Measuring the effectiveness of environmental dispute settlement efforts
-
Upload
susan-todd -
Category
Documents
-
view
216 -
download
0
Transcript of Measuring the effectiveness of environmental dispute settlement efforts
Dispute resolution and negotiations
Measuring the effectiveness of environmental
dispute settlement efforts
Susan Todd�
Department of Natural Resources, University of Alaska-Fairbanks, 349 O'Neill Building,
Fairbanks, AK 99775, USA
Received 1 February 1999; received in revised form 1 February 2000; accepted 1 February 2000
Abstract
Many authors have addressed the question of how to define effectiveness in
environmental dispute settlement (EDS). However, to date, we have no standard, measurable
definition of what constitutes effectiveness, and the criteria offered thus far are often more
philosophical than practical. This article reviews and integrates the literature on the criteria
for effectiveness. It translates these criteria into 26 indicators of effectiveness and presents a
pragmatic questionnaire that can be used across a broad array of cases to evaluate the
effectiveness of EDS processes. D 2001 Elsevier Science Inc. All rights reserved.
1. Introduction
Many authors have addressed the question of how to define success in
environmental dispute settlement (EDS). Most agree that the definition of success
must be multifaceted and that no single variable, such as reaching an agreement,
is adequate for evaluating the effectiveness of an EDS effort. However, to date,
the discipline has no common definition of success. According to Blackburn
(1995), the practice of EDS is well developed, but a major challenge to theory
development is `̀ the problem of defining `success' in mediation efforts.''
Because there is no common definition, most case studies present more descrip-
tion than evaluation and often lack empirical evidence (O'Leary, 1995). In
addition, those studies that do attempt to evaluate cases use different yardsticks
0195-9255/01/$ ± see front matter D 2001 Elsevier Science Inc. All rights reserved.
PII: S0 1 9 5 - 9 2 5 5 ( 0 0 ) 0 0 0 5 6 - 1
* Tel.: +1-907-474-6930; fax: +1-907-474-6184.
E-mail address: [email protected] (S. Todd).
www.elsevier.com/locate/eiar
Environmental Impact Assessment Review
21 (2001) 97± 110
in measuring success (Blackburn, 1995; O'Leary, 1995; Kleiboer, 1996). This
makes it difficult, if not impossible, to compare across cases.
There is also a conflict between a desire to promote EDS on the one hand and
a desire to conduct an objective evaluation on the other Ð a serious conflict that
has not been readily acknowledged. For example, Scher (1996) interviewed
several practitioners and found most were uncomfortable with evaluations,
although `̀ even the most skeptical'' conceded that there might be value in
evaluation for improving the practice. Most of the practitioners had little or no
faith that EDS would fare well in an assessment that looked for anything more
than subtle changes in the behavior of participants and the level of trust between
the parties.
Further, Scher postulates that both promotion and evaluation are legitimate
reasons for evaluating EDS cases. He states that evaluations can be used both `̀ to
improve and to promote the practice'' of EDS. Further, he states that two of the
most important steps in designing an evaluation are determining `̀ how the results
should be used'' and `̀ who the evaluation is for.'' He quotes Mike Elliott, a
professor at Georgia Institute of Technology, who perhaps inadvertently pointed
out why evaluations should not be conducted for the purpose of promoting the use
of EDS. Elliott claimed he had `̀ never seen a good evaluation'' of an EDS process
and was frustrated because, even though they might improve the practice of EDS,
`̀ negative evaluations might be damaging to the field.'' Another professor, Judith
Innes from Berkeley, said `̀ the biggest problem with evaluation is the risk that it
will underestimate the benefits of ADR'' (alternative dispute resolution). The
implication is that it is better not to evaluate at all than to risk a negative result.
However, there are also risks involved in failing to evaluate the practice
objectively. Credibility is hard-won and easily lost. To have credibility, evalua-
tions must be done objectively, even though some probably will be negative. We
must conduct them scientifically without concern for `̀ how the results will be
used'' or `̀ who the evaluation is for.'' Paying attention to these concerns implies
that it is okay to report negative results to practitioners, but not okay to report
those same results to politicians or agencies who might be in a position to fund an
EDS process. This would not constitute objective evaluation. We must get the
facts out and let the chips fall where they may.
Negative results do not mean that EDS does not work Ð only that it may not
always work. If a study is done well, and if EDS is as powerful as its promoters
claim, they have nothing to fear from evaluation. Unless EDS processes are
already perfect, we should welcome negative results for that is how the process
will be improved.
This conflict between promoting and evaluating EDS spotlights the need for
standards in judging effectiveness. Allowing each evaluator to select their own
favorite criteria after a case is finished is analogous to shooting the arrow first,
then painting a target around the point where it landed. Instead, we need to set the
target before conducting evaluations, and we can only do so by defining a
common set of factors for identifying effective EDS efforts.
S. Todd / Environmental Impact Assessment Review 21 (2001) 97±11098
This set must also be practical and Ð to the extent possible Ð measurable. It
will not help to have a broad criterion such as `̀ the wisdom of the agreement''
unless we also have practical means of assessing it. Such measurements will help
provide the data that some have found lacking in the studies to date. For example,
O'Leary (1995), in her assessment of the EDS literature, questions the claims of
success in many case studies, which she finds `̀ are too often reported with little,
if any, empirical evidence.'' She concludes that we need quantitative data,
questionnaire research, and additional comparative studies that examine not only
cases of dramatic success, but also less-effective cases. Likewise, Kressel and
Pruitt (1989), in their review of mediation research, call for more `̀ systematic
approaches to self-report data.'' To that end, this article provides a questionnaire
that researchers could apply to a wide variety of cases.
A change in terminology is also proposed here. Instead of `̀ criteria for
success'' or the `̀ definition of success,'' as most of the literature has called it,
this article recommends the phrase `̀ indicators of effectiveness.'' This terminol-
ogy is preferred for two reasons. First, unlike the term `̀ criteria,'' which implies
precision, the word `̀ indicator'' suggests that the factors are indicative Ð but not
a hard and fast measure Ð of success. Second, the word `̀ effectiveness'' is
preferred over the term `̀ success.'' The word `̀ success'' generally denotes
something unambiguous and near-perfect Ð an effort is either a successful or
it is not, and there is little room to equivocate. The word `̀ effectiveness,'' on the
other hand, is less black and white. It connotes something that is adequate and
useful, but not necessarily perfect.
This article reviews and integrates the literature into a fairly comprehensive
yet manageable list of indicators that should be considered in evaluating an EDS
case. Next, it discusses how these indicators could be used in the form of a
relatively simple Likert-scale questionnaire and proposes that the population for
the questionnaire should be those who took part in the negotiations. (I refer to
these active participants as `̀ the negotiators'' and/or `̀ the negotiating group.'')
Such a questionnaire should supplement rather than substitute for the qualita-
tive information and open interviews of the stakeholders that remain vital in
setting the context of a negotiation (these interviews usually include, but are not
limited to, the negotiators). A questionnaire would provide some of the quanti-
tative data that is noticeably lacking to date, it would facilitate the comparison of a
wide diversity of cases, and it would help to standardize the definition of what
constitutes an effective EDS process. The article concludes with a discussion of
who would administer and who would complete this questionnaire.
2. The indicators
Several commonalities exist in the literature on the definition of success. First,
many authors note that it is important to evaluate the effectiveness of both the
outcome and the process of an EDS effort. Perhaps the first to note this distinction
S. Todd / Environmental Impact Assessment Review 21 (2001) 97±110 99
was Bingham (1986), who categorized several of the factors listed below as either
`̀ process'' or `̀ outcome'' indicators. Moore (1996) and Wondolleck and Yaffee
(1994) also used factors that were classified as process or outcome criteria.
The list of indicators proposed here is organized by these `̀ outcome'' and
`̀ process'' categories, as well as several subcategories. Four subcategories are
adapted from Susskind and Cruikshank (1987), who identified four character-
istics of an effective EDS effort: fairness, efficiency, wisdom, and stability.
Several additional subcategories are recommended. These categories were helpful
in building a thorough and systematic list of indicators. Table 1 lists all the
indicators. The quetionaire in Table 1 was first developed based on a review of
the literature, and then it was used in evaluating three different EDS processes
[18]. It has since been refined and used in evaluating four additional cases. It
includes the indicators discussed below, listed as statements with a Likert scale
ranging from strongly agree to strongly disagree. The respondents can simply
mark their level of agreement with each statement. These indicators, and how
they can be used are discussed below.
2.1. Process indicators
2.1.1. Fairness
The best way to determine fairness, in Susskind and Cruikshank's (1987)
approach, is to `̀ evaluate the attitudes and perceptions of the parties most
affected.'' They suggest several tests, including: (1) Was the offer to participate
genuine and were all the stakeholders given a chance to be involved? (2) Was the
process perceived as legitimate after it ended, as well as when it began? and (3) In
the eyes of the community, was a good precedent set? Would the participants use
such a process again?
2.1.2. Efficiency
Susskind and Cruikshank (1987) determine efficiency in terms of whether
the process cost more than traditional processes, whether it took more time, and
whether any joint gains were left on the table. The findings of Buckle and
Thomas-Buckle (1986) corroborate the importance of this indicator: `̀ the most
concrete benefit indicated by the participants as contributing to their perception
that mediation had `succeeded' was reduction in time spent, expenses paid, and
delay encountered in mediation, when compared with adjudication or regula-
tory hearings.''
2.1.3. Changes in relationships
Several authors have stressed that signing a written agreement, although often
the goal, is not in itself an adequate measure of success because changes in
previously adversarial relationships are also important.
For example, the participants in a study conducted by Buckle and Thomas-
Buckle (1986) pointed out Ð to the surprise of their mediators Ð that they
S. Todd / Environmental Impact Assessment Review 21 (2001) 97±110100
Tab
le1
Indic
ators
of
effe
ctiv
enes
s:a
ques
tionnai
refo
rth
eneg
oti
atin
gte
am
Cat
ego
rya
Su
bca
teg
ory
aIn
dic
ator
Str
ongly
agre
e
(2)
Agre
e
(1)
Neu
tral
or
undec
ided
(0)
Dis
agre
e
(ÿ
1)
Str
ongly
dis
agre
e
(ÿ
2)
Not
appli
cable
(NA
)
Pro
cess
ind
icat
ors
Fai
rnes
s(1
)T
he
pro
cess
has
bee
nfa
ir.
(2)
All
pote
nti
ally
affe
cted
par
ties
wer
e
invit
edto
par
tici
pat
e.
(3)
Th
em
emb
ers
of
the
team
adeq
uat
ely
rep
rese
nte
dth
era
ng
eof
publi
copin
ion
on
this
issu
e.
(4)
The
pro
cess
of
choosi
ng
team
mem
ber
s
was
fair
.
(5)
Iw
ou
ldb
ew
illi
ng
topar
tici
pat
eon
such
ate
amag
ain
.
(6)
Th
isp
roce
ssse
ta
go
od
pre
ceden
tin
the
eyes
of
the
com
mu
nit
y.
Eff
icie
ncy
(7)
The
pro
cess
has
bee
nef
fici
ent
inte
rms
of
tim
ean
dm
on
ey.
Chan
ges
in
rela
tio
nsh
ips
(8)
Th
eef
fort
has
incr
ease
dth
ele
vel
of
trust
bet
wee
nth
ep
arti
es.
(9)
Th
ep
roce
ssim
pro
ved
my
insi
ght
into
the
inte
rest
san
dp
osi
tion
sof
all
the
dis
puta
nts
.
Com
par
edto
alte
rnat
ive
pro
cess
es
(10)
An
alte
rnat
ive
pro
cess
,su
chas
liti
gat
ion
or
leg
isla
tio
n,
wo
uld
not
hav
ew
ork
edas
wel
l
inse
ttli
ng
this
dis
pute
.
Ou
tcom
e
ind
icat
ors
Pu
rpo
se(1
1)
Th
ete
amm
etit
sst
ated
purp
ose
(to
dis
cuss
the
issu
eso
rth
ed
ata,
or
tore
ach
agre
emen
t,et
c).
Con
sen
sus
on
a
wri
tten
agre
emen
t
(12
)T
he
team
reac
hed
conse
nsu
son
a
wri
tten
agre
emen
t.
S. Todd / Environmental Impact Assessment Review 21 (2001) 97±110 101
Cat
ego
rya
Su
bca
teg
ory
aIn
dic
ator
Str
ongly
agre
e
(2)
Agre
e
(1)
Neu
tral
or
undec
ided
(0)
Dis
agre
e
(ÿ
1)
Str
ongly
dis
agre
e
(ÿ
2)
Not
appli
cable
(NA
)
Inte
rest
s(1
3)
My
ow
nin
tere
sts
wer
em
etby
the
pla
n.
(14
)T
oth
eex
ten
tp
oss
ible
,th
ew
ritt
enag
reem
ent
met
ever
yon
e's
inte
rest
s.
Wis
do
mo
fth
e
wri
tten
agre
emen
t
(15)
The
pro
cess
and
the
wri
tten
agre
emen
tat
tem
pte
d
tom
inim
ize
the
risk
sof
bei
ng
wro
ng.
(16
)T
he
pro
cess
consi
der
edth
epote
nti
alen
vir
on-
men
tal
impac
tsof
the
wri
tten
agre
emen
tan
d
fou
nd
them
acce
pta
ble
.
(17
)T
he
pro
cess
consi
der
edth
epote
nti
also
cial
and
eco
no
mic
imp
acts
and
found
them
acce
pta
ble
.
Sta
bil
ity
of
the
wri
tten
agre
emen
t
(18)
The
team
'sw
ritt
enag
reem
ent
isli
kel
yto
last
.It
wil
l`̀
stan
dth
ete
stof
tim
e.''
(19)
The
level
of
contr
over
syhas
dim
inis
hed
asa
resu
lt
of
this
dis
pute
sett
lem
ent
effo
rt.
(20
)T
her
eis
bro
adp
ub
lic
support
for
the
wri
tten
agre
emen
t.
(21
)T
he
effo
rth
asse
ttle
dth
ekey
issu
es.
(22
)I
bel
iev
eth
ew
ritt
enag
reem
ent
isbei
ng
or
wil
l
be
imple
men
ted.
(23)
Ther
ear
epro
vis
ions
for
reneg
oti
atio
nif
pro
ble
ms
aris
e.
Over
all
effe
ctiv
enes
s(2
4)
As
ate
amm
ember
,I
found
the
effo
rt
per
sonal
lyre
war
din
g.
(25
)I
thin
kth
eef
fort
was
asu
cces
s,by
my
ow
n
def
init
ion
of
the
term
.
(26
)P
leas
eex
pla
inb
elow
why
you
thin
kit
was
Ðor
was
no
tÐ
asu
cces
s.
aT
hes
eca
tegori
esan
dsu
bca
tegori
esw
ere
use
ful
inbuil
din
gth
eli
stof
indic
ators
,but
they
nee
dnot
be
incl
uded
on
the
ques
tionnai
refo
rth
eneg
oti
atin
gte
am.
S. Todd / Environmental Impact Assessment Review 21 (2001) 97±110102
viewed their EDS processes as effective even though no agreement was
reached. They identified several other significant ways in which mediation had
contributed to the successful processing of their conflicts. These included
personal rewards such as an increased personal ability to negotiate, an increase
in their level of trust and understanding, and insight into the interests and
positions of all the disputants. The authors concluded that we should `̀ account
a mediator effective to the degree that he or she enables the parties to increase
their affective and cognitive awareness about their relationship and the matter
at dispute.''
In a study of several EDS cases compiled by Crowfoot and Wondolleck
(1990), citizen members of EDS teams noted that they had `̀ gained valuable new
skills'' and that the processes had `̀ increased the credibility, legitimacy, and trust
between what had before been traditional adversaries.'' Bingham (1986), Lee
(1982), Moore (1996), Wondolleck and Yaffee (1994), and Zubek et al. (1992)
found that similar relationship-oriented criteria were important in participants'
assessments of their EDS processes.
2.1.4. Comparison to alternative processes
Most evaluations are relative. It is hard to judge the success of anything
without also answering the questions: `̀ Effective compared to what? What are the
alternatives?'' Comparing EDS to other alternatives is also needed. Would a
different process, such as litigation or legislation, have been a more effective
forum for settling the dispute? An EDS process may not appear totally effective
by itself Ð but if litigation and/or legislation were tried and failed to settle the
dispute, or if they cost far more in terms of money or time Ð the EDS process
may look quite effective in comparison.
If the stakeholders in a given case did try other forums, researchers should
make an effort to find out what happened and compare that to the EDS results. If
other forums have not yet been tried, then our only option may be to ask the
stakeholders themselves if they think an alternative forum would have worked
better in terms of the same indicators of fairness, efficiency, wisdom, etc.
2.2. Outcome indicators
2.2.1. Purpose
In Susskind and Cruikshank's approach, the fairness and the efficiency
indicators apply to the process, while wisdom and stability apply to the outcome.
However, the wisdom and stability indicators will not apply in every case because
many EDS teams are convened to discuss options or share information and are
not expected to reach an agreement. To deal with these cases, I added the
indicator `̀ Did the group meet its purpose?'' One criterion for the effectiveness of
many endeavors is whether the effort accomplished what it set out to do. If the
purpose was simply to discuss the issues, was that purpose met? If a team set out
to develop recommendations, did they do so?
S. Todd / Environmental Impact Assessment Review 21 (2001) 97±110 103
2.2.2. Consensus
First, a definition is in order. Consensus is often discussed but rarely defined in
the literature, despite the fact that it is often misunderstood by the general public
as an agreement that fully meets the concerns of all the parties to the fullest. I
define consensus as a package agreement Ð often a package of compromises on
the issues and subissues Ð which all members of the negotiating group can live
with because it meets their most important concerns. Unlike the typical vote Ð
which offers essentially two options: `̀ I love it'' or `̀ I hate it'' Ð consensus
allows a third option, if you will, the `̀ I can live with it'' option. This in itself
considerably expands the possibilities for agreement.
Because consensus is so fundamental to an EDS process, it is only natural to
view it as a key criterion of effectiveness. In fact, the pinnacle of effectiveness is
often viewed as a process in which the parties reached consensus on all of the key
issues and were able to see their consensus agreement implemented. However, it is
also quite respectable to reach agreement on many, but not all, of the issues.
2.2.3. Written agreement
In addition to reaching consensus, the literature also emphasizes the impor-
tance of putting the agreement on paper. A written agreement helps to ensure that
all parties are agreeing to the same thing and to eliminate some problems in
recollection and interpretation. In addition, while group members themselves may
move on, a document will remain and can have lasting impact. Therefore, if
reaching an agreement was part of the purpose of an EDS process, then it is
important to include this as an indicator of effectiveness.
If, however, the purpose of the process was simply to discuss information or
develop options, then it would be inappropriate to use a written agreement as a
measure of effectiveness (in which case the next two major criteria Ð the wisdom
and the stability of the agreement Ð would also be inappropriate measures).
2.2.4. Interests
A fundamental concept in EDS and consensus-building generally is that the
parties should attempt to meet their individual interests. The interviewees in
Moore's (1996) study identified this as a key factor in judging effectiveness.
However, as Moore (1996) and Wondolleck and Yaffee (1994) point out, a good
process attempts to meet the interests of all the parties, both individually and as a
group. Thus, the statements `̀ My own interests were met by the plan,'' and `̀ To
the extent possible, the agreement met everyone's interests'' were added to the list
of indicators.
2.2.5. Wisdom of the agreement
The wisdom of an agreement is surely the most difficult and subjective, yet
perhaps the most important, criterion to measure. Certainly, group effectiveness
should be related to the quality of their decision. As an expert in the evaluation of
management teams points out, if a group is established to make a decision, then
S. Todd / Environmental Impact Assessment Review 21 (2001) 97±110104
effectiveness for that group should be gauged in terms of the merits of their
decision (Guzzo, 1986).
As Susskind and Cruikshank (1987) point out, one difficulty with this criterion
is that it is generally only obvious in hindsight. Given time, we can look back and
evaluate the impacts of a decision. If a team decided to harvest timber on one site
and preserve another, we could examine both areas for many years to monitor the
environmental and social consequences of the decision. However, if little time
has passed, it will be difficult to assess the actual impacts.
However, we can determine whether the EDS process made a systematic
attempt to consider several alternative courses of action and their potential
impacts. Often, a good deal of the controversy on an EDS case results from
differences of opinion regarding the potential impacts of an action. Did the group
consider the possible environmental impacts (including cumulative impacts) or
several alternatives? Did they reach some agreement on the range of likely
impacts? Did they attempt to minimize the risk of being wrong?
In addition to the environmental questions, it is also important to consider the
social consequences. People are a resource, too, and most EDS cases are basically
struggles to balance the biophysical needs of an ecosystem with human needs. A
culture and a way of life can also be endangered. Thus, a question should be
included regarding whether the risks of social and economic impacts were
considered and whether there was an attempt to minimize problems.
2.2.6. Stability of the agreement
Stability is also a key component of effectiveness. `̀ An agreement that is
perceived as fair, is reached efficiently, and seems technically wise is nevertheless
unsatisfactory if it does not endure'' (Susskind and Cruikshank, 1987). Ideally,
the agreement will be implemented, and it will last because it diminishes the
controversy and, at best, settles the key issues. An agreement will also be more
stable if provisions are included for renegotiation, should that become necessary.
These points are discussed in more detail below.
2.2.6.1. Has the level of controversy diminished? Is there broad public
support? The level of controversy and the level of public support for a decision
are really opposite sites of the same coin; one looks for opposition, the other for
approval. It should be noted that at best, the underlying broad conflict (e.g.,
timber harvesting vs. preservation) is unlikely to disappear as a result of an EDS
process (Crowfoot and Wondolleck, 1990). However, it is hoped that the
controversy over the specific dispute (harvesting timber in a small, specific area
vs. preserving that area) can diminish Ð if not disappear Ð if the key issues have
been settled. Most importantly, the process should not create more problems than
it solves. As Carpenter and Kennedy (1988) put it, a basic principle of EDS is to
`̀ do no harm'' (i.e., do not exacerbate the controversy).
If the agreement is supported by a wide range of interests and if that support is
fairly solid, it is more likely to be stable and capable of withstanding attacks from
S. Todd / Environmental Impact Assessment Review 21 (2001) 97±110 105
extremes on both sides. Without broad support, an agreement signed by a few
representatives will not have far-reaching impact.
2.2.6.2. Have the key issues been settled? What could be a finer outcome than
that the key issues are settled? As one negotiator said of the key issue in her EDS
process, `̀ It's not an issue anymore. It's no longer front-page news; it's no longer
even back page news. I guess that's one of the best indicators we could hope for
that we did our job.''
If the key issues have not been settled, the controversy is likely to flare again. The
negotiators may have made significant progress in promoting understanding,
establishing relationships, and sharing information, but as long as key issues
remain, the controversy will not diminish and the agreement is unlikely to be stable.
2.2.6.3. Are provisions included for renegotiation? Because it is rarely possible
to anticipate all the types of problems the settlement may face, it is also important
for the disputants to have a good working relationship that will allow them to
return to the bargaining table to work things out if the need arises. `̀ If disputing
parties build a good working relationship, the prospects for stability are greatly
enhanced . . .. The parties are likely to come back to fix the agreement and make
it work if they feel positively about how they were treated'' (Susskind and
Cruikshank, 1987).
2.2.6.4. Has the agreement been implemented? Have the parties complied? Per-
haps the most common measure of effectiveness is whether the agreement has been
implemented and whether the parties complied with the agreement (Lee, 1982;
Zubek et al., 1992; Pearson and Thoennes, 1989). As Carpenter and Kennedy
(1988) point out, if a written agreement has been reached but not implemented, it
can create a feeling of betrayal among the parties and resentment over apparently
wasted effort.
2.2.7. Overall effectiveness
Finally, it is also useful to simply ask the stakeholders whether their process
was effective according to their own personal definitions of the term. This has
been done in many studies, including Buckle and Thomas-Buckle (1986),
Crowfoot and Wondolleck (1990), and Moore (1996).
3. Using the indicators
3.1. Interviews remain essential
Qualitative data will always have a place in the analysis of EDS cases. In fact,
I would argue for much more `̀ rich, thick description,'' as Geertz (1983) calls it,
in the presentation of case studies. Kressel and Pruitt (1989) also contend that
S. Todd / Environmental Impact Assessment Review 21 (2001) 97±110106
most investigations fail to provide enough detail about a case to make it clear just
what conflict management procedures were used.
However, qualitative data alone has some disadvantages that could be over-
come by supplementing interview data with a questionnaire completed by each of
the negotiating group members. In the process of evaluating a case, most
researchers conduct in-depth interviews of the major parties involved in the
negotiation (both those at the table and those on the sidelines), as well as any third-
party interveners. Most interviewers take care to avoid leading questions and to
allow the respondents to direct the discussion to their own points of concern. A
broad question such as `̀ Was this process effective and if so, in what ways?'' is
often used early in the interview to probe for the interviewee's own definition of
effectiveness. Such interviews should retain their place in the assessment of EDS
cases. In fact, the indicators suggested here also work well as an interview guide.
In my own work, I use the indicators as the basis for `̀ open-ended'' questions late
in an interview, after the respondent has shared what they feel is important with
minimal prompting or questioning from me. Combining both open-ended inter-
view questions and more structured, limited-response questions on the same topics
has proven useful for corroborating the data and interpreting the oral responses.
However, interview data alone has several disadvantages. First, it is largely
anecdotal and by itself, anecdotal information is not adequate to accurately assess
the effectiveness of EDS. Second, interview responses are not easily compared.
Because the interviews are intentionally loosely structured, the researcher often
asks the same basic question in slightly different ways in each interview. This
could potentially alter the responses, making it difficult to compare them. In
addition, in answering a question such as `̀ Was this process effective?'' respon-
dents will rarely use a simple `̀ yes'' or `̀ no.'' It is much more likely that each
respondent will use different wording Ð and often many paragraphs of colorful,
transcribed text Ð to answer the question. This detail is essential in providing the
context that is so important in qualitative research. Nevertheless, at the same time,
the detail makes it almost impossible to compare responses within a case Ð much
less across cases Ð or to make a statement such as `̀ eight team members thought
the process was fair and three did not.'' Such discreet data are also needed, and are
sorely lacking, in existing evaluations of EDS.
However, if the open interviews are supplemented with a questionnaire that
allows only a limited number of choices, several of these disadvantages are
overcome. First and most important, the questions are presented in exactly the
same words to each respondent and their answers are easily compared both
within a case and across a wide diversity of cases. The researchers is not required
to summarize Ð and in the process, possibly misinterpret Ð an interviewee's
soliloquy. If the questionnaire takes place following the preliminary analysis of
the interviews, it will also be possible to test the accuracy of the developing
conclusions. Finally, for those who object to the limited-response questionnaire,
it is possible to add a few open-ended questions and room for comment at the
end of the questionnaire.
S. Todd / Environmental Impact Assessment Review 21 (2001) 97±110 107
3.2. Who should administer the questionnaire?
By definition, an objective evaluation must be done by people who do not have
a personal stake in either promoting EDS or in the outcome of the negotiations in
the case being evaluated. Those involved in a case, either as an intervener or as a
stakeholder, can add to the store of knowledge by writing their memoirs of the case,
but their personal interests must be acknowledged up front. Further, such firsthand
accounts must not be our only source of information and evaluation of EDS.
People in the discipline are very firm on how important it is for a potential
mediator to be neutral on the issues. It is equally important for an evaluator to be
neutral. As stated above, there is a place for personal accounts, but objective
analysis by people who are not promoters of EDS, has nothing to gain whether
the evaluation is positive or negative, and were not involved in any way in the
case are essential.
3.3. Who should answer the questionnaire?
Very little has been written about the importance of being careful to `̀ sample''
each of the interests equally when choosing those to interview. It would not be
acceptable, for example, to interview only those who favored a proposal and none
of those who opposed it (or vice versa). All of us have probably seen such
`̀ evaluations,'' and they harm the credibility of the field.
With a written questionnaire, where results will be quantified according to
how many agreed, how many disagreed, etc., it is even more vital to ensure that
the sample is balanced. It is also important to define a bounded `̀ population''
from which you will take your sample. One of the easiest ways to determine who
to survey is to limit the `̀ population'' to the negotiating group itself, which
should be balanced in regard to the range of opinion on the issues (and the
questionnaire asks if they felt the group was balanced). This way, the population
is clearly defined, the research is less likely to be criticized as biased due to
selective sampling, and it is obvious what percentage of this small population
your sample represents (and you may have 100%).
The questionnaire in Table 1 was first developed based on a review of the
literature, and then it was used in evaluating three different EDS processes (Todd,
1995). It has since been refined and used in evaluating four additional cases. It
includes the indicators discussed above, listed as statements with a Likert scale
ranging from strongly agree to strongly disagree. The respondents can simply
mark their level of agreement with each statement.
4. Conclusions
Many authors have addressed the question of how to define success in EDS,
but to date, we have no standard, measurable definition of what constitutes
S. Todd / Environmental Impact Assessment Review 21 (2001) 97±110108
effectiveness. Most of the research has asked the negotiators to assess their
process in their own terms. Since each individual devises a slightly different
yardstick, we cannot compare across cases.
Clearly, it is time to adopt a uniform measure of effectiveness that incorporates
the primary criteria proposed in many different studies. Such a measure need not
be exclusive, but could supplement other approaches. This article reviews and
integrates the literature into a comprehensive yet manageable list of 26 indicators
that should be considered in evaluating an EDS case. These indicators can be
used in the form of a simple Likert-scale questionnaire. This questionnaire would
provide a uniform yardstick and the quantitative data that EDS evaluations
noticeably have been lacking, it would facilitate the comparison of a wide
diversity of cases, and it would help to standardize the definition of what
constitutes an effective EDS process.
References
Bingham G. Resolving environmental disputes: a decade of experience. Washington, DC: The Con-
servation Foundation, 1986.
Blackburn J. Environmental mediation theory and practice: challenges, issues, and needed research
and theory development. In: Blackburn J, editor. Mediating environmental conflicts, theory and
practice. Westport, CT: Quorom Books, 1995.
Buckle LG, Thomas-Buckle SR. Placing environmental mediation in context: lessons from `̀ failed''
mediations. Environ Impact Assess Rev 1986;6(3):55±70.
Carpenter S, Kennedy WJD. Managing public disputes. San Francisco: Jossey-Bass, 1988.
Crowfoot J, Wondolleck J. Environmental disputes: community involvement in conflict resolution.
Washington, DC: Island Press, 1990.
Geertz C. Thick description: toward an interpretive theory of culture. In: Emerson R, editor. Con-
temporary field research. Boston: Little, Brown and Company, 1983. pp. 37± 59.
Guzzo R. Group decision making and group effectiveness in organizations. In: Goodman P, editor.
Designing effective work groups. San Francisco: Jossey-Bass Publishers, 1986. pp. 34± 71.
Kleiboer M. Understanding success and failure of international mediation. J Conflict Resol
1996;40(2):360± 89.
Kressel K, Pruitt D. Conclusion: a research perspective on the mediation of social conflict. In: Kressel
D, Pruitt D, editors. Mediation research: the process and effectiveness of third-party intervention.
San Francisco: Jossey-Bass Publishers, 1989.
Lee K. Defining success in environmental dispute resolution. Resolve 1982;1(Spring).
Moore S. Defining `̀ successful'' environmental dispute resolution: case studies from public land
planning in the United States and Australia. Environ Impact Assess Rev 1996;16(2):151 ±69.
O'Leary R. Environmental mediation: what do we know and how do we know it? In: Black-
burn J, editor. Mediating environmental conflicts, theory and practice. Westport, CT: Quorom
Books, 1995. pp. 17,32.
Pearson J, Thoennes N. Divorce mediation: reflections on a decade of research. In: Kressel D, Pruitt D,
editors. Mediation research: the process and effectiveness of third-party intervention. San Francis-
co: Jossey-Bass Publishers, 1989. pp. 9 ±30.
Scher E. Evaluations: what for, by whom who pays? Consensus. 1996;5.
Susskind L, Cruikshank J. Breaking the impasse. New York: Basic Books, 1987.
Todd S. Designing effective negotiating teams for environmental disputes: An analysis of three wolf
management plans. Doctoral Dissertation, University of Michigan, Ann Arbor, 1995.
S. Todd / Environmental Impact Assessment Review 21 (2001) 97±110 109
Wondolleck J, Yaffee S. Building bridges across agency boundaries: in search of excellence in the
USFS. A research report submitted to the USDA-Forest Service Pacific Northwest Research
Station, in fulfillment of USDA Forest Service Cooperative Agreement #PNW 92-0215. Ann
Arbor: School of Natural Resources and the Environment, The University of Michigan, 1994.
Zubek J, Pruitt D, Pierce R. Disputant and mediator behaviors affecting short term success in media-
tion. J Conflict Resol 1992;36(3):546±72.
S. Todd / Environmental Impact Assessment Review 21 (2001) 97±110110