of behaviour, both during the grid session and

outside it, to suggest new ways of making sense of the new experience.

. Encouragingtheapplicationofexistingconstructsto situationsinwhichtheyhaven’t

previously been used, thereby increasing their range of convenience.

. Redefining themeaning and relative importance of existing constructs.

. Developing entirely new constructswhich have never been in the repertoire.

All this is fromWinter (1992: 240^245).

Yes. As you can see, most of these sources of change form part of the deliberate

personal interventions we call, variously, guidance, counselling, and therapy.

These are situations in which the individual is helped, sometimes quite

energetically, to construe the topic differently! You’ll need to look elsewhere

for material on this subject, which is beyond the scope of this guidebook. For

anyone who hasn’t any formal counselling or therapy training, a good place to

start would be Fonda (1982), followed by Jankowicz & Cooper (1982).

However, some of these changes can also occur spontaneously, as the

interviewee decides that existing constructs are insufficient to capture his or

her new experience. This is especially likely if the interviewee has decided that

the old constructs are invalid; that is, that they’re no longer acceptably

accurate in predicting life events (Kelly, 1963: 157–160). This assumption lies at

the heart of Kelly’s theory. His Experience Corollary states that: ‘a person’s

construction system varies as he successively construes the replications of

events’, this being seen as fundamentally a predictive, forecasting activity.

When this happens, the changes captured in a second grid can be quite

complicated, and the change over the two grids rather ‘messy’ to analyse. The

following is less of a procedure and more of a short list of initiatives by which

this messiness can be handled.

(1) Elicit two grids in succession from the same interviewee. Elicit two grids

in succession from scratch, using the basic procedure outlined in Section 3.1.2.

(2) Before analysis, make sure that any elements and constructs which both

grids have in common are in the same position in both grids. In other words,

shuffle the columns and rows of the second grid around (with their ratings, of

course!) so that the elements are in the same columns in both grids, as you look

at the columns from left to right, and that the constructs are in the same rows

as you read from top to bottom in both grids. As the second grid is a free-form

one, you will have elements and constructs left over (see Tables 9.2 and 9.3).

(3) If there are sufficient common elements and constructs to make it

worthwhile, treat that portion or both grids as a simple change grid (see

Section 9.1.1). Whichever of the analysis procedures from Section 9.1.1 that

you decide to use, don’t stop at that point, but look for links between the

existing constructs and the new ones, as follows.

(4) Examine the function of the new constructs (and elements, if any), with

the interviewee, putting them into context with the existing ones.

(5) Explore the process by which some of the constructs have been dropped.

Steps 3, 4, and 5 are in fact different facets of the same activity, for you are

trying to build a shared understanding of how, why, and how far your

interviewee is thinking differently. You might find that it simplifies the task if

you transfer all the constructs onto a single working grid sheet (see Table 9.3).

(6) Focus, first and foremost, on the constructs and their meaning: what’s

being said, and what has changed?

Additionally, to tackle steps 4 and 5, youmay find it helpful towork with an organising

framework: somemodelwhichwillprovideyouwitha systematicapproachbased ona

goodrationale for howpeople change their construing, andwillprovide a first step into

theory, that is, into understanding why the interviewee is changing. At least three

models are available in personal construct theory.

(a) The experience cycle.Construing over time consists of a cyclemade up of several

stagesas follows: anticipation, investment, encounter, assessment, and constructive

Table 9.2 One interviewee’s two grids, A1 and A2


Construct Dr JF Mr PMcS Prof. AW Ms AK Dr LT Dr TN

A1.1 Clear, understandable 1 3 3 2 1 5 Difficult to follow

A1.2 Makes it interesting 1 2 3 4 1 5 Dull and boring

A1.3 Easy-going 2 5 4 1 3 4 Tense and preoccupied

A1.4 An all-rounder 2 4 5 1 4 3 Very specialised

A1.5 Lenient marker 2 2 3 1 5 4 Strict marker

A1.6 Good tutorial skills 1 2 3 2 5 4 Doesn’t know how to discuss

A1.7 Good delivery 2 1 1 2 3 5 Talks too quietly and mumbles

A1.8 Makes me laugh 1 1 5 2 4 5 Rather solemn

A2.1 Can understand them


1 3 1 1 1 4 Difficult to follow

A2.2 Good delivery 1 1 1 1 4 5 Talks too quietly and mumbles

A2.3 Good in tutorials 1 2 1 2 4 4 Can’t hold a discussion

A2.4 Makes it interesting 1 2 1 4 2 5 Dull and boring

A2.5 Relaxed 1 4 3 2 3 5 Tense, preoccupied with


A2.6 Accessible to students 3 1 2 1 5 4 Never there to take a problem to

A2.7 Gives obscure references

3 5 4 3 2 1 Recommended texts easy to find

The topic is, ‘how I feel about my lecturers’; A1 was obtained at the start of the second year and A2 at the start of the third year of a three-year


Table 9.3 Single working grid sheet to summarise changes between interviewee’s two grids


Construct Dr JF Mr PMcS Prof. AW Ms AK Dr LT Dr TN

A1.1 and A2.1 Clear, understandable

1,1,0 3,3,0 3,1,2 2,1,1 1,1,0 5,4,1 Difficult to follow

A1.2 and A2.4 Makes it interesting 1,1,0 2,2,0 3,1,2 4,4,0 1,2,1 5,5,0 Dull and boring

A1.3 and A2.5 Easy-going 2,1,1 5,4,1 4,3,1 1,2,1 3,3,0 4,5,1 Tense and preoccupied

A1.6 and A2.3 Good tutorial skills 1,1,0 2,2,0 3,1,2 2,2,0 5,4,1 4,4,0 Doesn’t know how to


A1.7 and A2.2 Good delivery 2,1,1 1,1,0 1,1,0 2,1,1 3,4,1 5,5,0 Talks too quietly and


Sum of Diffs 2 1 6 3 3 2

A1.4 An all-rounder 2 4 5 1 4 3 Very specialised

A1.5 Lenient marker 2 2 3 1 5 4 Strict marker

A1.8 Makes me laugh 1 1 5 2 4 5 Rather solemn

A2.6 Accessible to


3 1 2 1 5 4 Never there to take a

problem to

A2.7 rev. Recommended texts

easy to find

3 1 2 3 4 5 Gives obscure references

1. The purpose is to simplify the messiness in Table 9.2. Common constructs have been shunted to the top of the table. The ratings are shown in

order: grid A1, grid A2, absolute difference.

2. Where necessary (construct A2.7), the poles are reversed to show all positively evaluated poles on the left.

revision.Readmore on this in Kelly (1970). If you have a background inmanagement,

you may recognise a similarity to the Kolb cycle (active experimentation, concrete

experience, reflective analysis, and abstract conceptualisation); see Kolb (1984) for

the basics, and Greenaway (2002) for a handy comparison of several experiential

change models,Kelly’s included.

(b) The creativity cycle. Fresh ways of thinking require alterations in the implicational

tightness of the constructs being used to predict a situation.‘Implicational tightness’

can be described as follows. Suppose you have the following two constructs about

supervisors at work:

Makeme feel nervous ^ I can relax in their company

Tend to distract me ^ Don’t disturbmy concentration

If, for you, they are implicationally tight, you’ll find you can only do a good job of any

task in which you’re engaged by concentrating very hard when a supervisor who

makes you feel nervousis looking at you.

Youmight findit difficult tothinkofalternativewaysofhandlingthissituation, solongas

the tight link between these two constructs is maintained. Of course, you’ve

developed an implicationally tight link between these two constructs for a good

reason. It helps you to know what you have to do to be effective ^ you have to concentrate!

But the strong implicational connection gets in the way of alternative, and possibly

more effective, ways of coping, unless you loosen the tightness of this link. As soon

as you imagine that it may be possible not to be distracted by people who make you

feel nervous, it becomes possible to search for new constructs.

And so the creativity cycle consists of an alternation between tight construing (in the

service of effective action), loose construing (in order to search for alternatives), and

renewed tight construing (efficient predictions using the revised construct set). See

Winter (1992: 13^14; 258^264) on how therapists utilise this cycle.

(c) The C-P-C cycle, which stands for Circumspection, Pre-emption, and Control. The

model asserts that, when making up their minds about an issue, an individual will

firstly tend to engage in circumspection by looking round in their personal repertoire,

and examining various propositional constructs whichmay be relevant to the various

issues in a situation. S/he then construes pre-emptively as s/he concentrates on a

single issue. Finally, s/he engages in control, that is, decides which pole of that

construct to use and relate to. (Youmay need to refer to Section 5.3.3 at this point, to

refreshyourmemoryonpropositionalandpre-emptive construing.) SeeWinter (1992:

13^14; 247^250) formore details.

And so, ‘exploring the process’ becomes a matter of drawing on any of these

models to structure an approach to understanding your interviewee’s change.

I don’t want to be too prescriptive here; you should simply take your

interviewee’s feedback, and do what seems sensible to you both at the time.

However, you might like to draw on the models by considering one or other of

the following.

(6) Drawing on the C-P-C cycle, the experience cycle, and the creativity cycle,

identify the kind of change which is taking place in your interviewee’s

construing of the topic, and act accordingly. If the issues concern a decision,

you might draw on the C-P-C cycle as follows. Does the second grid represent

an attempt to be more circumspect, pre-emptive, or controlling than in the first

grid? By and large,

. if circumspect: do the new constructs cover broader ground, a greater

variety of issues, than the original ones?

. if pre-emptive: are the new constructs minor variations of the other

constructs in the second grid, with the ratings being rather similar? What

would constructs with very different meanings, and very disparate ratings,

mean in these circumstances?

. if control: are the new constructs more behaviourally specific than the ones

they replace? Can you see more extreme ratings on the new constructs,

reflecting a greater salience with respect to your interviewee’s situation?

Perhaps the creativity cycle is relevant. Is the second grid produced as part of

an attempt to tighten or loosen construing?

. Generally speaking, are the ratings in the second grid more closely or less

closely related to each other than in the first grid? Check this out for key

comparisons as follows:

. Take one of the constructs which appears in both grids (call it construct 1), a

construct which seems highly related in the original grid (call it construct 2),

and any construct (call it construct 3) that seems to have replaced construct 2

in the second grid. How similar are the ratings between construct 1 and

construct 2 in the original grid? How similar are the ratings between

construct 1 and construct 3 in the second grid? Is the second grid an attempt

to loosen (lower similarity scores for the two constructs being compared) or

tighten (higher similarity scores between the two constructs being compared)?

Something similar is done, in a more structured way, using just one grid form

in the Implications Grid. The interviewee is asked, construct by construct,

what changes would occur in other constructs in a grid if they changed the

ratings they gave on a particular construct. Implications grids are beyond the

scope of this account, but you can find more on the technique in Chapter 3 of

Fransella et al. (2004).

If you have carried out a principal components analysis (see Section 6.3.1) of

each of the two grids separately, you will be able to compare the variance

contributions of the components in each grid, and see in gross terms to what

extent the construing is loose (many components accounting for a given

variance total) or tight (few components accounting for that level of variance).

Alternatively, you could try one or more of the different kinds of index

analysis, as mentioned in Section 5.1.1.

The Experience Cycle model can be drawn on if the two grids appear to be

part of an ongoing process of reflective analysis: the sort of process described

by authors such as Hunt (1987), Thomas & Harri-Augstein (1985), and

Whitehead (1994), an activity increasingly included in the objectives of

managerial doctorate programmes such as the DBA.

Questions to ask here would include the following.

. What stage in the cycle is represented by the transition between the two

grids? What kinds of conformation or disconformation are involved, and

what sort of revisions are being made?

. Feelings expressed during elicitation of, and strength of feeling as

represented by ratings in, the second grid compared with the first would

be worth exploring as you examine the investment phase. The particular

difference represented by constructs dropped from the first grid, and new

constructs used in the second grid, might be used to explore the assessment

and constructive revision phases.

. Alternatively, if you prefer to work directly with the Kolb cycle model, it

might be useful to see the two grids as two stages in the process of transition

from concrete experience to abstract conceptualisation. This provides a

direct expression of reflective analysis, and you’d concentrate on the reasons

for replacing old constructs with new ones. Alternatively, you might view

the two grids as a summary of the transition between abstract

conceptualisation and active experimentation. If so, you might want to

discuss how behaviourally explicit, and predictively robust, the constructs in

the second grid were in comparison to those in the first.

At this point, you should tackle Exercise 9.2,

to get a feeling for the clear presentation

of what might originally look like a rather

messy situation.