Front Cover
 Front Matter
 Table of Contents
 Ranking of browse species by cattlekeepers...
 Direct matrix ranking in Papua...
 Sustainability analysis
 Oral histories and local calen...
 Portraits and stories
 Bibliographic notes

Title: RRA notes
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00089570/00003
 Material Information
Title: RRA notes
Series Title: RRA notes.
Alternate Title: Rapid rural appraisal notes
Proceedings of RRA Review Workshop, Sussex
Proceedings of the Local Level Adaptive Planning Workshop, London
Participatory methods for learning and analysis
Physical Description: v. : ill. ; 30 cm.
Language: English
Creator: International Institute for Environment and Development -- Sustainable Agriculture Programme
Publisher: IIED, Sustainable Agriculture Programme
Place of Publication: London
Publication Date: December 1988
Frequency: irregular
completely irregular
Subject: Sustainable agriculture -- Periodicals -- Developing countries   ( lcsh )
Sustainable agriculture -- Methodology -- Periodicals   ( lcsh )
Genre: periodical   ( marcgt )
Dates or Sequential Designation: No. 1-
General Note: Title from cover.
General Note: Some issues have individual titles.
General Note: Latest issue consulted: No. 19, published in 1994.
 Record Information
Bibliographic ID: UF00089570
Volume ID: VID00003
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 24385692
lccn - sn 92015492
 Related Items
Succeeded by: PLA notes

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Front Matter
        Page 1
    Table of Contents
        Page 2
        Page 3
    Ranking of browse species by cattlekeepers in Nigeria
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
    Direct matrix ranking in Papua New Guinea
        Page 11
        Page 12
        Page 13
        Page 14
        Page 14a
    Sustainability analysis
        Page 15
        Page 16
        Page 17
        Page 18
    Oral histories and local calendars
        Page 19
        Page 20
    Portraits and stories
        Page 21
        Page 22
    Bibliographic notes
        Page 23
        Page 24
Full Text

/ C 7(13

RRA Notes

Number 3




This is the third of a series of informal notes on Rapid Rural
Appraisal (RRA). The aim is to share experiences and methods
among practitioners of RRA throughout the world.

We plan to publish brief informal pieces on any topic related to
RRA. We would like to hear news of meetings, workshops and
projects, both past and planned. In particular we are seeking
short accounts of experiences with RRA techniques in the field -
failures as well as successes. Please also send titles of
articles, papers and reports for listing under the new
publications section.

We will publish fairly regularly, depending on the availability
of material. As far as possible each issue will be put together
by a different editor and we would like to hear from volunteers
for this task.

The notes are being produced under the Sustainable Agriculture
Programme of IIED, which is financed by USAID and SIDA.

Gordon Conway
Robert Chambers
Jennifer McCracken
Jules Pretty

Material for inclusion in the notes should be sent to:

Jennifer McCracken
International Institute for Environment and Development
3 Endsleigh Street

Telephone: 01-388-2117
Telex: 261681 EASCAN G
Telefax: 01-388-2826




Ranking of Browse Species by Cattlekeepers
in Nigeria

Direct Matrix Ranking in Papua New Guinea

Sustainability Analysis

Oral Histories and Local Calendars
Portraits and Stories

Bibliographic Notes

Wolfgang Bayer

Robin Mearns

Iain Craig
Robin Mearns
Jules Pretty


This third issue of RRA Notes is concerned with methodologies
conducted in a number of different countries and at several
different stages of the development process. The first piece
illustrates very well the value of ranking techniques in
understanding the complexity and importance of browse as a forage
resource. Indeed the study threw up some surprising and
unexpected results; it also appears that the latter stages of the
ranking and analysis did not add significantly to the findings
generated very early on. It is also interesting that the
chemical analyses, taking about one year, were apparently
unnecessary the pastoralists knew best. The second piece is a
reanalysis of a repertory grid analysis used by the author in
Papua New Guinea there is of course much to be gained from this
kind of disciplinary cross-over and careful evaluation of
methodology. In the third piece the use of Sustainability
Analysis in Thailand is described within the context of
institutionalising RRA tools and techniques, and it appears to be
successful at creating consensus within a multidisciplinary and
multisectoral workshop it is not threatening, it is inexpensive
and effective. The final section concerns the representation of
rural people's knowledge in stories, histories and portraits.
These are colourful descriptions of situations encountered by an
RRA team or individual whilst in the field.

These Notes are now sent to some 200 practitioners worldwide: in
the next issue we plan to publish this network list to encourage
the distribution of interesting reports or material between
practitioners. Please, though, do continue to send in
contributions particularly if they are short. The success of
these Notes depends largely upon their informality.

Jules N Pretty
Associate Director
Sustainable Agriculture Programme



The Subhumid Zone Programme (SZP) of the International Livestock
Centre for Africa (ILCA) has been involved in livestock systems
research in West Africa since 1979. The SZP has been carrying
out long-term observations of livestock management by
agropastoralists and arable farmers in selected case study areas
in central Nigeria, parallel to testing and refining innovations
in animal husbandry and nutrition in collaboration with the
livestock keepers. Rapid appraisal techniques have been
periodically applied to gain a preliminary understanding of
certain elements of the livestock production systems. An example
is the ranking of browse species used within the agropastoral

A study of the grazing behaviour of cattle kept by settled Fulani
agropastoralists in two case study areas a farming region and a
grazing reserve had revealed that browse was an important
forage resource, particularly in the late dry season (Bayer,
1986). Whereas other forage resources such as natural range or
crop residues had already been studied in detail, little was
known about what constituted "browse": i.e. which trees and
shrubs were most commonly eaten by the cattle. Closer
observation of a grazing herd revealed that numerous different
species were being browsed. In order to identify whether and how
this forage resource could be improved, it was necessary that the
pattern of browse usage be identified and the relative importance
of the various browse species be assessed. To this end, the
following quick study was carried out.

Interview with Key Informants

A list of browse species in the study area was compiled on the
basis of observations of grazing cattle and interviews with key
informants. The SZP enumerators and herdsmen were men from the
main ethnic groups in the area (Kaje, Kamantan, Fulani, Hausa)
and had a good knowledge of the local vegetation. With their
assistance, a preliminary list of browse species was compiled
according to the species' local names. This list was checked with
some Fulani agropastoralists who were collaborating with the SZP
in on-farm ("in-herd") trials, and additional names of the browse
species were added. The resulting list was arranged
alphabetically and each browse species was given a number (code)
in order to facilitate recording of the ranking results. My
field assistant and I then prepared small cards with one species
name (and the corresponding number) per card.

Ranking Interviews

We took these cards to 6 Fulani cattlekeepers in each case study
area and, in individual interviews, asked them to rank the browse

species according to their importance as feed for their cattle.
As cattle herding is a male task among the Fulani, only men were
interviewed. The procedure was as follows:

1. The plant name on each card was read out to the Fulani man
(none was literate) and he was asked whether he knew the
plant and whether cattle eat it. If he did not know the
plant, the card was set aside. If he knew the plant, the
card was laid down on the ground in front of him.

2. In the case of each subsequent card, the man was asked
whether the plant named on the card was known to him and, if
so, if it was more or less important as cattle feed than the
plants named thus far. This was done by pointing to the
cards which had already been placed on the ground and
asking: it is more (less) important than this one (reading
out the name of the plant) ... than this one (ditto)... and
so on.

3. After all the cards had been laid down in order of
importance, the plant names in order of ranking from most to
least important was read back to the Fulani. He could then
change the order of the cards if he wanted to correct the

4. The man was then asked whether he knew of any other tree or
shrub species in the area which was eaten by cattle but had
not been mentioned thus far on the cards. If he named
additional plants, the name of each was written on a card
and he was asked to place the card as in Step 2.

5. We recorded the ranking by each man by writing down the card
numbers (rather than the plant names) in the order of
ranking. For example, if species No 5 was ranked as most
important, species No 12 as second most important, species
No 7 as third most important etc., we recorded the sequence
5, 12, 7 etc. Additional information given by the Fulani
about the browse species during the course of the ranking
exercise, e.g. about the plant parts preferred by the
cattle, was also noted.

6. As new plant names could be added during each interview, the
first Fulani interviewed did not have the chance to rank all
the species mentioned by the time we had interviewed the 6th
man in each case study area. In order to give each man the
opportunity to rank all the species, a second round of
ranking interviews was done with the same people. In this
second round, no additional species were mentioned by the

7. Tables of species ranking (one for each case study area)
were compiled by calculating an average from the rankings
given by the pastoralists (see Table 1). A pocket
calculator was used for this purpose.

Table 1: Simplified table of browse species ranking according to
importance as cattle feed*

Species Local Ranking by Fulani in farming area
name ------------------------------- Rank
Moh. Saleh Adamu Huss. average

Aftelia africana Kaawoo I 1 2 1 1.25 1
(Mahogony bean)
Khaya senegalensis Hadaaci 2 3 1 2 2 2
(Savannah mahogony)
Adenodolichos paniculatus Depaji 4 2 3 3 3 3
(Fire bean)
Oxytenenthera abyssinica Gooraa 3 5 4 4 4 4
Hucuna poggei Karara 5 4 5 6 5 5
(Cow itch)
Daniellia oliveri Maje 6 7 6 5 6 6
(African copaiba balsam)
Pterocarpus erinaceus Banwi 8 6 8 7 7.25 7
(African rosewood, African teak)
Cussonia barteri Tuwon 7 8 10 9 8.50 8
(Barter's cussonia) giwa
Vitex doniana Dinya 9 9 7 10 8.75 9
(Black plus)
Parinari curatellifolia Nawari 10 10 9 8 9.25 10
(Rough-skinned plum)

English names according to Dalziel (1937) or Gledhill (1972).

By the end of the ranking interviews, we had a list of about 30
species browsed by cattle. Subsequent evaluation of the ranking
results revealed that the 15 most important browse species in
both case study areas had already been included in the list after
the ranking had been done by the first 3 Fulani interviewed. The
"top 10" species are presented in Table 1. For us, the
importance of bamboo as a browse plant is somewhat unexpected.

In the late dry season i.e. during the period when browse is most
intensively used, samples were taken of each browse plant for the
purpose of estimating its quality as feed.

Identifying the Species

The ranking of browse species was done using local names of the
trees and shrubs on the cards. The lingua franca in Central
Nigeria is Hausa, but this is the second or third language of

most of the inhabitants. The Fulani agropastoralists' first
language is Fulfulde; the arable farmers in the case study areas,
who belong to the Kaje, Kamantan and Ikulu ethnic groups, each
speak their own language and have their own names for the local
plants. Some Fulani knew certain plants only by the Fulfulde
name, some only by the Hausa name, and a few plants were known
only in the language of the local farmers. Wherever possible, we
noted each plant name in Fulfulde, Hausa, and one of the other

To be able to interpret our findings in "scientific" terms, we
then sought the Latin names for each plant. Valuable aids in
this connection were Dalziel's Useful Plants of West Africa
(1937) and a list of vernacular (Hausa) names of trees and shrubs
which had been prepared by the Department of Forestry (Gbile,
1980). A final check of our "identification/interpretation" work
was made by a taxonomist from the National Animal Production
Research Institute (NAPRI) in Zaria 250 km north of the case
study areas who came to the field for this purpose.

Survey of Frequency of Browse Occurrence

With the aid of the taxonomist, we then made a quick survey of
the frequency of occurrence of the different woody species in the
case study areas. In each of the main vegetation/landuse types
(upland range, fallow land, cultivated fields, riverine areas,
shrubland) random quadrats were staked out and each tree/shrub
was counted, identified and classified according to height: knee-
high, waist-high, nose-high, and higher than the arm stretched
above the head. The measuring rule (the author) was 196 cm tall,
with an armstretch to a height of 2.4m.

The size of quadrat varied according to plant density; the
smallest quadrats were 10 x 10 m for recording shrub vegetation,
and the largest were 100 x 200 m to record trees in cultivated
fields. The number of quadrats per vegetation/landuse type
ranged from 2-6 in each case study area. In locations with few
tall trees and a dense understorey, we sampled a larger area for
the tall trees (e.g. 100 x 200 m) and then within this area we
sampled 3-4 quadrats of 10 m x 10 m to record the species up to
2.4 m in height.

During this quick survey, we found more than 100 woody species,
including all but two of the browse species mentioned by the
Fulani during the interviews. One species not found was Acacia
albida, which is very well known as a browse species by
scientists as well as the Fulani but is quite rare in the
subhumid zone. The other was a Veronica spp which is planted in
household gardens and used for seasoning soups but which is
rarely found in the fields. We also found some plants which had
not been mentioned by the Fulani as important browse species but
which are described in the literature as browse plants in other
parts of Africa (e.g. in Le Houerou, 1980).

When we expressed the occurrence of browse plants in percent of
total number of woody plants, we found that the percentage of
browse plants was higher in fallow and cultivated fields than in
natural savanna (upland range). In the case of small plants
(less than 180 cm or ca 6 feet in height) even the absolute
number of browse plants was higher in fallow fields than in
natural savanna. This would mean that the traditional way of
clearing fields for cultivation does not reduce the availability
of browse to the extent expected. The clearing practices and
shrub/tree regeneration in indigenous agricultural systems are
worthy of more investigation. Large agricultural development
schemes in Nigeria often involve wholesale clearing by
caterpillar. Much could be learned from local farmers about tree
protection (e.g. species and their uses, techniques of fostering
regeneration) to promote sustainable rather than destructive
forms of agricultural development.

Time Required for the Study

Two days were spent compiling the initial list of browse species
and preparing the cards. The interviews (30-60 minutes each)
were carried out during normal field visits for monitoring on-
farm trials and therefore stretched over a period of about 6
weeks. If we had concentrated solely on the browse ranking
interviews, they probably could have been completed within 3 days
in each case study area, i.e. in a total of 6 days. Two half-
days were spent in the field with the taxonomist in order to
verify species identification. Calculating the average species
rankings and compiling the ranking tables was a matter of 2 hours
in total.

The survey of how frequently the woody species occurred required
7 days of fieldwork, followed by 2 weeks verifying species names,
entering the data into a personal computer, and calculating
species occurrence per unit area in each vegetation/landuse type
in each case study area.

The samples of the browse species ranked by the Fulani were sent
to ILCA Headquarters in Ethiopia for chemical analysis. After
drying, the 30 samples had been milled in one afternoon, but it
took almost a year before the results of the analysis were sent
back to us in Nigeria. This aspect of the study could not,
therefore, be classified as "rapid".

Discussion of the Methods

In general, the "ranking interviews" gave us a fairly good idea
about the complexity and importance of browse as a forage
resource. Pastoralists were very willing to share their
knowledge about browse plants with us and appeared to enjoy the
interviews as much as we did.

What we failed to record systematically during the interviews
were the other uses of the trees and shrubs in addition to
fodder. Here, we managed to collect only incidental information.

Some of the browse species such as the savannah mahogany tree
(Khaya senegalensis) provide valuable hardwood. Others such the
Ficus spp are used for medicinal purposes. Still other trees
such as Vitex doniana are preferred for beehives; they also
produce edible fruits, and the leaves are used as vegetables.
These multiple uses of trees and shrubs could have been more
systematically recorded during the ranking interviews.

Looking back on how we conducted this rapid appraisal of browse
use and importance, the questions arise as to whether all parts
of the study were necessary and whether we might have conducted
some parts of the study more efficiently. The key aspects were
the identification of the species browsed and the cattlekeepers'
opinions about the relative importance of these species for
cattle nutrition. We chose to do the species ranking from 1 to
30, but it might have been easier and quicker for both the Fulani
and us to have grouped the cards into, say, 3 categories, e.g.
very important, important, less important. In fact, when we
divided the 30 species in the final ranking lists into 3 groups
of 10 species each, the "top 10" turned out to be the plants best
known to all pastoralists and were also plants with relatively
high nutrient value, the second-best group of 10 were also well
known to pastoralists but lower in nutrient value, and the third
group included species not known to all pastoralists and of
rather mixed nutrient value. Ranking in 3 categories would
probably have yielded similar results.

The survey of frequency of occurrence was done to gain more
information about "browse on offer". However, even a cattle herd
kept by settled pastoralists can use forage resources within a
radius of at least 5 km around the homestead, i.e. within an area
of almost 80 km To record the woody species with any degree of
precision within such a vast area would be quite a demanding task
in terms of time and personnel. The rapid survey within a small
number of quadrats in the main vegetation/landuse types gave us a
rough idea of the diversity of woody species in a subhumid
savanna environment, yielded some limited quantative data about
tree and shrub density in different vegetation formation and
under different forms of landuse, and clearly revealed which
species of trees are left in the fields when land is cleared for
cultivation in the traditional farming systems. The rapid survey
did not, however, yield figures which could be used to estimate
the total amount of browse available. The need for a specialist
in taxonomy capable of identifying the great majority of the
species in the field and the relatively time-consuming tasks of
verifying taxonomic names and data processing may limit the
applicability of such surveys.

It is doubtful whether chemical analyses of the browse species
was necessary. The results finally produced by the laboratory in
ILCA Headquarters agreed well with the values already published
in the literature (e.g. le Houerou, 1980).

Furthermore, there are considerable doubts as to the validity of
standard chemical analyses for estimating the feeding value of

browse plants, since substances such as tannins in the plant
parts may render certain elements (particularly nitrogen)
indigestible in the animal's stomach. As long as these problems
of analysing browse species have not been solved, it may be
sufficient to use published results of chemical analyses in order
to estimate the feed quality of the browse species ranked by the
livestock-keepers. Furthermore, the all-too-common delays in
processing the plant samples in laboratories can lead to great
delays in completion of reports. If results are to be produced
rapidly for immediate use, it is probably advisable in most cases
of rapid appraisal to avoid dependency on laboratories.

Literature Cited

Bayer, W. 1986. Agropastoral herding practices and grazing
behaviour of cattle in the subhumid zone of Nigeria. Addis
Ababa: ILCA Bulletin 24: 8-13

Dalziel, J.M. 1937. The Useful Plants of West Tropical Africa.
London: Crown Agents for Overseas Governments and Administrations

Gbile, Z.O. 1980. Vernacular names of Nigerian Plants. Lagos:
Federal Department of Forestry (Mimeo)

Gledhill, D. 1972. West African Trees. London: Longman

Le Houerou, H.N. (ed). 1980. Browse in Africa: The Current
State of Knowledge. Addis Ababa: International Livestock Centre
for Africa

Wolfgang Bayer
Rohnsweg 56
D-3400 Gottingen
Federal Republic of Germany


Tools like direct matrix ranking (DMR) become honed down,
improved and simplified over time, which is a very good thing.
Reading Robert Chambers' piece on DMR in Kenya and West Bengal
(RRA Notes No.l), I was struck by how much simpler the technique
sounded than the one I used in the Southern Highlands of Papua
New Guinea. However, the participatory learning process of
eliciting the matrices in each case was very similar.

My objectives were:

1. To compare a formal, 'scientific' land evaluation with local
people's own evaluation of the same land resources.

2. To show how social relations between clans in the area were
critical in controlling access to valued land types.

3. To highlight the differences between clans in access to
valued land resources, and to show how this was reflected in
disputes over particular land types.

I used DMR for the first of these objectives, to get to know
people and let them get to know me, and to break the ice for our
later and often sensitive discussions around the other issues.

Repertory grid analysis

At the time I called the technique I was using 'repertory grid
analysis', which should be seen as part of the same family of
techniques as DMR. The origins of repertory grid analysis are to
be found in the psychological literature on personal construct
theory, which as Robert Chambers pointed out, gets complicated
and difficult. But strip away the labels like 'mental constructs'
and 'semantic differentials', and what you are left with is
basically DMR, or so I had thought.

Other applications of repertory grid analysis in the context of
rural livelihoods in the South include:

- investigation of the utilities by which farmers evaluate
common weeds and local rice varieties in West Africa (Paul

- farmers' choice of crops in the Gezira, Sudan (John Briggs)

- small farmer perceptions of farming conditions and methods
in Trinidad (Barry Floyd)

- the perceived worlds of colonists of the Colombian
rainforest (Janet Townsend)

The application in Papua New Guinea

With people from seven different clans or sub-clans, in group
discussions, I drew up two sets of matrices, with the objects
(land types) along the top and the criteria down the side. One
set took local names for (and therefore definitions of) land
types as the objects, while the other took the land classes (Land
Mapping Units) defined in the formal land evaluation as the
objects. Both sets of matrices used exactly the same criteria for
distinguishing land types, which were the criteria people used
when I asked them to make choices between land types. We did this
for three land types at a time: "How is this one different from
the other two?", although pairwise comparisons would have done
just as well. Tables 1 and 2 are examples of the two matrices
elicited for one clan.

The discussions all took place in the field on hilltops which
overlooked the areas of land we were discussing. The groups were
either all men, all women, or mixed, and the number of
participants ranged from three to around 10.


There were at least two major problems in the way I applied the
ranking technique:

1. With DMR the objects to be ranked (tree species, crop
varieties etc) are often straightforward to define. Land
types are not like that. I tried to match outsiders'
definitions of land types (largely based on Western soil
science) to local people's own names and bounding of the
areas. Naturally there is no precise match, so any
conclusions based on this comparison could only be

For example, I could not be sure that when we talked about
eg. poi or kul (valley bottom wetland, in the two local
languages) we were actually referring to the same kind of
land that the land evaluation classified as "PB2" or "WK1".

In fact, this is precisely what I was trying to find out:

"How well do the outsiders' definitions of land types match
up to local people's own definitions?"

"How useful is this expensive land evaluation in terms of
the ways local people regard their own land resources?"

"If a land evaluation is supposed to be a tool for decision
making about land use, would decisions based on this
outsiders' evaluation really be meaningful for local

2. The values I entered in the matrices were scores, from 1 to
5, where 1 represented 'best' and 5 'worst'. The numbers

were written on cards which people would point to for each
object, and according to each criterion, in turn. In other
words, we did not the land types at all, in
relation to each other, but simply gave each a score in
relation to an abstract standard. In this way, many of the
land types would be given the same score on particular
criteria, in which case they cannot be ranked.

Looking back on it, I could kick myself for doing it this
wayl Although it is possible in many instances to rank the
land types from the matrices, at the time I did not see the
exercise as being primarily a ranking exercise. How
important 'labelling' can be: had I thought I was doing DMR
rather than repertory grid analysis, the process of
'ranking' would have been uppermost in my mind. Instead, I
got people to explicitly rank or make choices between land
types only informally, as a means of identifying the
criteria they used. This information, sadly, went

Some lessons learnt

1. Eliciting the matrices proved to be an excellent ice-breaker
and means of structuring discussions.

2. The design of the study was complex, which made interpreting
the results that much more difficult. However, it was
possible to identify which types of land people valued most,
and these were by no means always those land types which the
formal land evaluation classified as most productive for
food crops.

3. Perhaps more important or more interesting than the
intended outcomes of the analysis were the unforeseen
observations. Most notable among these were the differences
in the kinds of responses that different groups made:

a) where a village big-man was present in the group, he
would almost invariably dominate the discussion. Other
people would keep quiet or simply agree with his opinion.
This was true only with all-male or mixed groups.

b) in all-women groups, by contrast, it was more common for
there to be a lively exchange of opinions from all
participants, and for the final score to be the apparent
consensus view of the group.

c) generally men were more confident in their responses,
giving extreme scores of l's and 5's, while women would
give scores in the middle ranges.

d) mixed groups of both men and women were unsuccessful as
mixed groups, since in such cases the men would not allow
the women to speak.

4. Repertory grids were designed to be analysed using
sophisticated multivariate statistical techniques, based on
multi-dimensional scaling, cluster analysis etc., and most
applications of the approach have tended to use such tools
of analysis. But this would be highly inappropriate for the
kind of information generated by this exercise, which is
simply not up to this form of treatment. The use of numbers
in eliciting the matrices allocating scores to each land
type on the basis of a range of criteria means there is
too great a temptation to apply sophisticated statistical
analysis. It is like trying to drive in a pin with a
sledgehammer; more often than not the pin will break.

On reflection I should have used combinations of the symbols
+, -, ++, --, 0, etc. to imply the same scoring. The
matrices could be re-cast in this way, except that it would
be unfaithful to the original information. You cannot be
sure that if you use a range of --,-,0,+,++, instead of
1,2,3,4,5, people would give where they would otherwise
have said 2, or ++ for a 5. Better still, I should have
tried direct ranking instead of abstract scoring!

Robin Mearns
Institute of Development Studies, Sussex and International
Institute for Environment and Development





1 4
5 1
5 5

2 4
3 3
2 5
1 5

1 1 4
4 1 1
1 1 5

1 Karuka nuts, from customarily owned trees in the natural forest
I K~aruka nuts, from customarily owned trees in the natural forest





....5 (VERY POOR)

Gardens on alluvial or drained swampland
Gardens from grassland (mainly sweet potato)
Gardens from steeply sloping grassland (sweet potato and mixed
Gardens from forest or secondary regrowth ("greens" and mixed
crops with sweet potato
Lower montane forest
Wetlands (alluvial or swamp, undrained)






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

1 Karuka nuts, from customarily owned trees in the natural forest



....5 (VERY POOR)

The Land Mapping Units could be identified in the field from the land evaluation map

FKI Ki Floodplain, active floodplain area of recent alluvium and collo-alluvial
deposits around confluence of Mendi, Ki and Kwi rivers, wetland grasses and
sedges, Ishaemum dominant in drained areas, high intensity cultivation of sweet
potato and mixed crops

SN1 Nene Spurs, strongly dissected spurs at periphery of Birop Plateau, ridge crests
and plateau remnants, brown ash and olive ash soils, Miscanthus grassland, low
to medium intensity cultivation of sweet potato and mixed crops

PB1 Birop Plateau, high altitude volcanic ash plain, olive ash soils with brown ash
soils on steeper slopes, Hiscanthus grassland with patches of remnant forest,
low intensity sweet potato gardens with mixed crops on steep slopes

PB7 Birop Plateau, steep river gorges, associated with volcanic ash plain (see PB1)

FS Fault Scarps associated with volcanic ash plains, olive ash and brown ash
soils, lower montane forest and derived secondary regrowth, Miscanthus
grassland, 'bush' gardens of varying intensity

SB Wambul Footslopes, lower dip slopes and colluvial aprons associated with Tambul
Mountains, olive and brown ash soils, sedimentary soils, mixed ash soils,
Miscanthus grassland, medium intensity sweet potato and mixed gardens

--- -- -- -- --------"" -" ~------------ - - - - - - - - - - -



There is a growing interest throughout the world in simple,
systematic and innovative techniques designed to acquire, analyse
and effectively utilise information in rural development
programs. These techniques or 'tools' differ markedly from
conventional approaches, which are largely characterized by a
lack of flexibility, a concentration upon the collection of
quantitative data and their inability to respond to the real
needs of rural people.

Most of these new, innovative techniques have been developed by
practically-oriented academic groups, following Farming Systems
Research, Agroecosystem Analysis or Rapid Rural Appraisal
approaches. However, their adoption by national development
programs has been slow, mainly because potential users tend to
view them as mutually exclusive approaches, rather than as a
collection of component tools that can be assembled for a
particular purpose. Many of the tools also require
simplification if they are to become widely used, the challenge
is to accomplish this while still maintaining adequate scientific

The NERAD Project, in association with Thailand's regional
Universities, has been addressing these issues by reviewing,
analysing and simplifying the individual techniques, and
documenting them as 'tools' by means of user-oriented handbooks.
The handbooks have the common objective of assisting staff and
officials of rural development programs to choose suitable tools
according to their specific needs and available resources. By
this means, it is hoped that workers with little or no previous
experience, will be able to assemble the most appropriate 'user-
tailored package' and utilise the tools rapidly and effectively.

This report describes NERAD's attempt to do this for
'Sustainability Analysis', one of the first tools to be
documented as a handbook in both Thai and English. The article
is intended merely as a general summary of the major lessons
learned, and more details and specific guidelines for using the
technique can be found in the handbook, copies of which are
available from the NERAD Project Director.


SDevelopment needs Program planning
Planner -
Recommendation domains Site selection

SMajor problems Research design
Research needs Research planning

4 Siting of programs
Recommendation domains Siing of program
L Farmer selection
Farmer training
Recommended practices tra
Extension materials

Practical Application

Sustainability Analysis was one of the 16 tools reviewed,
simplified and documented as a handbook by the recent workshop
held for this purpose in Korat, Thailand (See RRA Notes No 1,
June 1988). The completed handbook has now been used by NERAD,
for training participants in the Project's technical review
workshops, before they use the tool to analyse the results of
trial technologies implemented by the Project.

In the first of these workshops, which covered integrated water-
resource development and utilisation, participants received
training in Sustainability Analysis techniques on the first
morning, and then broke into sub-groups to use the tool. Half a
day was allocated for the analysis of each technology, and
although progress was slow at first, skills improved rapidly as
experience was gained, and a total of ten technologies listed
below were analysed by the end of the workshop.

Group 1 : Groundwater

Shallow wells
Modified well

Group 2 : Fish Production

Fish in paddy
Village fishery
Village school fish ponds

Group 3 : Construction

Diversion weirs
Swamp rehabilitation

Group 4 : Integrated Use

Family water-use
Communal use

The outputs from the analyses are now being used as guidelines
for the production of extension handbooks. These will contain
recommended implementation-practices, recommendation domains and
future research and development requirements for each technology,
and will be published and distributed to the relevant line-
agencies for use in their regular work in the future. As an
example of the type of outputs obtained, a summary for the
technology of raising fish in the rice paddy, is presented in the
following table.


Choice of site:

Farmer training:

Preparation of

Fish release:


- Lack of security

- Poor water control

- Prone to flooding

- Construction delayed

- Farmers unwilling to
dig trenches/ponds

- Lack of knowledge on
fingerling sources

- Netting/construction
too expensive

- Lack of water

- Lack of fish feed

- Effect of fish on
rice and vice versa

- Lack of local markets

- Site close to house

* Modified shallow well

- Mid-elevation paddy

- Closer supervision

- Farmer to farmer

- Media broadcasts
* Local fish spawning
by farmers

- Loans for farmers

* Modified shallow well
- Nursery ponds deeper

* Use farm by-products

* Study of rice/fish
ecology needed
* Change fish species

- Farmer marketing

Denotes a potentially promising but untested solution,
that warrants a high research priority.


Sustainability Analysis has proved useful to NERAD for a number
of reasons. First, it is simple and inexpensive to use and can
be applied to analyse any type of technology. Secondly, it is
not threatening to either research or extension personnel, and
thus promotes a frank and honest analysis. Finally, its
effectiveness can be significantly enhanced by using it in
conjunction with other tools, in particular: On-Farm and Multi-
location Trials, Topical Agroecosystem Zoning, Triage and Farmer

Sustainability Analysis undoubtedly still has many potential uses
that have not yet been tested, and further applications of the
technique should therefore be encouraged. Use of the tool in
Thailand has demonstrated the generally poor level of
understanding of the system properties of productivity,
stability, equitability and sustainability, and promoting a wider
acceptance of these as important measures of system performance,

will enhance the utility of the tool. Perhaps the most pressing
development need for Sustainability Analysis is to find a more
appropriate name for it. This is by no means a pedantic
requirement, as the purpose of the tool is often perceived as
assessing only the key sustainability properties of the system,
and as a consequence, it has been significantly under-utilised.

Key References and Sources of Information

Craig, I.A. 1988. Sustainability Analysis. Agricultural
Development Tools, Handbook No 17. Northeast Regional Office of
Agriculture, Tha Phra, Khon Kaen, Thailand (In English)

Chouangcham, P. 1988. Sustainability Analysis. Agricultural
Development Tools, Handbook No 17. Northeast Regional Office of
Agriculture, Tha Phra, Khon Kaen, Thailand (In Thai)

Craig, I.A. and Pisone, U. 1988. A Summary of the NERAD
Promising Processes, Methodologies and Technologies for Rainfed
Agriculture in Northeast Thailand, NERADICS Report TO Northeast
Regional Office of Agriculture, Tha Phra, Khon Kaen, Thailand

Pretty, J.N., Craig, I.A. and Chouangcham, P. 1988.
Familiarisation Notes on Simple and Innovative Tools for
Agricultural Development Programs. NERAD Project, NEROA, Tha
Phra, Khon Kaen, Thailand

* For further information or copies of these reports please
contact lain Craig

lain Craig
Farming and Cropping Systems Specialist, NERAD Project, Northeast
Regional Office of Agriculture, Tha Phra, Khon Kaen 40260,


Local calendars and nutritional studies

Anthropometric means of assessing nutritional status among
children are often criticised for comparing unlike with unlike.
Jelliffe's standard tables, for example, are scarcely valid if
you are comparing the nutritional status of school aged children
in the Highlands of Papua New Guinea, whose diets are typically
deficient in protein, with that of children of a similar age in
Europe or North America. However, anthropometric techniques,
relying on measurements of a child's height-for-age or weight-
for-age, provide a useful means for comparing the nutritional
status of two sample groups of children within the same overall
population. Measuring height and weight in the field presents few
problems other than logistical ones. Telling someone's age
however, can be a whole new ball game.

In many parts of the Southern Highlands in Papua New Guinea,
there are only very incomplete written records of the birth dates
of children older than six or seven years. For those children at
school, the school register does include a column for age or year
of birth. You only notice the problem when a first-year
schoolchild, aged seven, stands before you with a healthy stubble
on his chin! Naturally, if he is in Grade One at school, it
stands to reason that he is seven years old. For children not at
school, you do not even have the school register guesstimate to
go on.

Constructing a local calendar is a good way of overcoming the
problem of putting ages to people. The oral record of important
local events is often very accurate and widely known. For
example, I was able to match the ages of some people to the date
the local airstrip was built, in 1961, an event which heralded
great change in the area. Their parents' ages you could sometimes
judge by the dates of the 'first contact' patrols to pass through
these remote parts of New Guinea, in this case during the 1930s.
Once a range of reference ages had been set in this way, you
could ask mothers whether their second-born was born before or
after another woman's child known to her, and of known age, and
so on. Laborious work perhaps, but fascinating as well as
fruitful in the end.

Oral history

Oral history is an engrossing subject in itself. In one area
where I worked, Upper Mendi in the Southern Highlands, a number
of old men I spoke to remembered very clearly the first contacts
they had with white outsiders. Kumape of Egari village and Pondo
of Komia village remembered them best. There were three very
distinct events which had made a very deep impression on them.
Two of these were Australian patrols during the 1930s, one

administrative, and an earlier one by gold prospectors. I spoke
to several people in different villages about these, and tried as
far as possible to match their recollections to the reports of
the patrol officers and the script of a radio interview given by
one of the prospectors. Suffice to say there was a mismatch.
Perhaps the latter's story was censored for Australian radio,
though I doubt it.

The "third time", so the old men's stories go, was during the
Second World War when a number of fighter planes flew over Upper
Mendi travelling from Wewak on the north coast towards Port
Moresby on the south coast. The noise of the aircraft was at
first baffling. Everyone looked around on the ground to see where
it was coming from. The eventual sight of many planes stacked in
the sky proved to be unforgettable; the men took great delight in
demonstrating with their hands what it looked like.

Finally, oral history serves at least two additional purposes for
the fieldworker. Not only can it be used for dating and putting
ages to people, as mentioned above for nutritional studies, and
as I also found in trying to put a timescale on local vegetation
change, but it is also an excellent way to break the ice with
people, where you may want to discuss many other things more
directly related to "development research". While I was
researching access to land through clan linkages, land disputes
and vegetation change, getting people to tell me their stories
made a refreshing change when we got bored with these matters!

Robin Mearns
Institute of Development Studies, Sussex and International
Institute for Environment and Development, London


The notion of using stories recounted by people met in the field
as an important source of information described by Robin Mearns
above was also explored in a recent zoning exercise conducted in
North West Pakistan. The Malakand Fruit and Vegetable
Development Project conducted a Rapid Agroecosystem Zoning (RAZ)
of Alpuri Sub-Division, with the principal objective of producing
strategies for the future development of the region. There were
4 major principles to the RAZ: local people helped to define the
zones; the information was gathered by a multi-disciplinary team
working in the field; the process involved several iterative
phases of progressive filtering of information, joint workshop
analysis and focussing on insights; and finally the zonal
boundaries and strategies produced will always be subject to
change in the light of information generated in the future.

Amongst the techniques used were portraits and stories, which are
short, colourful descriptions of situations encountered by an RRA
team in the field or stories recounted by people met there. They
describe information that is difficult to incorporate into
diagrams, help to bring to life the conditions of rural people,
and in particular draw attention to how rural people perceive
problems and opportunities.

1. A farmer from Besham-Karora valley with 2 wives and 19
children. When asked, he himself had to ask a son the
number of girls. Only the boys went to school: there is no
local girls school. He owns 60 kanals (3 ha) of mostly
irrigated land; producing mainly rice, but some maize and
wheat. Until 10 years ago his mother cultivated vegetables
and worked in the field, but stopped due to old age. She
complained that the young generation were not interested in
working in the field and her skills were being wasted.
Neither wife was interested "why should they work in the
fields if they can buy vegetables in the bazaar?" Since the
forest were felled 15-20 years ago, the farmer believes that
there is currently less rainfall and more soil erosion.

2. A farmer and his elderly father by the Indus described the
impact of the building of the Karakhoram Highway, completed
in 1974. The forests were rapidly depleted by local people,
determined to maximise revenue. The hillsides are now
almost empty, save for a few Acacia, and pine at higher
elevations. The forests used to contain leopard, bears and
jackals. But porcupine is still a serious pest, regularly
damaging maize. These are located in their holes by trained
dogs, and then shot on emergence. In the Shang area about
10 were shot last year.

3. We met an orchard grower in Shahpur who felt that orchards
were profitable, but there still remained an obligation to
supply fruit to family, friends and neighbours free of

charge. There was some evidence of pests and diseases, but
as no extension officer has ever visited him, he does not
take any precautionary plant protection measures.

4. We visited one farmer close to the top of Kandao pass, at
about 7000 ft, and asked about his major problems they
were stability of fruit production, pests on maize, and
fuelwood. We asked about all the nearby pine, did he not
use this for wood? He said no, even though all those nearby
belonged to him. He had first to receive written permission
from the forest department before he could cut his own
trees. So he walked long distances up into the hills, well
away from the road, to chop pine. Unusually for the area,
he was also growing poplar on field boundaries for fuelwood.

5. A farmer with 2 acres in Martung, 1 under wheat followed by
maize and the other under rice, had recently planted 6 fruit
trees near his house, consisting of apple, apricot and plum.
They were purchased as seedlings in Mingora and are
flourishing. Over the past three years he has planted 20
trees. Some time back a commercial nursery brought about 500
fruit tree seedlings to Martung but they were too dry and
died. So the farmer went to Mingora to buy his plants. He
first began to appreciate fruits during his visit to Mingora
when he saw them in the shops.

Jules N Pretty
IIED, London


Khon Kaen University. 1988. RRA in Northeast Thailand Case
Studies. KKU-Ford Rural Systems Research Project. Faculty of
Agriculture, Khon Kaen 40002, Thailand

A companion volume to the earlier Proceedings of the 1985
International Conference on RRA. The case studies cover a wide
range of topics.

Kumar, K. 1987. Rapid, Low-Cost Data Collection Methods for
AID. USAID Program Design and Evaluation Methodology No 10

This guide focuses on various interviewing techniques, direct
observation and informal surveys in general. It is directed
principally at project managers.

NERAD. 1988. Preference Ranking. Agricultural Development Tool
Handbook No HI. NERAD Project, Tha Phra, Khon Kaen, Thailand

NERAD. 1988. Sustainability Analysis. Agricultural Development
Tool Handbook No H17. NERAD Project, Tha Phra, Khon Kaen,

NERAD. 1988. Agricultural Triage. Agricultural Development
Tool Handbook No H 5. NERAD Project, Tha Phra, Khon Kaen,

The three handbooks translated into English arising from the
meeting reported in RRA Notes No 1.

MFVDP. 1988. Alpuri: Rapid Agroecosystem Zoning. Report of
Zoning of Alpuri Sub-District, Swat, Northwest Frontier Province,
Pakistan by Malakand Fruit and Vegetable Development Project and
International Institute for Environment and Development, London

A report on the Zoning of Alpuri, conducted by the Malakand Fruit
and Vegetable Development Project at the beginning of the
project. The zoning produced a first iteration of boundaries,
together with hypotheses for development phrased on strategies or
key questions.

Grandin, B. 1988. Wealth Ranking in Smallholder Communities: A
Field Manual. Intermediate Technology Publications, London

A method for enabling pastoralists or villagers to rank
households by welath or other criteria. Developed in East
Africa. It works in segmented societies. A report in RRA Notes
4 from Ruth Grosvenor-Alsop indicates that it also works in
segmented societies.

McCracken, J.A., Pretty, J.N. and Conway, G.R. 1988. An
Introduction to Rapid Rural Appraisal for Agricultural

Development. International Institute for Environment and
Development, London

Lays out rationale and principles of RRA, together with
descriptions of generic types of RRA. Includes Bibliography and
Kew Workers list.

Carson, B. 1988. A Soil Conservation Manual. KEPAS, Bogor,

A first attempt to institutionalise rapid zoning techniques into
an extension service for the purpose of soil conservation.

Molnar, A. 1988. RRA and Participatory Planning methods for
Land-Based Natural Resource Management Projects. Draft. Report
for FAO Forest Department and World Bank (ASTWE)

A recent balanced and well-informed review, partly besed on
interviews with RRA practitioners.

Ampt, P. 1988. RRA in Forbeshire, Australia. University of
Sydney, New South Wales

An RRA in Australia by 10 scientists over 5 days which used
techniques developed in Guatemala and Thailand which identified
the diversity of farming systems and the innovativeness of

Panya, Opart et al. 1988. Charwal in Northeast Thailand. KKU-
Ford Rural Systems Research Project, Khon Kaen University

A classical RRA in the Khon Kaen tradition of a multidisciplinary
team, triangulation and use of several methods. 51 person days
in the field over 2 months.

Scheuermeier, U. 1988. Approach Development. LBL. Ch-8315
Lindau, Switzerland

An important contribution to participatory development developed
by the Tinau Watershed Project in Nepal. Makes very good use of
diagrams and suggests a good formula for the production of
working hypotheses.

McCracken, J.A. 1988. Participatory Rapid Rural Appraisal in
Gujarat: A Trial Model for the Aga Khan Rural Support Programme
in India. International Institute for Environment and
Development, London

Report of recently conducted RRA for village level planning which
has special emphasis on active invovlment of villaers at all
stages of the work.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs