Assignment: Action Research Survey

Action research by its very nature is participatory and should allow all interested parties to participate in voicing their opinions, concerns, or ideas. Survey creation is often an attractive method by which to collect information from groups of people (though it is not the only method). Since action research is often less formalized than other types of research, survey creation can be more casual and conversational. The goal is for teams or groups to collect enough information from the surveys that they can then begin the next phases of action research: interpreting and implementing.

In this Assignment, you develop a survey that could be used in an action research project.

To Prepare

Imagine that you are implementing a well-being program for the staff at the human or social services organization at which you currently work or one at which you might work in the future. Consider how you might use an open-ended survey to determine the needs for this program.

The Assignment (1–2 pages):

• Briefly describe the human or social services organization (real or hypothetical) for which you are creating your survey.
• Create a 10-item, open-ended question survey that you could use to determine the needs for a well-being program for the staff.
• Briefly explain why you selected each question for the survey.
• Finally, describe the medium you would use to deliver the survey.

Support your Application Assignment with specific references to all resources used in its preparation. You are asked to provide a reference list for all resources, including those in the Learning Resources for this course.

1
RESEARCH IN PROFESSIONAL

AND PUBLIC LIFE1
THE PURPOSES AND APPLICATIONS OF ACTION
RESEARCH: WHO DOES ACTION RESEARCH, AND
WHY DO THEY DO IT?
Action research is a systematic approach to investigation that enables
people to find effective solutions to problems they confront in their
everyday lives. Unlike experimental or quantitative research that looks
for generalizable explanations related to a small number of variables,
action research seeks to engage the complex dynamics involved in any
social context. It uses continuing cycles of investigation designed to
reveal effective solutions to issues and problems experienced in specific
situations and localized settings, providing the means by which people in
schools, businesses, community agencies and organizations, and health
and human services may increase the effectiveness and efficiency of
their work. In doing so it also seeks to build a body of knowledge that
enhances professional and community practices and works to increase
the well-being of the people involved.

For many people, professional and service occupations—teaching, social
work, health care, psychology, youth work, and so on—provide
appealing avenues of employment. These occupations have the potential
to provide meaningful and fulfilling work that is intrinsically rewarding.
Increasingly, however, people in these sectors find their work to be more
demanding and less satisfying. They often struggle to balance growing
demands on their time and energy as their workloads continue to
expand, and they are routinely confronted by problems rarely
encountered 20 or 30 years ago.

The pressures experienced in professional practice reflect tensions in
contemporary society. The complex influences that impinge on people’s
everyday social lives provide a fertile seedbed for a proliferating host of
family, community, and institutional problems. Professional
practitioners and agency workers are increasingly held accountable for
solutions to problems that have their roots in the deeply complex
interaction between the experiences of people and the realities of their
social lives: stress, unemployment, family breakdown, alienation,
behavioral problems, violence, poverty, discrimination, conflict, and so
forth.

Although adequately prepared to deal with the technical requirements of
their daily work, practitioners often face recurrent crises outside the
scope of their professional expertise. Teachers face children disturbed by
conflict in their homes and communities, youth workers encounter
resentful and alienated teenagers, health workers confront people
apparently unconcerned about life-threatening lifestyles and social
habits, and social and welfare workers are strained past their capacity to
deal with the impossible caseloads spawned by increasing poverty and
alienation.

There is an expectation in social life that trained professionals, applying
scientifically derived expertise, will provide answers to the proliferating
problems that confront people in their personal and public lives.
Community responses to crises that arise from drug abuse, crime,
violence, school absenteeism, and so on invariably revolve around the
use of a social worker, youth worker, counselor, or similar type of
service provider whose task it is to eradicate the problem by applying
some intervention at an individual or programmatic level. These
responses have failed to diminish growing social problems that have
multiplied much faster than human and financial resources available to
deal with them. Moreover, evidence suggests that centralized policies
and programs generated by “experts” have limited success in resolving
these problems. The billions of dollars invested in social programs have
failed to stem the tide of alienation and disaffection that characterize

many areas of social life in modern industrial nations.

If there are answers to these proliferating social problems, it is likely that
centralized policies will need to be complemented by the creative action
of those closest to their sources—the service professionals, agency
workers, students, clients, communities, and families who face the issues
on a daily basis. Centralized policies, programs, and services, I suggest,
should allow practitioners to engage the human potential of all people
who contribute to the complex dynamics of the contexts in which they
work. Policies and programs should not dictate specific actions and
procedures but instead should provide the resources to enable effective
action appropriate to particular places. The daily work of practitioners
often provides many opportunities for them to acquire valuable insights
into people’s social worlds and to assist them in formulating effective
solutions to problems that permeate their lives.

We therefore need to change our vision of service professionals and
administrators from mechanic or technician to facilitator and creative
investigator. This new vision rejects the mindless application of
standardized practices across all settings and contexts and instead
advocates the use of contextually relevant procedures formulated by
inquiring and resourceful practitioners. The pages that follow describe
some of the ways professional and community workers can hone their
investigative skills, engage in systematic approaches to inquiry, and
formulate effective and sustainable solutions to the deep-rooted
problems that diminish the quality of professional life. This volume
presents an approach to inquiry that seeks not only to enrich professional
practice but also to enhance the lives of those involved.

As a young teacher, I had the rare experience of being transferred
from the relative security of a suburban classroom to a primary
school in a remote desert region of Western Australia. My task was
to provide education for the children of the Aboriginal people who,
at this time, still lived as they had for millennia, moving through the
land in small family groups, hunting for their food, and sleeping in
leaf shelters. On my first day in class, I was confronted by a wall of

silence that effectively prevented any possibility of teaching. The
children refused to respond verbally to any of my queries or
comments, hanging their heads, averting their eyes, and sometimes
responding so softly that I was unable to hear what they said. In these
discomfiting circumstances, I was unable to work through any of the
customary routines and activities that had constituted my
professional repertoire in the city. Lessons were abbreviated,
disjointed, and seemingly pointless, and my professional pride took a
distinct jolt as an ineffective reading lesson followed an inarticulate
math period, preceding the monotony of my singular voice through
social studies.

The silence of the children in the classroom was in marked contrast
to their happy chatter as we walked through the surrounding bush
that afternoon, my failing spirits leading me to present an impromptu
natural science “lesson.” In this and following lessons, the children
taught me a great deal about the living bushland that was their
natural home—the small animals and birds that, though unseen by
my city-bred eyes, were everywhere; the plants, fruits, roots, and
berries that were edible; the places where water could be found
(precious commodity in this desert environment); and how to survive
when the weather was very hot. My “teachers” were a mine of
information, and they revelled in the opportunities to demonstrate to
me their knowledge about the environment and their skill in being
able to move so easily in what I saw as a hostile setting.

I discovered that I was able to make use of this knowledge and
interest in the classroom, fashioning a range of learning activities
related to literacy, mathematics, social science, and natural science.
In small ways I was thus able to accommodate my approach to
teaching in this unique educational environment, but the experience
endowed me with an understanding that the regular routines of
teaching were unsuited to my current circumstances. All the taken-
for-granted assumptions of my professional life rang hollow as I
struggled to understand the nature of the problems that confronted

me and to formulate appropriate educational experiences for this
vibrant, independent, and sometimes fractious group of students.
Texts, curricula, teaching materials, learning activities, classroom
organization, speech, interactional styles, and all other facets of
classroom life became subjects of inquiry and investigation as I
sought to resolve the constant stream of issues and problems that
emerged in this environment. To be an effective teacher, I discovered
that it was necessary to modify and adapt my regular professional
routines and practices to fit the lives of the children.

The legacy of that experience has remained with me. Although I
have long since left school classrooms behind, the lessons I learned
there still pervade all my work. I engage all professional,
organizational, and community contexts with a deep sense of my
need to explore and understand the situation. An attitude of inquiry
enables me to engage, examine, explore, formulate answers, and
devise responses to deal more effectively with each context I
engage—and the diverse experiences and perspectives of the people
within it.

In these situations, I now cast myself as a research facilitator,
working with and supporting people to engage in systematic
investigation that leads to clarity and understanding for us all and to
provide the basis for effective action. In many places in the United
States, Canada, Australia, East Timor, and Singapore I use
techniques and procedures that can be fruitfully applied to the day-
to-day work of people in schools, organizations, and community
settings. I am now a practitioner–researcher.2

As becomes evident in the following sections, action research is not
merely a tool for applying a standardized set of procedures to
professional, organizational, or community life, nor a “job” intended as
the provenance of trained researchers. Neither is it a superficial set of
routines that legitimate any set of social or professional practices. Far

from providing a set of fixed prescriptions to be applied in any context,
action research provides a flexible and practical set of procedures that
are systematic, cyclical, solutions oriented, and participatory, providing
the means to devise sustainable improvements in practice that enhance
the lives and well-being of all participants.

A Handbook for

Participatory Action Research,

Planning and Evaluation

Jacques M. Chevalier and Daniel J. Buckles

SAS
Dialogue

Copyright for A Handbook for Participatory Action Research, Planning and Evaluation is held by the
authors, Jacques M. Chevalier and Daniel J. Buckles. The work is licensed under the Creative Commons
Attribution-Noncommercial 2.5 Canada License, and is available in pdf format from our website at

www.participatoryactionresearch.net.

You are free to make a limited number of copies of A Handbook for Participatory Action Research, Planning
and Evaluation on condition that it is reproduced in its existing format, without reference to Third Parties,
and that these copies are not used for commercial purposes. If you are interested in producing multiple

copies of the handbook or purchasing additional copies, contact the authors at www.sas2dialogue.com for
permission and discussion of appropriate terms and conditions.

Copyright © Jacques M. Chevalier and Daniel J. Buckles (2011).

Correct Citation: Jacques M. Chevalier and Daniel J. Buckles. 2011. A Handbook for Participatory Action
Research, Planning and Evaluation. Ottawa, Canada: SAS2 Dialogue.

Copyright SAS
Dialogue

http://www.sas2.net

http://www.sas2.net

http://www.sas2dialogue.com

http://www.sas2dialogue.com

Module 6

Understanding Systems

119

SAS

Dialogue

12

0

SAS
Dialogue

Purpose To describe how people view a domain or topic area, and create new learning opportunities based on this understanding.

PRINCIPLES

The theory of human understanding underlying Domain Analysis is a social adaptation of Personal
Construct Psychology, a well-known theory in Psychology and the Cognitive Sciences developed in the

1950s by George Kelly. The key assumption is that people understand a domain by dividing it into
parts and creating a description of the whole based on comparisons (or degrees of similarity and
difference) between the parts. For example, to know the meaning of ‘tasty food’ requires not only a
sense of what ‘tasty foods’ have in common but also words and ideas to describe the opposite. In
Personal Construct Psychology, domain parts are called elements and the contrasting characteristics
are called constructs. The social adaptation presented below builds on this perspective by showing
how stakeholder groups create and organize elements and their contrasting characteristics for a domain or topic area. The method
uncovers ways people make sense of reality in a particular context and helps create opportunities for problem solving and learning.

Domain Analysis can be applied to any topic including things in nature (Ecological Domain), activities (Activity Domain), problems (Problem
Domain), stakeholder profiles (Social Domain), and options for action (Option Domain). Following are detailed instructions for the tool, which can

be adapted for these specific applications (see examples below). Information gathering and analysis can be done manually, as described below,
or using the software RepGrid (http://regrid.com).

Step 1 Define the domain or topic area and identify at least six elements and no more than 12 that belong to the domain. These should
be concrete, distinct and clearly defined. If the elements are vague, use the Laddering Down method in Active Listening to make
them more specific and meaningful. Ask “What do you mean by this?” or “Can you give an example of this?”. Another option is to

use description and storytelling to explore the topic, and then use this information to identify the elements. Write or draw each

element on its own card with a brief description on the back of the card.

Step 2 Decide on a rating scale with a range from 1-5 or 1-7 (see Scoring Tips). Create a table on the floor or wall with the term
‘Characteristics’ at the top of Column 1.

Step 3 If necessary, discuss or provide one key characteristic participants want to explain in light of a problem-solving exploration of the

domain. Write the key characteristic on a card, using one or two key words and give it a score of 1. Then, identify the opposite of
the key characteristic on the same card and give it a score of 5 (or 7). Place the card showing these two opposite characteristics
and the corresponding scores in the second row of the first column. (This step and the next two steps are optional.)

Step 4 Rate all the elements using the key characteristic and its opposite and the rating scale (from 1 to 5, for instance). Discuss the score for
each element until participants agree. Record each score on its own card and write the reason given for each score on the reverse side of
its card or on a flip chart. Place each score card in the row for the key characteristic, below the corresponding element.

Reconstructing models of reality

Domain

Analysis

12

1

SAS
Dialogue

http://re

http://re

Step 5 To facilitate interpretation of the table, reorganize all the elements in order based on the ratings given for the key characteristic.

Step 6 To elicit other characteristics from participants, choose three element cards from the top row at random. Identify two of them (a pair)

that are the same in some important way, and different from the third. Identify what it is these two elements have in common that is also

relevant to the topic. Write the characteristic on a new card and give it a score of 1. Then, identify the opposition or contrast that makes

the third element different from the pair. Write this opposite or contrasting characteristic on the same card and give it a score of 5 (or

7). Examples of opposite characteristics are: a good leader – an ineffective leader; reliable – unreliable; safe – risky; etc. Place the card

showing these two opposite or contrasting characteristics and the corresponding scores in the third row of the first column.

Step 7 Repeat the process described in Step 6 to identify other sets of opposite or contrasting characteristics and add a new row for each set.

Step 8 Rate all the elements using each characteristic and its

opposite and the rating scale created in Step 2. Discuss

the score for each element until participants agree.

Record each score on its own card and write the reason

given for each score on the reverse side of its card or on

a flip chart. Place each score card in its row, below the

corresponding element.

INTERPRETING THE RESULTS

Step 9 To interpret the results, start with a review of the process, including the way that participants interacted and reached decisions at each

step. Also review the substance of the exercise, including the topic that participants selected, the elements and the characteristics

identified, and the kind of information or knowledge used to rate the elements. Summarize the main points on a flip chart.

Step 10 Review the column scores that describe the elements. Look for obvious features such as whether the scores tend to be in the middle or

closer to the poles. Also look for the elements that have similar scores for most characteristics, including the key characteristic.

Summarize the characteristics they share and draw lines connecting elements with similar column scores to show that they are part of the

same cluster or family of elements.

Step 11 Review the row scores that describe the characteristics. Look for obvious features such as scores that vary little and others a lot, or

characteristics that are more meaningful compared to others. Also look for matching characteristics. There is a match between two or

more characteristics when row scores are similar or show an inverse relationship to each other. Summarize the matches and draw lines

connecting characteristics with similar (or inverse) row scores. Characteristics that match the key characteristic (identified in Step 3) can

help explain important aspects of the topic area.

Characteristics Conflict A Conflict B Conflict C Conflict D Conflict E Conflict F

Rarely (1)

Often (5)
1 1 2 3 5

5

Legal (1)

Personal (5)
4 5 3 2 1 2

Interests (1)

Values (5)
1 3 2 4 5 4

122

Reconstructing models of reality

Domain AnalysisSAS
Dialogue

RETHINKING THE ANALYSIS

Step 12 Modify, delete or add to the list of elements, characteristics, and scores at any time during the process.

Look for an extra characteristic and opposite if two elements that are very similar need to be distinguished from each other more

sharply. To do this, find a meaningful difference between the two elements. Use this difference to create a new characteristic and

its opposite and rate all the elements on this characteristic.

Look for an extra element if two characteristics that are closely matched need to be distinguished from each other more sharply.

To do this, find a new element within the domain that brings together the characteristics that are rarely matched. Insert the new

element in a new column and rate it for each characteristic and its opposite.

Step 13 Review and summarize key comments concerning the domain or topic made during

the exercise. Then identify the learning opportunity (see Learning Opportunities,

below) and develop a strategy to act on this understanding.

Be sure to review in detail the Scoring Tips. These are critical to proper application of

Domain Analysis.

TIPS ON ELEMENTS

Supply or negotiate some or all the elements or elicit them from the participants, depending on

the purpose of the exercise and the facilitator’s role.

The list of elements can include an ideal or a problematic element that can be compared with other elements.

TIPS ON CHARACTERISTICS (CONSTRUCTS)

Supply or negotiate any characteristic and its opposite or elicit them from the participants, depending on the purpose of the exercise and the

facilitator’s role.

When using characteristics to describe the elements, do not interpret the descriptions as statements of facts that are either right or wrong.

Statements about elements should be accurate only in the sense of truly reflecting people’s views and understanding of reality.

Characteristics should be relevant to the topic area, focused and clear. They should usually consist of concrete nouns, actions or verbs

ending in ‘–ing’ rather than abstract terms, qualities or ideas.

Characteristics and their opposites can include responses or concrete actions related to each element (see Problem Domain).

!

!

12

3

Reconstructing models of reality
Domain AnalysisSAS
Dialogue

TIPS ON CHARACTERISTICS (CONSTRUCTS) continued

If the characteristics are vague or sound like clichés, use the Laddering Down technique in Active Listening to make them more meaningful and

detailed. Ask “What do we mean by this?”, “Can we give an example of this?”, “How can we tell this?”, or “In what way is this true?”.

Don’t use negative phrases, such as ‘not legal’ to describe the opposite of or contrast with a characteristic such as ‘legal.’ Negative phrases tend to

be vague and meaningless. Opposites or contrasts phrased more precisely will describe people’s views on a domain in a more meaningful way.

If necessary, some of the characteristics may involve a single pole or reference point against which all the elements are rated. For example,

‘cost’, ‘importance’ , ‘priority’, ‘feasibility’ may go from low to high (see Option Domain).

If participants cannot identify what it is that two elements have in common or what makes the third element different from the pair, ask in

another way, apply the Laddering Down technique (see Active Listening), choose another three elements at random or choose two cards instead

of three.

You can use other elicitation tools to identify characteristics and their opposites, without comparing elements chosen at random. A simple

procedure is the catchall question: ‘Can you think of some new, different characteristic and its opposite?’ Another option is the full context

procedure: review all elements and find two that have a characteristic in common, and then the element that is the most different from these and

in what way. Use this procedure to identify one or more characteristic and its opposite. Another option is to use description and storytelling to

explore the topic (for example, by describing examples of success and failure), and then use this information to identify the elements as well as

their characteristics organized into opposites.

To identify several characteristics and their opposites in less time, divide all participants into groups of two or three. Ask each group to choose

three elements at random and to identify a relevant characteristic and its opposite. Collect these new characteristics and their opposites, discuss

and clarify their meaning, and group together those that are the same (see tips in Social Domain).

Don’t use a characteristic together with its opposite more than once. However, a particular characteristic can be used more than once if it is paired with

a different opposite characteristic (such as ‘legal’ as opposed to ‘personal’ in one case, and then ‘legal’ as opposed to ‘political’ in the other case).

Characteristics can be grouped together into appropriate categories supplied by the facilitator or created and defined by the participants (see Free List

and Pile Sort). They can also be ranked by order of importance. This will help with interpretation of the table at the end of the exercise.

TIPS ON RATING

If the characteristic and its opposite do not apply to an element, don’t provide a score. If a characteristic does not apply to many elements,

try rewording it or leave it out of the analysis.

If the scores for a characteristic and its opposite are nearly the same across all elements, redefine the characteristic or leave it out of the analysis.
124

Reconstructing models of reality
Domain AnalysisSAS
Dialogue

TIPS ON RATING continued

The rating of elements can be done without focusing
attention on the table. To do so, place a card representing

a characteristic and some distance apart another card

representing its opposite or contrast. Then take each
element card or an object representing the element and ask

participants to locate the element somewhere on the continuum

between the two characteristic cards. Convert this location into a
rating, and track the scores separately in a table or directly in RepGrid.

Repeat this exercise for each characteristic and its opposite.

TIPS ON INTERPRETING

When comparing elements, focus on those row characteristics and

relationships that are more important or interesting. Don’t assume that all

relationships are meaningful. This would be over-interpreting the results.

As noted in Step 5, use the ratings for the key characteristic (identified

in Step 3) to reorganize all element cards (row 1) and score cards (row

2) from the lowest score to the highest. The reorganized table will help
explain the key characteristic.

Group together similar elements by moving the columns around and
placing them side by side (use masking tape to stick the column cards

together). Do the same with matching characteristics, by moving the

rows around and placing them one above the other.

Where you find high matches between row scores or sets of characteristics and

their opposites, discuss whether one row set is an example or the effect of the

other row set, or if it has the same meaning or the same cause as the other set.

To focus on two characteristics and their opposites only, create a diagram by drawing a vertical line that crosses a horizontal line of equal

length. If your scale is from 1 to 5, write 1 and 5 at opposite ends of both the horizontal line and the vertical line; indicate what these minimum

and maximum scores mean. Write 3 where the two lines cross. For each element, locate the score for one characteristic and its opposite on the
horizontal line, and then the score for the other characteristic and its opposite on the vertical line. Connect the scores from the two lines, and

write the name of the element where they meet. The closer two elements are in the diagram, the more similar they are.

x Conflict A

Values

x Conflicts D, F

Interests

Personal, Interests

Legal, Interests

Personal, Values

Legal, Values

Legal

Personal

x Conflict

B

x Conflict

C

x Conflict E

125

Reconstructing models of reality

Characteristic 1 2 3 4 5 Characteristic

Good organizer Good listenerJohn S.

Domain Analysis
1

51

5
3
SAS
Dialogue

TIPS ON INTERPRETING continued

To help people participate actively in the analysis, prepare and distribute

copies of the element cards among the participants. Then ask participants

to identify other elements with row scores that are identical or very similar

to theirs. Give special attention to similarities in the key characteristic and

other characteristics important to the domain. Groups formed around

similar elements can then prepare and present a brief description of what

the elements have in common. Following this, all participants can discuss

the main differences observed between groups (see tips in Social Domain).

TIPS ON THE MATHEMATICS

The software RepGrid (http://repgrid.com/) performs the calculations

described below. The Focus command creates a cluster analysis.

Elements that have the most similar ratings are placed side by side.

Characteristics that are closely matched also appear side by side, with

inverse relationships converted into positive relationships. A diagram with

lines outside the table meeting at various points indicates the levels of

similarity between elements and between characteristics.

The PrinGrid command creates a graph with calculations based on

principal component analysis. The graph is a two dimensional

representation of multidimensional relationships among elements and

characteristics. Dots show the location of each element in relation to all

other elements and to characteristics represented by straight lines. The

shorter the characteristic line, the less the ratings for the characteristic

vary. Closer relationships between elements (dots), between characteristics

(lines), and between elements and characteristics are shown by their

distance from each other. The main horizontal line (principal component 1)

and vertical line (principal component 2) are summary variables for these

multidimensional relationships. The percentages at the end of each line

indicate the extent to which each component explains these

multidimensional relationships. (See examples.)
126

Reconstructing models of reality

100
90
80
70

100 90 80 70

SRI

ALYSSE

CLAIRE

MELISSA
W.

GENEVIEVE
DWAINE

MARTINE

ELAINE
EMILY
W.

FRANCOISE

LUCIE

EUNICE

JOCELYNE

ANDREW
IMARA
KATE
SHANNON
EMILY
K.

MELISSA

LOREDANA

NAVA
KARINE

WANGARI
YVES

One-to-one
discussion Group
discussio
Improvisation

Planning

Facilitation Analytisis

People-oriented Task-oriented
Face-to-face Distance

1 1 2 2 3 4 4 4 3 3 3 3 3 2 3 3 3 2 2 2 2 1 2 2
2 3 3 4 2 3 3 4 4 4 3 3 3 2 2 2 2 2 2 3 4 4 1 1
2 3 2 1 3 3 3 4 4 3 3 3 3 3 3 2 2 2 3 4 4 4 4 3
1 1 1 3 4 4 2 3 3 3 3 3 2 2 2 2 4 4 4 4 4 4 4 5
1 1 2 1 1 3 2 2 2 1 1 1 1 1 2 3 3 3 3 3 3 4 4 5

CLUSTER ANALYSIS

Level of similarity

Analysis of

collaborative

inquiry skills

2:
22.3%

1:
47.6%

Planning

Group
KARINE

NAVA
DWAINE

LOREDANA
WANGARI
YVES

At
distance

Analysis
MELISSA

Face-to-face

People-oriented

Facilitation

CLAIRE
SRI
MELISSA
W.
JOCELYNE
LUCIE
EUNICE
MARTINE
FRANCOISE

ELAINEEMILY
W.

KATEANDREW

IMARAALYSSE

One-to-one EMILY
K.
Improvisation SHANNON

GENEVIEVE Task-oriented

PRINCIPAL COMPONENT ANALYSIS

Domain AnalysisSAS
Dialogue

http://repgrid.com/

http://repgrid.com/

TIPS ON THE MATHEMATICS continued

To manually calculate the level of difference between two column elements, calculate the sum of differences (SD) between same-row scores

(leave out rows that have empty squares). Then calculate the total maximum difference for all scores (this is MS, the maximum score, minus 1,

multiplied by C, the number of row characteristics that got ratings). The level of difference between two elements is SD divided by the total

maximum difference for all scores multiplied by 100. To turn this level of difference into a percentage similarity score, subtract it from 100. In

other words: [100 – (SD x 100)] / [(MS-1) x C]. Using the table created in Step 7 as an example, the sum of differences (SD) between the recorded

scores for elements E and F is 2 and the total maximum difference is 12, or [(5 – 1) x 3]. This results in a difference of 16.7% (2/12 x 100).

Looking at it another way, the two elements are similar at a level of 83.3%.

To manually calculate the level of difference between two row characteristics, calculate the sum of differences (SD) between same-column

scores (leave out columns that have empty squares). Then calculate the total maximum difference for all scores (this is MS, the maximum

score, minus 1, multiplied by E, the number of elements that got ratings). The level of difference between two characteristics is SD divided by

the total maximum difference for all scores multiplied by 100. To turn this level of difference into a percentage similarity score, subtract it

from 100. In other words: [100 – (SD x 100)] / [(MS-1) x E. Using the table created in Step

7 as an example, the sum of differences (SD) between the recorded scores for the last

two rows is 14 and the total maximum difference is 24, or [(5 – 1) x 6]. This results in a

difference of 58.3% (14/24 x 100). Looking at it another way, the two elements are

similar at a level of 41.7%.

If the level of similarity between two sets of row scores is very low, this indicates an

inverse relationship. This means that if participants choose a characteristic at one

end of the continuum in one row then they tend to choose the characteristic at the

opposite end in the other row. When this happens, turn the inverse relationship into a

positive one by reversing all the scores in one row (from 2 to 4 or from 5 to 1, in a

scale from 1 to 5, for instance). Positive relationships are easier to interpret. For

instance, by reversing the scores for the last row in the table already presented, the

level of similarity between the last two rows is 83.3%.

127

Reconstructing models of reality
Domain AnalysisSAS
Dialogue

91

LEARNING OPPORTUNITIES

Domain Analysis helps to identify learning opportunities based on an understanding of multidimensional

relationships among elements and characteristics within the domain or topic area. Opportunities may involve

structural learning, communicational learning, temporal learning or adaptive learning. Understanding the nature

of the learning opportunity helps with development of an action strategy.

STRUCTURAL LEARNING

Convergence

There is convergence in the system when the row scores in the table are closely matched. In this case, most

characteristics can be regrouped into two categories that are opposite each other, with the elements falling

somewhere along the continuum from one set of opposites to another. If convergence in the system is limiting,

search for new elements that combine the characteristics in novel ways. Give special attention to novel ways of

combining elements with the key characteristic identified in Step 3 (see example in Activity Domain).

Polarization

There is polarization in the system when one group of elements has one set of column scores and the other group of

elements is opposite in all respects. In this case, most elements can be regrouped into two categories that are opposite

each other. If polarization in the system is limiting, search for new elements that combine the characteristics in novel

ways. Give special attention to novel ways of combining elements with the key characteristic identified in Step 3.

Dispersion

There is dispersion in the system when very few elements or characteristics are closely matched. This indicates that each element is

entirely different and there is no pattern in the system. If dispersion in the system is limiting, search for other elements or characteristics

that may be missing and needed to introduce some meaningful pattern into the system (see example in Social Domain).

Vagueness

There is vagueness in the system when the scores for the elements do not vary much. If this is limiting, search for the likely cause. Some

possibilities are: participants have very different views of the elements and negotiated the differences by assigning average scores;

participants emphasize the connections and similarities between the elements, not the differences; participants have limited knowledge of

the domain or topic area; the elements chosen are too general.

!
!

128

Reconstructing models of reality
Domain AnalysisSAS
Dialogue

COMMUNICATIONAL LEARNING

Disagreement

There is disagreement when people give very different scores to the same elements

using the same characteristics. To measure levels of agreement and disagreement

between two tables or sets of scores, total the differences between same-square

scores and divide this number by the total maximum difference between all squares

(this is MS, the maximum score, minus 1, multiplied by E, the number of elements

that got ratings). If disagreement is a limitation, identify the key area(s) of

disagreement and the likely causes. Continue discussion of the causes until the

scores reflect a common assessment of the situation.

To compare many characteristics and tables representing the views of different individuals or groups, reorder the row characteristics in each

table from top to bottom, with those at the top matching the ratings of the key characteristic identified in Step 3. These key matching

characteristics represent what each individual or group has in mind when thinking about important aspects of the topic. Then, look for key

matching characteristics that participants agree or disagree with across the sample. If the tables contain many characteristics, they can be

grouped into categories (see Tips on characteristics), reordered from top to bottom within each category, and then assessed for key match

agreements and disagreements across the sample within each category. The software RepGrid will also compare tables that contain some or all

the same elements and characteristics. Levels of agreement may be combined with levels of understanding (below) to produce the six possible

scenarios outlined in Disagreements and Misunderstandings.

Misunderstanding

There is misunderstanding when a party with a particular profile (such as men) fails to predict how a party with a different profile (such as

women) will rate certain elements. To measure levels of misunderstanding, each party must try to guess how the other party will rate the

same elements using the same characteristic(s). Then, total the differences between the original scores and the scores each group predicted

for the other. Divide this number by the total maximum difference for all squares (this is the maximum score minus 1, multiplied by the

number of elements). If misunderstanding is a limitation, identify the key area(s) and the likely causes of misunderstanding. Compare and

discuss the scores until a better understanding of each other’s views is created. Levels of understanding may be combined with levels of

agreement (above) to produce the six possible scenarios outlined in Disagreements and Misunderstandings.

Confusion

There is confusion among people when the parties use different elements or characteristics to describe the same domain or topic. If confusion is

a limitation, search for common elements or shared characteristics to create some basis for mutual understanding and agreement.

!
!

129

Reconstructing models of reality
Domain AnalysisSAS
Dialogue

TEMPORAL LEARNING

Instability

There is instability in the analysis when the way people view a domain or

topic and characterize its elements changes quickly or frequently over

time, without any clear justification. If instability is limiting, identify the

factors that may explain this. Look for elements or characteristics that are

more meaningful, or take more time to discuss the ratings or to gather

the information needed to complete the exercise.

Resistance to change

There is resistance to change when people become aware of specific

learning opportunities described above yet prefer to leave the views

expressed in their analysis unchanged. If resistance to change is limiting,

identify the factors that may explain this or take more time to discuss the

topic, the elements, and their characteristics. Note that elements and

characteristics (which reflect how people think) are generally more

difficult to change compared with element ratings (which reflect what

people think about the elements and characteristics).

ADAPTIVE LEARNING

Failure to predict

There is a failure to predict when experience and real events do not confirm

the characteristics and the ratings applied to the elements in the analysis. To

assess the predictive value of the analysis, select key characteristics and their

opposites, and then identify indicators that define the meaning of each

number on your rating scale. Collect reliable information on these indicators

related to each element to see if the characteristics are relevant and the

ratings are confirmed. If the failure to predict is limiting, change the ratings

or look for characteristics that have better predictive value.

130

Reconstructing models of reality

LESS
VISIBLE

V
project

Quality
control

Web
site

DEVELOP
NEW
TOOLS

INFORM

Planning

Data
bank

PROFITABILITY

INFORM

Management
 information

QUALITY
CONTROL

EFFICIENCY

File
processing

Modalities

MORE
VISIBLE

Decentralization

SPECIFY

SATISFY
CLIENTS

Support

Process

2:
24.3%

1:
40.1%

PRINCIPAL COMPONENT ANALYSIS: Assessment of Program Activities

Domain AnalysisSAS
Dialogue

Ecological Domain examines how people view existing elements in nature using terms and characteristics that participants choose and

negotiate. The tool may be used to classify things in nature (such as apple varieties or soil types) or ecological processes (such as indicators

of climate change). The understanding of the domain may help people innovate, solve problems or test views against experience or other

sources of knowledge.

Summary of this example: In March 2009 COPAGEN held in Dakar, Senegal a West African colloquium on strategies to preserve and

promote peasant varieties of food plants, partly in response to the spread of genetically modified organisms (GMOs) in Africa.

Participants tested Ecological Domain to see how the technique could help develop a strategic and methodical approach to promoting

local knowledge on peasant seeds (different from the conventional use of questionnaires and interviews). To start the analysis, the

participants identified six strategic and vulnerable food plants grown in their respective countries. They also identified a series of

characteristics and their opposites that reflected three basic questions: in what way are the plants strategic, what makes them

vulnerable, and what kind of action is being taken to preserve them. The results represented in the two graphs reveal that 4 of the 6

plants chosen by the participants are strategic because they produce rich and tasty food, serve multiple usages, and are vulnerable to

drought. Actions to preserve two of them (Niebe peas, Red sorghum) involve marketing measures and customary rules of farmer

behavior. The other two (Souna millet, Moutini millet) are preserved mostly through technical measures. By contrast, the remaining

varieties (Laboko yam, Red Fyfe wheat) are particularly vulnerable to being contaminated by GMOs. These patterns, represented in the

Principal Component graph, account for about 77% of the variance within the observed system (see percentages on the horizontal and

vertical axes). Considering these findings, participants decided to explore other actions to preserve plants vulnerable to drought.

Ecological Domain

131

Reconstructing models of reality

100
90
80
70

60

50

100 90 80 70 60

Niebe
peas
(Nim
San
Ugni)
Red


Sorghum

Moutini
(Local
Millet
Djelgogui)
Souna
Millet
Red
Fyfe
Wheat
(Canada)

Laboko
Yam

Market
strategy
to
preserve

ZAI
technique
to
preserve

Behavior
to
preserve Technique
to
preserve
(neem
oil)
Vulnerable
to
worms

Vulnerable
to
insufficient
water

Good
taste Rich
food,
mutilple
usages
Staple
food Resistant

Vulnerable
to
drought

Vulnerable
to
GMO

1 1 3 3 3 3
1 2 3 4 3 1
3 4 4 3 3 1
3 3 3 3 4 1
3 3 1 2 5 2
1 1 1 1 5 5

2:
24.7%

Vulnerable
to
drought
Red
Sorghum
Souna
Millet
Moutini
(Local
Millet
Djelgogui)
Niebe
peas
(Nim
San
Ugni)
Staple
food

Vulnerable
to
insufficient
water

Market
strategy
to
preserve
1:
53.2%

ZAI
technique
to
preserve

Vulnerable
to
worms

Behavior
to
preserve

Good
taste

Laboko
Yam
Vulnerable
to
GMO

Red
Fyfe
Wheat
(Canada)Technique
to
preserve
(neem
oil)

ResistantRich
food,
mutilple
usages

CLUSTER ANALYSIS

Levels of similarity

PRINCIPAL COMPONENT ANALYSIS

SAS
Dialogue

Activity Domain

Activity Domain examines how people view existing activities or

actions using terms and characteristics that participants choose

and negotiate. The tool may be used to identify different types

of actions or activities and explore associated levels of difficulty,

forms of knowledge, benefits, the values or skills involved, etc.

An understanding of the activity domain may help people

innovate, solve problems or test views against experience or

other sources of knowledge.

Summary of this example: In this organization, most

knowledge sharing (KS) activities fall into two categories. On

the one hand, KS that participants consider more useful to

their work (on the left hand side) includes ‘Structured

reflection’ (rated first), ‘Writing report articles’ and ‘Invited

guests’ (both rated second), and ‘Evaluation committee

meetings’ (rated third). These activities tend to be planned

(‘purposeful’) and are done episodically. They involve an

active sharing of information and filtered feedback on

existing projects. Except for ‘Writing report articles’, more

useful KS activities involve real-time teamwork. On the other

hand, more time and resources are dedicated to less useful

KS activities (on the right hand side) that are regular and

unplanned (byproducts). These activities include ‘Circulating,

posting and storing written information’ (rated fifth, the least

useful) as well as ‘Written/verbal reports’ (on conferences,

visits, etc.) and ‘Regular program staff meetings’ (both rated

fourth). Except for ‘Regular program staff meetings’, these

activities involve a passive sharing of knowledge, they are

done individually, not in real-time (sequentially), and they

contribute less to innovation. These patterns, represented in

the Principal Component graph, account for about 83% of the

variance

within the observed system (see percentages on the

horizontal and vertical axes). Based on this analysis,

participants plan to allocate more time to useful KS activities,

and do the less useful ones differently.
132

Reconstructing models of reality

100
90
80
70
60
50

40

100 90 80 70 60

Regular
program
staff
meetings

Circulating/posting/storing
written
information

Wrtten/verbal
re
conferences,
visits
Invited
guests

Evaluation
committee
meetings

Structured
reflection

Writing
report
articles

INDIVIDUAL

GROUP

SEQUENTIAL INTER

ACTIVE

CONTRIBUTES
LESS
TO
INNOVATION

CONTRIBUTES
MORE
TO
INNOVATION

PASSIVE ACTIVE

BY-PRODUCT

PURPOSEFUL

DIRECT
FEEDBACK

FILTERED
FEEDBACK

REGULAR EPISODIC
LESS
USEFUL
TO
OUR
WORK

MORE
USEFUL
TO
OUR
WORK

MORE
TIME-RESOURCES

LESS
TIME-RESOURCES

5 1 1 2 4 5 1
5 1 1 4 5 5 2
4 2 2 5 4 4 5
5 1 2 4 4 5 5
3 1 3 4 4 5 5
3 1 3 1 4 5 5
1 1 2 2 3 5 4
2 1 2 4 3 5 4
1 1 3 5 4 4 4

1:
60.5%

Circulating/posting/storing
written
information
Structured
reflection
ACTIVE
FILTERED
FEEDBACK
PURPOSEFUL
CONTRIBUTES
MORE
TO
INNOVATION
Evaluation
committee
meetings
GROUP

2:
22.6%

MORE
USEFUL
TO
OUR
WORK
LESS
TIME-RESOURCES

Invited
guests
Wrtten/verbal
re
conferences,
visits

INDIVIDUALWriting
report
articles

SEQUENTIALEPISODIC

MORE
TIME-RESOURCESRegular
program
staff
meetings

INTERACTIVE

REGULAR
BY-PRODUCT
DIRECT
FEEDBACK
CONTRIBUTES
LESS
TO
INNOVATION
PASSIVE

LESS
USEFUL
TO
OUR
WORK

CLUSTER ANALYSIS
Levels of similarity
PRINCIPAL COMPONENT ANALYSIS
SAS
Dialogue

Problem Domain examines how people view existing problems using

terms and characteristics that participants choose and negotiate. The

tool may be used to identify different types of problems, levels of

difficulty, responses adopted in the past, etc. The understanding of the

problem domain may help people innovate, find appropriate solutions or

test views against

experience or other sources of knowledge.

Summary of this example: About 25 representatives of

French-speaking African countries working on issues of

natural resource management (NRM) identified the most

frequent types of NRM conflicts occurring in their respective

countries, such as between pastoralists and agriculturalists,

elected locals and administrators, men and women, etc. They

also identified contrasting characteristics to describe these

conflicts. Each kind of conflict was rated against each

characteristic and its opposite, using a scale of 1 to 9. The

analysis showed that conflicts amongst agriculturalists and

between agriculturalists and pastoralists are the most

intense. Clashes between ethnic groups are also intense,

although less so. All of these conflicts usually involve

conflicts of status and interests and are addressed through

management solutions. By contrast, tensions between

funders and governments and between technical services

and pastoralists are much less intense, they involve conflicts

in power and ‘mission’, and they are addressed through

technical solutions. Patterns represented in the Principal

Component graph account for about 88% of the variance

within the observed system (see percentages on the

horizontal and vertical axes). Discussion focused on ways to

introduce management solutions in less intense conflicts,

and technical solutions in more intense conflicts, as

complements to current strategies.

Problem Domain

133

Reconstructing models of reality
100
90
80
70
60
100 90 80 70

Pastoralist
v.
agriculturalists
Agriculturalists
v.
agriculturalists

Ethnic
group
v.
ethnic
group
Pastoral
v.
agricultural
populations

Men
v.
women

Technical
services
v.
pastoralists

Funders
v.
government
Elected
locals
v.
administrators

Administrators
v.
village
heads

MANAGEMENT
SOLUTION

TECHNICAL
SOLUTION

STATUS POWER

INTERESTS MISSION
MORE
INTENSE LESS
INTENSE

1 1 1 1 1 8 9 4 2
1 2 1 2 3 4 9 9 8
1 4 4 3 5 7 9 8 8
1 2 4 6 6 6 5 3 8

CLUSTER ANALYSIS
Levels of similarity
Administrators
v.
village
heads

Elected
locals
v.
administrators
Funders
v.
government

Technical
services
v.
pastoralists
TECHNICAL
SOLUTION

LESS
INTENSE
POWER
MISSION

1:
72.7%

MANAGEMENT
SOLUTION

MORE
INTENSE

STATUS

INTERESTS

Pastoralist
v.
agriculturalists
Agriculturalists
v.
agriculturalists
Ethnic
group
v.
ethnic
group
Pastoral
v.
agricultural
populations
Men
v.
women

2:
15.1%

PRINCIPAL COMPONENT ANALYSIS
SAS
Dialogue

Option Domain examines how people view different proposed actions (options) using terms and criteria that participants choose and

negotiate. The tool may be used to identify different kinds of options, evaluate them on specific criteria, establish priorities, and

support decision making. The understanding of the option domain may help people innovate, solve problems or test views against

experience or other sources of knowledge.

Summary of this example: About 2000 artisanal fishers exploit

shellfish in the Common Fishery Zone of Ancud in central coastal

Chile. The Fund for Fisheries Research invited some 57 fishers,

officials and scientists to a two-day meeting to discuss better

fishery management strategies in the zone. Participants identified

seven possible actions together with seven criteria that could be

used to evaluate the proposed actions. A scale of 1 to 7 was

applied to each criterion. Participants noted that restricting access

to the fishery may not be costly but will take time, is less feasible

legally, and will generate some conflict, at least at the beginning.

Better enforcement measures, while more feasible legally, are not

going well and represent a costly, longer-term approach that

depends more on other actors. As for raising government funding,

this is and will continue to be difficult and depends on others. On

the whole, mobilizing support for better management practices and

forming representative bodies received the most favorable ratings.

These patterns, represented in the Principal Component graph,

account for about 70% of the variance within the observed system

(see percentages on the horizontal and vertical axes).

Option Domain

134

Reconstructing models of reality
100
90
80
70
60
100 90 80 70 60

Restrict
access

Mobilize
support
for
implementation
Form
representative
bodies
Restock
Rotate
fishing
effort
Effective
enforcement

Raise
government
funding

DEPENDS
MORE
ON
FISHERS

DEPENDS
LESS
ON
FISHERS

SHORT
TERM LONG
TERM

EASY

DIFFICULT

MORE
CONFLICT LESS
CONFLICT

LESS
FEASIBLE
(LEGALLY) MORE
FEASIBLE
(LEGALLY)
GOING
WELL

GOING
BADLY

LESS
COSTLY MORE
COSTLY

5 1 3 2 2 5 7
5 4 2 5 4 6 4
5 5 4 3 5 7 7
4 5 7 6 6 5 7
2 7 6 7 7 7 7
2 3 3 7 7 6 5
2 1 5 5 6 7 2

CLUSTER ANALYSIS
Levels of similarity

LESS
FEASIBLE
(LEGALLY)
MORE
CONFLICT
LESS
COSTLY
GOING
WELL

Restrict
access
Raise
government
funding
DEPENDS
LESS
ON
FISHERS

2:
26.7%

EASY

DEPENDS
MORE
ON
FISHERS

DIFFICULT

SHORT
TERM

1:
43.5%

Form
representative
bodies
Mobilize
support
for
implementation

GOING
BADLY

MORE
COSTLY
MORE
FEASIBLE
(LEGALLY)
LESS
CONFLICT

Rotate
fishing
effort
Restock

Enforcement

LONG
TERM

PRINCIPAL COMPONENT ANALYSIS
Enforcement
SAS
Dialogue

Social Domain examines how people view themselves and others using terms and characteristics that participants choose and negotiate.

The tool may be used to identify different groups or categories of stakeholders based on the types and levels of interests they have in a

project or program; the forms and levels of organization or power they can apply to a situation; the degrees and ways in which they are

trusted or viewed as legitimate by others; the actions or positions they take in a conflict; or the information, skills, values or leadership

styles they might apply in a situation. The understanding of the social domain may help people innovate, solve problems or test views

against experience or other sources of knowledge.

Summary of this example: Farmers grow tobacco on some 80,000

acres of agricultural land in Bangladesh, mainly under direct contract

with the British American Tobacco Company. While tobacco is a cash

crop for farmers, tobacco farming causes a wide range of

environmental, social and health problems in farming communities.

The Bangladesh non-governmental organization UBINIG is working

with tobacco farmers who have expressed a desire to move away

from tobacco into other kinds of farming. As it cannot work with all

households at the same time, the project needed to form subgroups

that could conduct and assess alternatives to tobacco. Social Domain

was used to design strategies that reflect different farmer profiles.

The exercise revealed that farmers were made up of households with

one of four profiles: young tobacco farmers; older farmers with small

areas of tobacco and food crops; tobacco traders with limited

tobacco production of their own and; older, land-rich farmers with

the flexibility to avoid tobacco farming. It also suggested that being

involved in the tobacco trade is particularly important to land-poor

farmers (such as Razzak, Azizul and Huq), giving them a distinct

profile that should be taken into account when evaluating

alternatives to tobacco production. These patterns, represented in

the Principal Component graph, account for about 78% of the

variance within the observed system (see percentages on the

horizontal and vertical axes). A plan was developed to monitor the

impact of alternatives to tobacco on the livelihood of households

with these four distinct profiles.

Characteristics Aminul Hakim Razzak Azizul Nazmul Alim Abu Taleb Huq Salam

No/little tobacco (1)

Large tobacco fields (5)
1 2 2 2 2 2 3 3 6

No/little farmland (1)

Big farm (5)
6 4 1 2 6 4 2 3 6

Few food crops (1)

Many food crops (5)
5 3 4 4 5 2 2 4 3

Rare tobacco trade (1)

Frequent tobacco trade (5)
1 1 5 4 2 1 1 6 6

Young (1)

Old (5)
3 6 4 2 4 6 5 4 2

Social Domain

135

Reconstructing models of reality

Large
tobacco
fields

HUQ

RAZZAK

AZIZUL

Frequent
tobacco
trade

Big
farm

Few
food
crops

No/little
tobacco
Old

ABU
TALEB
ALIM

HAKIM
Rare
tobacco
trade

1:
48.8%

2:
29.4%

YoungMany
food
cropsNAZMUL

AMINUL SALAM

No/little
farmland

SAS
Dialogue

ADAPT Social Domain can also be facilitated without the use of a
table, thereby focusing attention on the discussion and the

active engagement of participants in describing meaningful

similarities and differences between them. To achieve this,

Step 1 Divide all participants into random groups of three. Ask

each group of three to identify two people in the group

(a pair) that are the same in some way relevant to the

domain or topic, and different from the third. Find a

characteristic that is shared by the pair, and then the

characteristic that makes the third person different.

Step 2 Make a list of the distinctions between characteristics and

their opposites obtained from all the groups. Discuss and clarify the meaning

of each distinction. Group together the distinctions that are the same. Reduce

the list to 4 to 6 distinctions that matter the most in the domain or topic area.

To help interpret the results of the analysis, rank the pairs of characteristics in

order of importance (see Tips on characteristics).

Step 3 Each participant rates himself or herself on each characteristic and its opposite,

from 1 to 5. Ensure that participants have a common understanding of what the

numbers on the scale mean for each characteristic and its opposite, or develop

indicators. Each actor can record their ratings on a card showing the same

characteristics, in the same order, and with the same format (see example card).

Step 4 Ask each participant to find others that have cards with many row scores that are identical or similar (only one point apart

in most rows) to theirs. Give special attention to similarities in the rows that describe the most important characteristics.

Encourage all participants to compare their cards with others until groups or ‘families’ with similar profiles are formed.

Step 5 Groups formed around similar cards can then prepare and present to the whole group a brief description of the characteristics group

members have in common. When a group presents their profile, others groups can move closer if they feel they are similar in significant

ways or distance themselves if the differences are more important than the similarities. At the end of the exercise, participants should

discuss the main differences observed between groups and plan strategies that draw on different but complementary profiles.

Actor’s card: John S.

Characteristics 1 2 3 4 5 Characteristics

Good listener x Good speaker

Organized x Creative

Efficient x Committed

Rallying person x Visionary

Experienced x Adventurous

Social Domain

136

Reconstructing models of reality
SAS
Dialogue

Systems Thinking

Purpose To identify entry points into a system based on an assessment of how elements in the system interact to create specific

behaviors and situations.

PRINCIPLES

A system is a set of interacting and interdependent parts forming an integrated whole. Each part can best be understood in the context of

relationships with other parts and the whole system, rather than in isolation. System Dynamics helps understand how people define and

understand: 1) differences between parts of a system; 2) how parts interact with each other and relate to the whole and; 3) opportunities to

challenge and improve both the parts and the whole.

Efforts to think and act ‘holistically’ depend on how people divide and define the parts of the whole. System parts and their relationships cannot be

understood through universal categories that apply to all possible settings. They are always expressed with local color and meaning.

The method presented below is an adaptation of the input-output matrix used in economics to depict the interaction of sectors in an economy.

Following are detailed instructions for the tool, which can be adapted and applied to any topic, including systems in nature (Ecological Dynamics),

activities (Activity Dynamics), problems (Causal Dynamics), skills (Skill Dynamics), stakeholder behaviors (Network Dynamics), values (Value

Dynamics) and social systems involving the interaction of actors, problems and actions (Social Dynamics) (see examples below).

Step 1 Define the topic area and identify the key elements or component parts of the system involved (see Free List and Pile Sort). These

should be concrete, distinct and clearly described. If the elements are vague, use the Laddering Down method in Active Listening to

make them more specific and meaningful. Ask “What do you mean by this?” or “Can you give an example of this?”. Another option is to

use description and storytelling to explore the topic, and then use this information to identify the elements. Write or draw each

element on its own card, with details on the back of the card or on a flip

chart. When using a standard matrix (see Tips, below, for alternatives),

make a copy of each element card.

Step 2 Create a table on the floor or wall. Place one set of element cards in

the top row and the other set (showing the same elements in the

same order) in the first column.

Step 3 Decide on a rating scale to indicate the level of contribution that

each element makes to other elements (for example, from 0 for no

contribution to 10 for a critical contribution).

Elements A B C D

Total

Contribution

A

x

B x

C x

D x

Total

Dependence

System Dynamics

137

SAS
Dialogue

Step 4 Use the scale created in Step 3 to rate the level of contribution that each element currently makes to each other element. Ask ‘At what

level does this (name the column element) contribute to that (name the row element)?’ Clarify the question and adapt it to the topic (see

specific applications of System Dynamics). As in all rating exercises, the same score can be given to two or several elements.

Proceed with the rating exercise one column after another. Start by rating the extent to which element B contributes to the element

heading the column A. This will ensure that the direction of the contribution is clear and consistent. If participants invert the question

and indicate how A contributes to B, insert the score in the appropriate cell and return to the questioning by column.

Record each score on its own card and write the reason given for each score on the reverse side of its card or on a flip chart. Place

the score cards in the appropriate rows and columns of the table. Leave empty all cells that combine an element with itself (A

contributes to A), unless the element interacts with itself (as do members within a stakeholder group, for instance).

Step 5 Once the table is complete, total all scores in each row and write Total Contribution

at the top of a new column to the right. Insert the total scores in this new column, in

the appropriate rows. The column shows the total contribution of each row element

to all other elements. (A different term for this column is used in Ecological

Dynamics, Causal Dynamics and Network Dynamics.)

Step 6 Total all scores in each column and write Total Dependence at the beginning of a

new row below. Insert the total score in this new row. This indicates the total

dependence of the column element on all other elements. (A different term for this

sum is used in Ecological Dynamics, Causal Dynamics and Network Dynamics.)

Step 7 Calculate the dynamic interaction between all elements by totaling all contribution

scores (or dependency scores) and dividing the result by the maximum total score that

could be obtained if all cells in the row (or the column) received the highest rating in the

range. Insert the resulting percentage figure at the bottom of the last column.

Step 8 Create a diagram by drawing a vertical line that crosses a horizontal line of equal length. Write or draw a symbol representing the

topic (identified in Step 1) above the diagram. Write at opposite ends of the vertical and horizontal lines the minimum score (usually 0)

and the maximum possible score that could be obtained if all cells in a row or column received the highest rating in the range (for

instance, the maximum total score that can be obtained with four elements interacting, using a scale of 0 to 10, is 30). Insert the

number that represents the middle score (the sum of maximum scores in a row divided by two) where the lines cross. The vertical

line indicates the Total Contribution of an element (its row total) and the horizontal line, its Total Dependence (or column total).
138

Systems Thinking

System DynamicsSAS
Dialogue

Step 9 Label the four corners of the diagram with the result obtained by combining the possible outcomes of each axis: elements that contribute and

depend more (top right); those that contribute more and depend less (top left); those that contribute less and depend more (bottom right); those

that contribute and depend less (bottom left). To facilitate the analysis, find an idea or a symbol to represent each corner of the diagram. Elements

that contribute and depend less may be important even if they interact little with other elements in the

system.

Step 10 To locate each element in the diagram, mark where the element’s total contribution score is located on the vertical line and the

element’s total dependence score is located on the horizontal line. Draw a line from each location and insert the name of the element

where the two lines meet.

Step 11 Include in the diagram other information that may be useful for the analysis, such as the overall level of control that stakeholders have over

each element in the system, the time and level of effort it would take to act on it or the order in which people plan to act on certain elements.

Use a code (such as capital letters, numbers, colors or circles) to identify elements with these characteristics (see examples below).

Scores that contradict the main tendencies of the diagram may also

be important and affect the interpretation of results; one element that

contributes little to other elements may still contribute a lot to one

important element, for instance. To identify these contradictory

scores, compare each cell score appearing in the rating table with the

average row score to see if both scores are on the same lower side or

upper side of the middle point of the scale (5 in a scale of 0 to 10, for

instance). If a cell score is not on the same side as the average row

score, compare the score with the average column score to see if both

scores are on the same lower side or upper side of the middle point of

the scale. If the cell score is not on the same side again, draw a circle

around it. Once these contradictory scores are identified, draw arrows

in the diagram to indicate the relationships that contradict the main

tendencies of the system. Use continuous arrows for scores above

the middle point of the scale. These indicate bottom-side elements

that contribute significantly to some elements located on the left side

of the diagram (see example in Skill Dynamics). Use broken arrows

for scores below the middle point. These indicate upper-side elements

that do not contribute significantly to some elements located on the

right side of the diagram (see example in Causal Dynamics).

139

Systems Thinking
System Dynamics

Depends

a lot

Contributes

little

Contributes a lot

Depends

little

Element A

Contributes a lot

Depends little

Contributes little
Depends little
Contributes a lot

Depends a lot

Contributes little
Depends a lot
SAS
Dialogue

INTERPRETING THE RESULTS

Step 12 Discuss the overall level of dynamic interaction of the elements calculated in Step

7 and review the location of the elements in the diagram, considering three

possible scenarios: integration, hierarchy or dispersion.

There is integration in the system when many elements are located in the top-

right section of the diagram. This usually reflects a high score for dynamic

interaction (above 60%, as calculated in Step 7). In an integrated system

increasing or decreasing the contribution of one element in the top-right

section may in turn affect the level of contribution of all other elements located

in the same section. The result is a chain effect that influences the dynamic

interaction of all elements, including the element that receives initial attention

(see example in Causal Dynamics).

There is hierarchy in the system when the diagram consists mostly of top-left

elements and bottom-right elements. This usually reflects a middle score for

dynamic interaction (between 40% and 60%, as calculated in Step 7). In a

hierarchical system, attention to elements in the top-left section will automatically

have an influence on the bottom-right elements (see example in Social Dynamics).

There is dispersion in the system when the diagram consists mostly of elements in the bottom-left section of the diagram. This

usually reflects a low score for dynamic interaction (below 40%, as calculated in Step 7). Elements in this section may be important

even if they interact little with other elements in the system. In a dispersed system, the elements interact little and can only be

modified through direct actions (see Activity Dynamics).

Step 13 Summarize the scenario or combination of scenarios that best describe the results in the diagram. Discuss the way that

participants reached decisions at each step, the elements included and left out of the analysis, the kind of information or

knowledge used to rate the elements, the contradictions identified and the other information added in Step 11. If need be, modify

one or several elements considering the discussion, and recalculate the overall interaction of all elements (see Step 7). When

completed, use this analysis to identify system entry points, rethink priorities or modify some elements so that they interact

differently with the other elements.

140

Systems Thinking
System DynamicsSAS
Dialogue

TIPS

Be sure to review in detail the Tips for Free List and Pile Sort, Ranking and Rating.

These are critical to proper application of System Dynamics.

The elements used in System Dynamics can be real or proposed.

If some elements have a negative impact on other elements, use a scale that has

negative scores (from –10 to 10, for instance; see Ecological Dynamics). Negative

scores reflect conflict in the system.

To focus on the rating discussion rather than the table, use a flip chart to represent

each column element. On each flip chart place the rating cards that indicate the

contributions other elements make to the flip chart element. Once the flip charts are

completed, compile the scores in a table and go on directly to the diagram in Step 8.

Another option is to make only one set of element cards and place these in a column in

plain view of all participants. When discussing the elements, move the top card to one

side and begin by asking to what extent do the remaining column cards contribute to the element set to one side. Continue this line of

questioning down the column, always referring to the isolated element card. Once these relationships have been scored and recorded in a

table, return the top card to the column and pull out the next element card. All cards remaining in the column can then be discussed as

elements contributing to the isolated card. Continue until all interactions have been assessed and recorded. Once the scores are compiled

in a table review the process and go on to the diagram in Step 8. This procedure lends itself to a direct conversational style of facilitation

focusing on rating of the elements rather than the construction of a table. It also makes it easier to use objects instead of element cards,

and work in a smaller space.

To compare current levels of interaction between elements with levels people are aiming for in the future, divide each cell of the table

created in Step 2 into two parts and insert a score in each part: the first score to describe the actual contribution that an element makes to

another, and the second score to describe the ideal contribution it should make.

141

Systems Thinking
System DynamicsSAS
Dialogue

142

Elements
A B

C

D

Total Contribution

A
x

B
x
C
x
D
x

Total Dependance

Systems Thinking
System DynamicsSAS
Dialogue

Ecological Dynamics helps describe how the components of an ecological system interact with each other. The tool may be used to support

systems thinking concerning things in nature (such as plant species and varieties) or ecological processes (such as soil degradation or the

dynamics of pollution). The understanding of the system may help people decide where to focus attention and what relationships to change.

Ecological Dynamics begins by defining an ecological system and listing the components of the system. The rating scale can include negative as well as
positive values (for example, – 10 to + 10). It focuses on the extent to which one component provides benefits to or harms other components in the
system, and the extent to which each is helped by or harmed by other components. These can be seen as relations of cooperation (each component
derives a benefit) or relations of exploitation or competition (each component benefits
at the expense of the other). When rating, ask ‘To what extent does this component
(name the row component) provide benefits to or harm that component (name the
column component)?’ When both situations apply, estimate the net effect. The resulting
matrix produces an index for helps/harms other components (vertical axis) and an
index for is helped by/harmed by other components (horizontal axis). See System

Dynamics for

detailed instructions.

Elements Rice Maize Sorghum Barbaty bean Pearl millet Black gram Sesame Pigeon pea Green gram Total
Contribution

Rice x 0 0 0 0 -3 0 0 -3 -6

Maize -2 x 0 5 2 0 3 0 0 8

Sorghum -3 0 x 5 0 -4 0 -4 -4 -10

Barbaty bean -5 -3 0 x 0 -3 -4 -2 -3 -20

Pearl millet -4 -5 0 5 x -3 0 -4 -3 -14

Black gram -3 0 0 0 0 x 0 0 0 -3

Sesame -5 2 -3 -5 -5 0 x -1 0 -17

Pigeon pea -5 0 -2 4 -2 0 0 x 0 -5

Green gram 0 0 0 -3 0 0 0 0 x -3

Total

Dependence
-27 -6 -5 11 -5 -13 -1 -11 -13 -70

Ecological Dynamics

143

Systems Thinking
SAS
Dialogue

Summary of this example: In this Indian mixed cropping

system the most important crops (marked with circles) are

rice, pigeon peas, and sorghum. The analysis shows that some

crops interact in positive ways. For instance, maize generally

affects other crops positively and is also positively affected by

sesame cultivation. Also, the growth of barbaty bean vines

benefits significantly from climbing on the stalks of maize,

sorghum, millet, and pigeon pea (see arrow). On the whole,

however, the diagram indicates that most crops affect other

crops in slightly negative ways. Farmers reduce these

exploitative relationships by adjusting how much of each crop

they sow. For example, they may increase the ratio of rice in

their field while reducing the ratio of pearl millet. They also

assume that unpredictable environmental factors will cause

some crops to produce little or fail. When this happens,

competition is also eliminated, allowing the remaining crops

to produce better. (Source: Colin Lundy, 2006. Growing Seed

Knowledge: Shifting Cultivation and Agricultural Biodiversity

among Adivasi Communities in India. MA Thesis in

Anthropology, Carleton University, Ottawa)

Harms a lot

Helped

a lot

Harmed

a lot
Sorghum

– 30

Harms a lot
Harmed a lot

Helps a lot

Helped a lot

Harms a lot
Helped a lot

Helps a lot

+ 30

– 30

+ 300

Helps a lot
Harmed a lot

Rice

Maize

Barbaty bean

Pearl millet

Green & black gram

Sesame

Pigeon pea

Ecological Dynamics

144

Systems Thinking
SAS
Dialogue

Causal Dynamics

Causal Dynamics helps assess how factors related to a key problem interact. The tool may be used to support systems thinking

concerning how to act on a problem through particular factors in the system (entry points).

Causal Dynamics focuses on relationships of cause and effect rather than relations of contribution and dependence explored in most other

applications of System Dynamics. It begins by defining a key problem and listing the factors involved. Include the key problem in the rating

matrix if it interacts with other factors directly. Leave the key problem out of the rating matrix if the factors are manifestations or examples of
the key problem.

When rating, ask ‘To what extent does this (name the row element) cause that (name the column element)?’ or ‘At what level does this (name the

row element) produce that (name the column element) as a consequence?’ The resulting matrix produces a cause index at the end of each row in
the table (vertical axis in the diagram) and an effect index at the bottom of each column (horizontal axis in the diagram). Label the four corners of

the diagram with the result obtained by combining the possible outcomes of each axis: factors that are pure causes of other factors (upper-left

corner of the diagram), factors that are pure effects of other factors (bottom-right corner), factors that are both causes and effects (upper-right
corner) and factors that are independent of each other (lower-left corner). See

System Dynamics for detailed instructions.

ADVANCED VERSION

Apparent and Real Weight

Some factors at the root of a key problem may have to be addressed directly even if they interact with other factors. To identify these,

distinguish between the apparent and real weight of each factor.

After defining the key problem and identifying the factors involved (Step 1), estimate how important each factor is in relation to the key
problem. This is the apparent weight of each factor, and reflects initial thinking about the weight of factors in a given context. Estimate

the apparent weight using a rating scale of 1 to 10 and write the result in the corresponding cell in the top row of the table and the sum
in the last cell. Factors with apparent weights of less than 3 are very weak causes of the key problem and should be left out of the analysis.

Complete Steps 4 to 8 and then revisit the weight of each factor. Estimate how important the factor would be if all the other factors were

eliminated or did not exist. This is the real weight of each factor and reflects thinking informed by the rating exercise regarding the
weight of each factor in isolation from other factors included in the analysis. Use the same rating scale, making sure that the real weight is

less than or the same as the factor’s apparent weight. Write the score next to the apparent weight in the corresponding cell in the top row

of the table and the sum in the last cell.

Complete other steps including a diagram with the results (Steps 8 to 11). Review the apparent and real weight for each factor and adjust the

size of the dot assigned to each factor. Use bigger dots when the real weight of a factor is the same or close to its apparent weight as

this indicates that it will remain significant even when other factors are eliminated. Give special attention to these factors when interpreting
the results. Factors that do not loose much of their real weight when other factors are addressed are persistent causes and may require more

direct attention than initially thought.
145

Systems Thinking
SAS
Dialogue

Causal Dynamics

146

Systems Thinking

Factors Poor KM* Lack of innovation Quantitative approach RBM* Weak partnering Poor HRM* Donor dependency Cause Index**

Weight: apparent, real 10, 2 7, 4 7, 4 6, 4 7, 6 8, 3 6, 5 51, 28

Poor KM x 0 8 8 8 2 2 28/60

Lack of innovation 10 x 6 8 4 4 2 34/60

Quantitative approach 10 10 x 4 4 10 0 38/60

RBM* 8 2 2 x 4 4 0 20/60

Weak partnering 8 8 8 0 x 8 0 32/60

Poor HRM* 8 10 10 4 8 x 6 46/60

Donor dependency 6 4 6 10 4 6 x 36/60

Effect Index 50/60 34/60 40/60 34/60 32/60 34/60 10/60 234/420

* RBM = Result-Based Management. KM = Knowledge Management. HRM = Human Resource Management.

** The Cause Index and the Effect Index correspond to the total (factor) contribution to and dependence on other factors, respectively.

Summary of this example (see next page): This organization feels that the way it manages knowledge is
not as useful to its members as it should be. Using the Causal Dynamics technique (and a rating scale of
0 to 10), participants choose to focus on the key factors in the top right of the diagram — factors that are
both causes and effects of the problem. They discover that their non-strategic management of human
resources (poor human resource management strategy) is a major contributing factor. Since they have
some control (marked in green) over this factor, they decide to free up some resources and use them to
innovate in the field of Knowledge Management (KM). They can innovate despite their donor’s accounting
approach to KM and the organization’s overemphasis on periodic accounts of measurable results, factors
over which they have little control (marked in red). Once these initial actions (numbered 1 in parentheses)
are taken, the organization will explore better ways to involve their partners in KM activities, a goal that
will take time. Other objectives, such as rethinking the organization’s dependence on a principal donor,
are less urgent. In the long run, the organization may want to act on this independent factor directly or
through causes not identified in this analysis.

SAS
Dialogue

Factor Integration Level

Step 7 in System Dynamics involves the calculation of the

dynamic interaction between all elements. In the advanced

version of Causal Dynamics this calculation may be influenced

by persistent factors (factors with a real weight that is similar to

its apparent weight). To calculate the Factor Interaction Level

(FIL), multiply the Total Cause Index % (the percentage figure at

the bottom of the last column) by the Total Real Weight

Reduction. The Total Real Weight Reduction is the Total

Apparent Weight (the sum of all apparent weights recorded in

the last column) minus the Total Real Weight (the total of all real

weights recorded in the last column), divided by the Total

Apparent Weight. In short: FIL = Total Cause Index % x (Total

Apparent Weight – Total Real Weight) / Total Apparent Weight. In

the example provided, the Total Cause Index % is 55.7%, or

234/420. The Real Weight Reduction is 45.1%, or (51 – 28)/51.

Thus the Factor Interaction Level is about 25%, or 55.7% x 45.1%,

a moderate FIL. This measure helps to guide interpretation

considering the three possible scenarios described under System

Dynamics: integration, hierarchy and dispersion.

Causal Dynamics

147

Systems Thinking

Legend: The size of each dot indicates the real weight of the

factor. Green means participants have some control over the

factor; red indicates little or no control. Numbers in parentheses

reflect the order in which participants plan to act on each factor.

Broken arrows indicate a weak causal relationship (contradicting

main tendencies in the diagram).

Cause Index

high

Effect

Index

high
Effect
Index

low

Causes Causes & effects

Cause Index low

60

600 30

0 EffectsIndependent factors

Poor HRM strategy (1)

Lack of innovation (1)

Quantitative
approach (1)

Poor KM

RBM (3)

Donor dependency (4)

Weak partnering (2)

SAS
Dialogue

Activity Dynamics

Activity Dynamics helps

describe how activities

in a project or program

interact with each other.

The tool may be used to

support systems

thinking concerning

how to increase synergy

among activities and

improve the overall

efficiency and

effectiveness of the

system.

Activity Dynamics begins by defining a set of actions, a

project or a program and listing the activities involved.

It focuses on the extent to which one activity contributes

to or depends on other activities. When rating, ask ‘To

what extent does this activity (name the row activity)

contribute to that activity (name the column activity)?’ The

resulting matrix produces an index for contributes to

other activities (vertical axis) and an index for depends

on other activities (horizontal axis). See System

Dynamics for detailed instructions.

Summary of this example: This project involves research
and action mostly, with some training. On the whole, the
interaction between the corresponding activities is very weak;
each activity makes a limited contribution to other activities.
Data collection and analysis contributes the most, and
lobbying depends the most on other activities. Changing how
these activities are carried out could increase synergies.

ACTIVITIES RESEARCH ACTION TRAINING Total
ContributionData collection/

analysis

Publishing Reports Green

manure

Local

initiatives

Lobbying

Data collection/analysis x 8 6 1 3 8 3 29/60

Publishing 0 x 2 1 1 6 0 10/60

Reports 0 2 x 0 0 2 0 4/60

Green manure experiments 2 3 5 x 2 3 1 16/60

Local initiatives 1 2 4 2 x 4 2 15/60

Lobbying 0 0 0 2 4 x 7 13/60

Training 0 0 0 6 3 0 x 9/60

Total Dependence 3/60 15/60 17/60 12/60 13/60 23/60 13/60 96/420

Contributes little

Depends
a lot

Depends
little

Contributes a lot
Depends little

Contributes a lot
Depends a lot

Contributes a lot
60
600 30

0

Contributes little
Depends a lot

Contributes little
Depends little

Data
collection/analysis

Lobbying
Local initiatives

Training

Publishing

Green manure

Reports

148

Systems Thinking
SAS
Dialogue

Skill Dynamics

Skill Dynamics helps assess how each skill applied to a set of activities,

project or program contributes to other skills and depends on them at

the same time. The tool may be used to support systems thinking

concerning the skills required in a situation and how to mobilize and

create synergies between the skills of different actors in the system.

Skill Dynamics begins by defining a set of activities, a project or a

program and listing the skills involved. It focuses on the extent to

which one skill contributes to or depends on other skills. When rating,

ask ‘To what extent does this skill (name the row skill) contribute to

that skill (name the column skill)?’ The resulting matrix produces an

index for contributes to other skills (vertical axis) and an index for

depends on other skills (horizontal axis). See System Dynamics for

detailed instructions.

Summary of this example: Training and analysis (circled) are the skills with the
highest levels of satisfaction. Together with writing they contribute the most to
other skills. Skills in theory and visual design are helpful when doing analysis,
and languages are helpful when doing networking (see arrows in the graph). By
contrast, networking skills contribute little to other skills in this system.

Skills Analysis Training Languages Networking Visual design Theory Writing Total

contribution

Analysis x 8 2 0 7 9 8 34/60

Training 5 x 4 8 5 6 9 37/60

Languages 1 9 x 9 1 2 8 30/60

Networking 0 4 3 x 0 0 0 7/60

Visual design 6 9 0 0 x 4 6 25/60

Theory 7 5 0 2 4 x 7 25/60

Writing 7 6 7 3 1 7 x 31/60

Total dependence 26/60 41/60 16/60 22/60 18/60 28/60 38/60 189/420 (45%)

Contributes little

Depends
a lot

Depends
little
Contributes a lot
Depends little
Contributes little
Depends little
Contributes a lot
Depends a lot
Contributes little
Depends a lot
Contributes a lot
60
0
600 30
Analysis

Theory

Training

Networking

Visual design

WritingLanguages

149

Systems Thinking
SAS
Dialogue

Network Dynamics

Network Dynamics helps assess the network of influence, trust or information that exists between stakeholders involved in a

particular situation or project.

Network Dynamics begins by defining a situation or project and listing the stakeholders involved. It focuses on one kind of network

at a time (influence, trust or information) and assesses the extent to which one stakeholder networks with other stakeholders. See

System Dynamics for detailed instructions.

A network of influence (or power) is a set of connections where people use their prestige, wealth, knowledge or position to affect other people’s

decisions. When rating, ask ‘To what extent does this stakeholder (name the row stakeholder) influence that stakeholder (name the column

stakeholder)?’ The resulting matrix produces an index for influences others (vertical axis) and an index for is influenced by others (horizontal axis).

A network of trust is a set of connections where people show confidence in other parties and rely on them to provide support, to behave

in appropriate ways, and to do what they are expected to do. When rating, ask ‘To what extent does this stakeholder (name the row

stakeholder) trust that stakeholder (name the column stakeholder)?’ The resulting matrix produces an index for trusts others (vertical

axis) and an index for is trusted by others

(horizontal axis).

A network of information is a set of connections where people pass on knowledge or views to other people. When rating, ask ‘To what

extent does this stakeholder (name the row stakeholder) provide information to that stakeholder (name the column stakeholder)?’ The

resulting matrix produces an index for informs others (vertical axis) and an index for is informed by others (horizontal axis).

Stakeholders Small
farmers

Municipal
authorities

Ranchers
association

Agricultural
laborers

Catholic
Church

NGO Teachers Trusting

Small farmers x 2 1 2 7 5 5 22/42

Municipal authorities 3 x 6 1 4 0 2 16/42

Ranchers association 4 7 x 3 2 0 3 19/42

Agricultural laborers 2 0 2 x 6 4 4 18/42

Catholic Church 6 5 5 7 x 5 7 35/42

NGO 5 0 0 3 3 x 2 13/42

Teachers 5 4 4 7 6 5 x 31/42

Trusted 25/42 18/42 18/42 23/42 28/42 19/42 23/42 154/294 (52.4%)

150

Systems Thinking
SAS
Dialogue

Social Dynamics, also known as Symphony, helps assess the ways in which key stakeholders, key problems and significant actions

influence each other in a particular situation.

Social Dynamics begins by defining a situation and listing the key

stakeholders, problems and actions involved. It focuses on the extent to

which one element in the situation interacts with others. When rating, ask ‘To

what extent does this (name the row element) affect or influence that (name the

column element)?’ The resulting matrix produces an index for influences other

elements (vertical axis) and an index for depends on other elements

(horizontal axis). See System Dynamics for detailed instructions.

Summary of this example: The federal government is actively supporting current
plans to expropriate half of the communal lowlands. This has led to acts of violence,
which may affect the federal government’s public image and power to expropriate
the land (see continuous arrow in the diagram). Violence is the result of a threat to
the communal land tenure system, yet this response (together with more information
on communal land entitlements) may force the federal government to pressure the
municipal authorities to endorse the small farmers’ proposal to redistribute the
remaining communal uplands as individual plots.

Social Dynamics

Factors Lowland
expropriation

Upland
redistribution

Lack of
information

Violence Small
farmers

Municipal
authorities

Federal
government

Total
contribution

Lowland expropriation x 8 7 8 7 9 10 49/60

Upland redistribution 2 x 4 2 9 7 1 25/60

Lack of information 6 5 x 5 8 8 2 34/60

Violence 7 2 2 x 8 8 2 29/60

Small farmers 2 8 1 4 x 4 2 21/60

Municipal authorities 2 7 3 4 7 x 4 27/60

Federal government 8 7 8 8 8 8 x 47/60

Total dependence 27/60 37/60 25/60 31/60 47/60 44/60 21/60 232/420 (55%)

Systems Thinking

Influences little

Depends
a lot
Depends
little

Influences a lot

Influences a lot
Depends little

Influences little
Depends little

Influences a lot
Depends a lot

Influences little
Depends a lot

60
600 30

Violence

Lowland expropriation

Small farmers

Lack of information

Municipal
authorities

Federal government

Upland redistribution

0

151

SAS
Dialogue

Value Dynamics helps assess how the values, moral principles or
rules of ethical conduct that people adopt when taking a position or
acting on a key problem interact with each other. The tool may be
used to describe and reflect on the integration of supporting values

(acting as means) and values expressing end goals.

Value Dynamics begins by defining a key problem or a set of actions

and listing the values people apply and refer to in relation to the

problem or actions. It focuses on the extent to which one value supports
and is supported by other values people apply to the problem or action.

When rating, ask ‘To what extent does the application of this value (name

the row value) support the application of that value (name the column
value)?’ The resulting matrix produces an index for supports other

values (vertical axis) and an index for is supported by other values

(horizontal axis).

The value system resulting from the analysis can be interpreted considering three

possible scenarios: integration, hierarchy and fragmentation. In an integrated value

system, values support each other, acting as rules of ethical conduct and end goals at
the same time (top-right section). In a hierarchical value system, top left rules of

ethical conduct support bottom right end goals. In a fragmented value system, moral

principles and rules of ethical conduct interact little and are applied to the key
problem independently of each other. See System Dynamics for detailed instructions.

Summary of this example: In this project
achieving peace and a better understanding of
reality are supporting values and end goals at
the same time. They represent core values that
support and are supported by other values in
the system. Dialogue and fairness play the role
of supporting values (or means), while
development is mostly at the receiving end of
other values (an end goal).

Values Fairness Dialogue Development Understanding Peace Total

contribution

Fairness x 4 7 2 10 23/40

Dialogue 8 x 8 9 10 35/40

Development 0 2 x 2 1 5/40

Understanding 7 2 6 x 6 21/40

Peace 3 7 6 8 x 24/40

Total dependence 18/40 15/40 27/40 21/40 27/40 108/200 (54%)

Supports little

Highly

supported

Little

supported

Supports highly

Supports highly
Little supported

Supports little
Little supported

Supports highly
Highly supported

Supports little
Highly supported

40
0

400 20

Peace
Fairness

Understanding

Dialogue

Development

Value Dynamics

152

Systems Thinking
SAS
Dialogue

• SAS2_Module6.1_Sept11_En
• SAS2_Module6.2_Sept11_En