ArticlePDF Available

Community Participation in Urban Sanitation: Experiences in Northeastern Brazil

March 2007
Journal of Planning Education and Research 26(3):284-300

March 2007
26(3):284-300

Authors:

Texas Southern University

This article explores how well community participation explains variations in the performance of sanitation projects in Brazilian cities. Case studies of condominial sewers in Recife and Natal, Brazil, reveal a variety of activities that we categorize into four forms of participation: mobilizing, decision making, construction, and maintenance. Mobilizing and decision making were associated with performance in the cases we examined. Our results suggest that participation form as well as community influence are important. The hypothesis that increased participation enhances performance is an oversimplification for condominial sewers.

Robustness of the performance categories to different configurations of the data. Note that although the levels of project performance change when measured with subjective versus objective data, these changes are unimportant because both sets of data are robust in their categorization of performance compared to case R2.

…

. Levels of authority of condominial sewer users.

…

Figures - uploaded by Earthea Nance

Content may be subject to copyright.

Content uploaded by Earthea Nance

Content may be subject to copyright.

284

The participation of users—in designing and implementing projects and managing

water and sanitation services—is now being built into [World] Bank funded projects

with the aims of increasing efficiency, equity and cost recovery and facilitating the exten-

sion of service coverage to poor communities.

Watson and Jagannathan (1995, 1)

Community participation is frequently identified by scholars and practitioners as

central to success in delivering physical infrastructure services (e.g., World Bank 2004).

However, much of the community participation literature does not carefully explain

the impacts of different forms of participation on project effectiveness. It is often

assumed that participation in all aspects of service delivery (e.g., project selection,

design, construction, and maintenance) contributes to more successful project out-

comes. Moreover, findings from the community participation literature are sometimes

applied across a broad spectrum of service sectors (e.g., water, sanitation, education,

and health care) and settlement types (e.g., rural, urban, suburban, and periurban).

Although we were guided by the literature to believe community participation would

contribute positively to project effectiveness, we sought to examine the concept of par-

ticipation more carefully by distinguishing among forms of participation and by view-

ing participation in infrastructure projects as part of a broader network of relationships

among citizens, sewer practitioners, and officials.1

Most of the empirical evidence concerning the performance outcomes of projects in

which community members participate is based on studies of rural water projects (e.g.,

Isham and Kahkonen 1999; Narayan 1995; Ostrom 1992; Prokopy 2002, 2005; Wade

1988; Sara and Katz 1997) and from other public service sectors where the concept of par-

ticipation may be viewed as a form of “co-production”; that is, citizens provide inputs to

services that are traditionally produced exclusively by public agencies2(Brudney and

England 1983; Percy 1984; Sundeen 1985; Levine 1984). For example, citizens have been

involved in the co-production of metropolitan police services (e.g., Ostrom, Parks, and

Whitaker 1978; Parks et al. 1982) and primary education (e.g., Ostrom 1996). Support

for participation has emerged in each of these sectors. One reason is that some degree

of social change—new habits, skills, or awareness by the beneficiaries of the service—is

often an important aspect of service delivery outcome (Whitaker 1980). Community par-

ticipation has been found to be effective in the provision of public or quasi-public services

(as opposed to commodity goods), networked services that are difficult to distribute, and

services in which contributions from citizens and government agencies are complements

Abstract

This article explores how well community

participation explains variations in the

performance of sanitation projects in

Brazilian cities. Case studies of condo-

minial sewers in Recife and Natal, Brazil,

reveal a variety of activities that we catego-

rize into four forms of participation: mobi-

lizing, decision making, construction, and

maintenance. Mobilizing and decision

making were associated with performance

in the cases we examined. Our results sug-

gest that participation form as well as com-

munity influence are important. The

hypothesis that increased participation

enhances performance is an oversimplifi-

cation for condominial sewers.

Keywords: community participation; Brazil;

Latin America; municipal and public services;

sanitation; condominial sewers

Journal of Planning Education and Research 26:284-300

DOI: 10.1177/0739456X06295028

Earthea Nance is an assistant professor of

urban environmental studies at Virginia

Tech. During the 2005/2006 academic

year, she served as a Martin Luther King

fellow and visiting assistant professor in

the Department of Urban Studies and

Planning at Massachusetts Institute of

Technology. Her main areas of research

and teaching are community participa-

tion and grassroots organizing, urban

infrastructure, environmental justice,

environmental planning and manage-

ment, anthropological research methods,

and Brazil/Latin America.

Leonard Ortolano is the UPS Foundation

Professor of Civil Engineering at Stanford

University. Between 2003 and 2006, he

was also director of the Haas Center for

Public Service. His research and teaching

focus on environmental planning and

policy implementation, with an emphasis

on issues linked to air and water quality

management and the delivery of water

and sanitation services to disadvantaged

communities.

Community Participation in Urban Sanitation

Experiences in Northeastern Brazil

Earthea Nance & Leonard Ortolano

Community Participation in Urban Sanitation 䉳285

and not substitutes (Ostrom 1996). Brazil’s “condominial” sew-

ers are a good example of co-produced service delivery, because

citizens work directly with government agencies to produce and

provide the service.

䉴Condominial Sewers in Brazil

Brazilian engineers began experimenting in the early 1980s

with a decentralized, low-cost approach to wastewater collec-

tion known as condominial sewers. These systems are less

expensive than conventional sewers, because sewer laterals are

installed under sidewalks or yards of homes instead of under

streets. Costs are lower because laterals can be made with less

expensive material than is used in conventional sewers. Costs

can be further reduced by having project beneficiaries partici-

pate in constructing and maintaining portions of the system,

most commonly the household connections—sections of the

sewer line running from the house to the clean-out box that is

part of the condominial sewer line.

Condominial sewers were developed in Brazil to expand

household wastewater collection service to unserved (often low-

income) population groups (Sinnatamby 1983; World Bank

1994). More specifically, condominial sewers were adopted to

overcome many of the obstacles associated with conventional

sewers, including (a) the high cost of conventional sewer sys-

tems, (b) low household connection rates to conventional sys-

tems because of a lack of demand at prevailing prices, and (c)

people’s aspirations for a flexible sewerage solution that could

be upgraded to a conventional sewer as incomes rose (Bakalian

and Jagannathan 1991). An implicit hypothesis of the condo-

minial sewer approach is that community participation in

project planning, design, construction, and maintenance con-

tributes to better project outcomes. In their comments about

condominial sewer technology, for example, Bakalian and

Jagannathan (1991, 1) note that “its success hinges on the readi-

ness of the community to participate in the various phases of

the project cycle.” In addition to their use in Brazil, condo-

minial sewer systems have been constructed in Bolivia,

Guatemala, Zambia, and Pakistan (Mara 1996).

A key research objective was to determine whether and

how different forms of community participation are associ-

ated with project outcomes. We were motivated to investigate

community participation in condominial sewers because the

literature does not offer any project-scale case studies on how

people participate in condominial sewers, nor does it indi-

cate which forms of participation are associated with various

project outcomes. On the first point regarding how people

participate, we conceived of participation in condominial

sewer projects in the same way that practitioners in sanitation

agencies in Brazil think about it, that is, in the context of a

four-part “project cycle” that provides a framework for man-

aging the great number of tasks involved in taking a project

from conception to operation. The four parts consist of plan-

ning, design, construction, and maintenance. Condominial

sewer practitioners in Brazil have devised ways to engage the

community in each of these project phases. Mobilization,

which is the form of participation that occurs (more or less)

during the planning phase, involves planners bringing

together residents to explain the costs and benefits of condo-

minial sewers. The degree to which residents can be mobi-

lized is felt to reflect demand for condominial sewers, and

many analysts believe that demand is linked to project

success.3Decision making, where local insights and local

knowledge are brought to bear on project decisions, is the

form of participation that takes place (more or less) during

the design phase. The potential link to project success is that

this local knowledge will lead to designs that are more sensi-

tive to local conditions and the needs of the community to be

served. Community participation in project construction is

thought to enhance the likelihood of project success by

reducing project costs and bringing about a sense of owner-

ship among the participants. Community participation in

project maintenance takes place during the final stage of the

project cycle, and regular and timely maintenance can clearly

be linked to positive project outcomes.

Community participation in construction and mainte-

nance activities, which are normally performed by sanitation

agency staffs or contractors, can potentially lead to better cost

recovery and higher connection rates. However, having resi-

dents perform this work would not necessarily be more effec-

tive in terms of the operability and health impacts of the

system because the quality of construction and maintenance

work performed by residents may not be high. This might

appear to be an obvious statement, but it has not been

addressed in the participation literature because the concept

of participation has rarely been disaggregated by form.

Our research question may be stated as follows: Do differ-

ent forms of participation contribute differently to project

outcome, and if so, how? In investigating this question, we

gave particular attention to the way in which project outcome

is conceptualized and measured because we felt that different

forms of participation might affect different types of project

outcomes. Sewer performance evaluations most often use

cost recovery and coverage (e.g., the percentage of a com-

munity served by sewers) as measures of efficiency-based out-

comes, but these overlook important effectiveness-based

outcomes, such as the impacts on health and livelihoods that

are a consequence of how sewer systems actually function.

The effectiveness-based performance outcomes used in our

cases are detailed in Nance (2005) and summarized in a later

section of this article. By disaggregating participation into

four components and focusing on effectiveness-based out-

come measures that include impacts on the health and liveli-

hoods of the communities served, a nuanced approach to

studying the impacts of participation is possible.

The literature also lacks a discussion of the relative impor-

tance of community participation compared to other factors

that may explain the performance of condominial sewers,

and this is a subject explored later in this article. Our

research builds on Gabrielle Watson’s (1995) empirical study

of condominial sewers.4The objective of Watson’s study was

to assess “institutional issues, with the aim of highlighting the

achievements of the condominial system and pointing to

areas that require more attention” (Watson 1995, 7). She

described condominial sewer experiences in seven Brazilian

cities; however, her study did not assess participation and per-

formance at the project (i.e., neighborhood) level. Based on

the mixed performance results reported in Watson’s survey,

follow-on research at the project level is crucial to under-

standing how participation shapes condominial sewer perfor-

mance. We build on Watson’s study by tracing participation

experiences at the level of the neighborhoods served by con-

dominial sewer projects.5

By design, condominial sewerage in Brazil in the 1980s

and early 1990s was often organized in a decentralized fash-

ion, and this reduces the value of a city-level representation

of participation practices and performance outcomes.

Condominial sewers were produced in different ways by dif-

ferent communities and agencies, and they were installed in

highly variable circumstances. For example, in some cities,

condominial sewer services were delivered one project at a

time in individual neighborhoods that were not unified by a

public trunk sewer system. Many Brazilian cities did not have

trunk sewers that served the entire city. Instead, the trunk sys-

tems typically served only the wealthier areas of cities. The

majority of homes either relied on individual on-site systems

(i.e., pit latrines, septic tanks, and leach pits), or they released

wastewater to storm drains, street gutters, and open ditches.

In many cities, when sewer coverage was expanded using con-

dominial sewers, those sewers were not linked to the trunk

system. Sewage from the condominial sewers was discharged

untreated into ponds, canals, and storm drains, often shifting

sanitation problems to other parts of the city rather than solv-

ing them. The service provided was not uniform across a city,

and the level of community involvement in condominial sew-

ers and the quality of service outcomes were generally neigh-

borhood specific. Given this local variability, household and

neighborhood level information is needed to fully under-

stand people’s activities in project development and to reason

about how their participation affected project performance.

In addition to variation at the neighborhood (i.e., project)

level, there are important differences at the city level that can

contribute to our understanding of the conditions in which par-

ticipation affects project performance. A city that has condo-

minial sewer projects delivered by a single agency using a

coordinated approach (and standardized procedures for involv-

ing residents in project design and implementation) would be

expected to produce a different set of project outcomes from a

city that has several uncoordinated agencies implementing pro-

jects in an ad hoc manner. The two cities included in our inves-

tigation—Recife and Natal, Brazil—employed these two

contrasting approaches during the period studied. In Natal,

condominial sewer projects were implemented by one entity,

the North Rio Grand State Water and Sanitation Agency, which

embraced condominial sewer technology and was committed to

community participation. This agency implemented new con-

dominial sewer projects as part of a plan to expand the trunk

sewer network. In Recife, four agencies were involved in con-

dominial sewer implementation: the State Water and Sanitation

Agency of Pernambuco, the State Housing Agency, and two

municipal bodies—the Urban Planning Agency and the Public

Works Agency. These four agencies did not always cooperate,

and they did not follow a cohesive urban development plan

with regard to sanitation. The agencies also differed in their

support of condominial sewers as an appropriate technology

and in their commitment and approaches to community

participation.

In the remainder of this article, we explore the following

questions at the project level: What forms of community par-

ticipation were used in each of several case study condo-

minial sewer projects in Recife and Natal, and which forms

were associated with enhanced sewer service outcomes? We

begin by presenting our overall research approach.

䉴Research Methods

Preliminary fieldwork was conducted in May 1994. Most of

the data were obtained during ten months of fieldwork from

October 1994 to July 1995, and follow-up observations were

made in January and February 2006.6The fieldwork was con-

ducted in Brazil’s poorest region, the Northeast, in the capi-

tal cities of Recife and Natal.7Because of its low cost,

condominial sewerage was particularly appropriate for the

Northeast region, and at the time of the initial fieldwork, the

region had the most widespread experience in Brazil with

condominial sewers. We selected Recife and Natal as research

sites because each city had over five years of condominial

sewer experience and at least five condominial sewer projects

in various stages of implementation, and each city had offi-

cials and agency staffs willing to cooperate with us. In addi-

tion, some written information was available on the

condominial sewers in each of the cities.

䉴Use of Purposeful Sampling to Select Cases

We chose not to use a large-sample, statistically oriented

research approach. Although that strategy would have facili-

tated generalization, it would not have allowed us to obtain the

detailed understanding of project design and operations that

286 Nance & Ortolano

Community Participation in Urban Sanitation 䉳287

we felt we needed. We wanted details on the role of community

participation in each project so that we could investigate links

between different forms of participation and project outcome.

We also wanted to examine how factors unrelated to participa-

tion might have influenced outcomes. A multiple case study

approach allowed us to obtain an appropriate level of detail.

Our strategy for selecting cases was to identify contingently

a small number of projects that, in the judgment of condo-

minial sewer practitioners at local sanitation agencies in Recife

and Natal, represented examples of particularly good and bad

performance. In the course of our detailed case studies, we

could verify whether the projects selected were in fact good or

bad in terms of performance. The cases employed a multi-

stakeholder approach to develop a comprehensive assessment

of what constituted high- and low-performing projects.

This case selection technique, which purposefully selected

illustrative good and bad projects, is grounded in the standard

literature on qualitative research design.8Patton (2002, 231)

characterizes the approach as a form of purposeful sampling

that he refers to as “extreme sampling.”

On the basis of our conversations with condominial sewer

practitioners in Recife and Natal, we knew (in advance of

selecting our cases) that some projects that practitioners con-

sidered good performers had poor community participation,

and some projects that they considered poor performers had

good community participation. In selecting our case studies,

we used an additional criterion related to whether, in the judg-

ment of condominial sewer practitioners, the projects in their

cities had levels of community engagement in their participa-

tion programs that were “good” or “bad.” This information on

levels of participation, together with information on practi-

tioners’ opinions regarding performance of projects in their

cities allowed us to array twenty-one different projects in a two-

by-two table, with cells representing good and bad project per-

formance and good and bad participation (see Table 1).9We

then selected cases such that we had at least one case from

each cell.

We considered participation in choosing cases because we

wanted to avoid the possibility of ending up only with good per-

forming cases that had good participation and bad performing

cases that had bad participation. Our principal interest in doing

this case study work was to explore relationships between com-

munity participation and project performance. A set of cases

involving projects from all four cells of the two-by-two table

would provide us with the most information about possible rela-

tionships between participation and performance. Moreover,

this selection would allow us to avoid having only cases where

good projects had good participation and bad projects had bad

participation, even though that was consistent with our initial

hypothesis (i.e., what we expected based on the community par-

ticipation literature). We make no claim that the cases we exam-

ined are a representative sample in a statistical sense. They were

not selected to be representative but to maximize our opportu-

nities to learn.10 These aspects of our approach must be kept in

mind when considering our conclusions.

A particular challenge in this approach is that we could not

confirm the judgments about best and worst cases offered by

sanitation agency staffs until after we conducted detailed case

studies. As explained later, sanitation agency staffs have their

own views on what is effective, and those views may not corre-

spond with what is observable “on the ground,” or to the views

of community members using the sewers. For this reason, we

sought viewpoints from multiple stakeholder groups to con-

firm that we were maximizing differences among the cases.11

This type of challenge is common in qualitative research

design. In comparison to quantitative researchers who are con-

cerned about a form of systematic error known as selection

bias, qualitative researchers are often more concerned about

selecting cases that clearly illustrate a phenomenon (of which

the case is intended to be an example) within its context. But

the phenomenon of interest is frequently underdetermined to

some extent. Project X might be selected for study because it is

thought to be a case of success, but whether the case is a success

cannot be determined until the case study is completed. To

solve this problem of underdetermined outcome, case study

researchers deliberately select cases on specific variables of

interest—often on the extreme values of the variables—knowing

Table 1.

Two-by-two matrix of candidate projects

from which the cases were selected.

Good Bad or No

Participation Participation

Good performance Alemao

Norte e Sul

Playa Bebera

Lago Santa Paula

Rio Jorge Jardim Sepa

Entra Paradiso Triburi Alto

Vila Triburi Santa Ilena

Bad performance Rosa

Quatros

Bom Freire

Brasilia Jose Pico

Vila Maria Bambino

General Fabio Cabanga

Casa Jeri Grand Favela

Source: Nance (2004).

Note: Underlined projects are located in Recife; projects in italics

are located in Natal. At the time these data were gathered, Recife

had many small projects, and Natal had fewer large projects. Also,

the projects in Recife had a higher rate of performance failure

compared to the projects in Natal. Not listed are nineteen addi-

tional projects for which practitioner judgments of performance

and participation were ambiguous (ten in Recife and nine in

Natal). Original project names have been changed to preserve

the anonymity of the communities.

288 Nance & Ortolano

that their subsequent analysis may challenge their initial pre-

sumption about the variables. Moreover, in contrast to statisti-

cal studies, the qualitative case selection process does not

presume a normally distributed population of cases. Some of

the most interesting single-case studies address not the average

case but an unusual, critical, or revelatory case that needs to be

explained. These cases would be ignored in a quantitative selec-

tion process because they would be considered outliers. For

small-sample research designs, polar cases (i.e., case selection

on the extremes of the variables of interest) frequently offer a

superior frame of observation than do average cases (i.e, selec-

tion on average or middle values) or randomly selected cases.12

Seven Projects in Recife and Natal

We chose three projects in Recife (i.e., cases R1, R2, and R3)

and four projects in Natal (i.e., cases N1, N2, N3, and N4) for

in-depth study. We assigned pseudonyms to the cases to main-

tain household and neighborhood anonymity and to comply

with privacy requirements for research involving human sub-

jects. Table 2 provides an overview of the seven cases.

Participants in the study included residents living in the case

study project areas; community leaders; engineering, commu-

nity participation, and maintenance staffs of the sanitation agen-

cies; elected officials, agency managers, and appointed agency

directors involved in planning and funding condominial sewer

projects in Recife and Natal; and other relevant personnel such

as employees of consulting and construction companies who

worked as contractors for the implementing agencies. Except

for residents, we used a snowball sampling technique to identify

study participants based on their experience, knowledge, posi-

tion, and involvement in local condominial sewer projects.

Representative samples of households were randomly selected

in each case study project area.13 In all, we conducted in-depth,

semistructured interviews with residents from 297 households

Table 2.

General information about the seven selected case study projects.a

Percentage of Percentage of Sewer

Case Implementing Implementing Total Estimated Income CondominialeOverall Sewer Performance

Study Period Agency(s) bHomescPopulation LeveldSewer Coverage Coverage fIndex

R1 1986-1988 OBRAS, 293 1,327 Middle 79 82 67

COMPESA

R2 1987-1992 URB, 158 705 Low 48 48 12

COMPESA

R3 1987-1990 COHAB, 1,349 6,340 Poor 100 100 71

COMPESA

N1 1987 CAERN 384 1,789 High 36 92 68

N2 1986-89 CAERN 757 3,535 Low 76 96 66

N3 1986-89 CAERN 964 3,663 Lower-Middle 76 93 55

N4 1987 CAERN 152 668 High 52 100 72

Source: Nance (2004).

a. The information in this table is based on data gathered during the main fieldwork period in 1994 to 1995. On a follow-up visit to the

case study areas in January and February 2006, we observed the following: a general increase in commercial activity in all the neighbor-

hoods, the addition of new buildings and the physical degradation of older buildings, significantly increased crime in the case N2 neigh-

borhood, a significant improvement in the functioning of the treatment plant near the case N3 neighborhood, and a slightly increased

presence of gray water and sewage in all of the neighborhoods except in the case R2 neighborhood. The open ditches in the case R2

neighborhood had been lined with concrete half pipes by the city, which had the effect of drying the surrounding soil and improving the

overall environment compared to previous conditions but had little to no impact on the relative performance of the case R2 sewer sys-

tem compared to the other cases. On the basis of these recent observations and by comparison to the original observations in 1994 and

1995, we suspect that if we were to reanalyze the performance of these projects today, they would remain in the same relative performance

categories even though the value of each sewer performance index would likely change. In any event, the 2006 observations reported

above have no bearing on the analysis presented in this article, which concerns the periods of project implementation and project oper-

ations approximately five years after construction.

b. During the implementation period, OBRAS was Recife’s public works department, COMPESA was Recife’s state water and sanitation

agency, URB was Recife’s city urban planning department, COHAB was Recife’s state housing agency, and CAERN was Natal’s state water

and sanitation agency.

c. To determine the total number of households in each project area, we conducted a neighborhood census in which every household in

each project area was counted.

d. Income levels are categorized as follows: high income is >10 minimum salaries per household, middle income is >5 and ≤10 minimum

salaries per household, low income is >3 and ≤5 minimum salaries, and poor is >1 and ≤3 minimum salaries per household.

e. Sewer coverage (i.e., the percentage of households served) was estimated based on interviews with a representative number of ran-

domly selected households in each project area (at ρ≤0.10). Sample sizes (by case) ranged from thirty-three to fifty-nine households

with a median of forty-one.

f. Overall sewer coverage includes both condominial and conventional sewer connections.

Community Participation in Urban Sanitation 䉳289

and with thirty-one agency staff members

and officials who were each knowledgeable

about condominial sewer projects.14

Condominial sewer performance, an

important concept in our study, was

defined as the sum of the operational effec-

tiveness and beneficial impacts of an indi-

vidual condominial sewer project,

expressed as a sewer performance index

(SPI). This index consisted of twenty-seven

indicators of performance—twenty indica-

tors of operational effectiveness and seven

indicators of impacts—that came from

direct observation of the systems in opera-

tion and from information obtained in

interviews with three groups of stakehold-

ers: residents, sanitation agency engineers,

and sanitation agency maintenance staff

members. Examples of operational effec-

tiveness measures include the operability of

household connections and the existence

of sewer line blockages caused by soil,

trash, and damaged pipes. Indicators of

impacts included ratings by residents of satisfaction with the

sewer system and direct observations of sewage on the surface

of the ground. Each indicator was converted to a score on an

ordinal scale from 0 (least favorable) to 100 (most favorable

performance possible with the measurement instruments

used). All of the scores were integrated into a single index of

performance by adding the median of the twenty operability

scores and the median of the seven impacts scores. Each

median was weighted equally so as not to place arbitrary

emphasis on operability, for which more indicators were avail-

able. For purposes of this discussion, SPI scores in the range of

0 to 33 are characterized as low; scores between 34 and 67 are

characterized as mid-range; and scores from 68 to 100 are char-

acterized as high. Results from these calculations are shown in

the last column of Table 2. For an in-depth explanation of the

conceptualization, measurement, and calculation of the sewer

performance index, see Nance (2005).

Figure 1 compares the results of recalculating the sewer per-

formance index using all subjective indicators (indicators 1-16

and 21-24) and all objective indicators (indicators 17-20 and

25-27). Although the data can be combined in many different

ways, by any reasonable combination of the individual indica-

tors, case R2 is a poorly performing project, and the other cases

all achieved significantly higher levels of performance. Put

another way, different stakeholders scored projects differently

but ranked them similarly. Evidence for these claims is pre-

sented in Nance (2005). She found that “different stakeholders

scored the same project differently . . . [b]ut these results do

not indicate any pattern in stakeholder variability between cases

with poor and good performance” (p. 496). Based on her over-

all analysis of performance, Nance found that engineering staff

judgments of performance were unreliable and that engineers

had lower internal agreement about performance than project

area residents and maintenance staff members. (Engineers play

a central role in planning, design, and implementation, but

compared to maintenance staffs and residents, they are less

familiar with project operations.) Also, maintenance staff mem-

bers had high internal agreement about performance but were

far more critical of performance than the other stakeholder

groups. For these reasons, Nance (2005) concluded that a mul-

tistakeholder approach to project evaluation (using multiple

indicators) was superior to reliance on the viewpoints of a single

stakeholder group.

Measuring Community Participation

Community participation was defined at the outset of the

study as the involvement in and contributions to a project by

individual residents, households, and the community as a

whole.15 We disaggregated this variable into four forms of par-

ticipation relevant to condominial sewers: involvement in

mobilizing support for a project, involvement in decision mak-

ing, contributions to construction work, and contributions to

maintenance work.These were the forms of participation

mentioned by practitioners and beneficiaries as relevant to

implementing various phases of the project cycle for condo-

minial sewer systems. We represented each form of participa-

tion as an index consisting of various participation measures.

Each measure was expressed on an ordinal scale from 0 to

100. A total of thirty-one participation measures were used,

and scores were assigned based on data obtained from case

study documents and interviews with residents and sanitation

100

R1 R2 R3 N1 N2 N3 N4

Subjective Objective

Figure 1. Robustness of the performance categories to different configurations of the data. Note

that although the levels of project performance change when measured with subjective

versus objective data, these changes are unimportant because both sets of data are

robust in their categorization of performance compared to case R2.

290 Nance & Ortolano

agency staffs involved in community participation activities.

The interview questions are presented in Tables 3, 4, and 5;

conceptualizations of the different forms of participation are

described below.

The term mobilizing represents activities in which the

implementing agency described to residents the nature of

condominial sewers and their advantages in terms of

improved health and sanitation and otherwise engaged resi-

dents. The goal was to foster widespread support for the pro-

ject by actions geared toward building a community-wide

base of knowledge about condominial sewers (e.g., house-to-

house visits to teach every resident about what condominial

sewers are and how they function) and by gatherings where

residents could express their needs and preferences con-

cerning sanitation in general and wastewater collection in

particular. Case R2 residents were not well mobilized prior to

or during project implementation. In contrast, residents in

each of the other six cases were either well mobilized prior to

the start of the project or mobilized effectively during imple-

mentation. These cases demonstrate the variety of ways in

which people participated in mobilization activities for con-

dominial sewer projects. Mobilization activities were used to

express or to create demand for a project, to exchange infor-

mation, to educate residents about the value of sewerage,

and to give residents a forum to complain collectively about a

project. Sometimes residents organized themselves, and at

other times, they were mobilized by the implementing agency.16

We conceived of participation in decision making in terms

of the extent to which residents took part in project-related

decisions, such as choosing among backyard, front yard, or

sidewalk condominial sewer layouts. In case R2, the case with

poor performance, residents did not participate in any

project-related decisions. Residents were not asked to sign,

authorize, or accept the condominial sewer, and they were not

involved in selecting the level of service (which corresponded

with the tariff level) or the system layout. In contrast, residents

in each of the six cases with good performance were involved in

some aspect of project decision making. Residents in all cases

except R2 were involved in one or more of the following key

decisions:

•Deciding whether to connect to the sewer line;

•Selecting the type of service and tariff level;

•Deciding the layout of pipes and cleanout boxes;

•Selecting the materials of construction; and

•Determining which household facilities would be con-

nected to the condominial sewer.

Participation in construction and maintenance work was

measured based on whether beneficiaries contributed time,

money, tools, materials, and labor for project construction

Table 3.

Interview questions for agency staff who

implemented participation programs.a

Form of Participation Interview Question

Mobilizing Were residents mobilized?

Did residents receive an orientation to

the project?

Did residents receive education about

sanitation?

Were user groups formed?

Did residents attend meetings?

Were residents visited in their homes?

Did residents receive project literature?

Decision making Did residents decide the layout of the

system?

Did residents decide to accept the service

level?

Construction Did residents construct condominial

work sewers?

Did residents purchase materials?

Were houses connected to the

condominial sewer?

Were residents responsible for

constructing house connections?

Maintenance Did residents accept responsibility for

work maintenance?

Did residents perform maintenance?

Source: The entire semistructured interview instrument for

participation staff is available (in Portuguese) in Nance (2004,

app. A).

a. Affirmative answers were further explained by agency staff to

reduce ambiguity in the meaning of terms.

Table 4.

Interview questions for residents. a

Form of Participation Interview Question

Mobilizing Did you participate in mobilizing

residents?

Did you participate in project meetings

with residents?

Did you participate in project meetings

with agency staff?

Did you discuss the project with agency

staff?

Were you involved in a sanitation

education program?

Were you involved in a health program?

Did you respond to questionnaires from

agency staff?

Decision making Were you involved in selecting the

layout for the project?

Construction work Did you authorize project construction?

Did you contribute labor, money,

property, tools, materials, etc.?

Did you construct your house

connection?

Maintenance work Who fixes your household’s condominial

sewer when there is a problem,

including paying for the expense?

Source: The entire semistructured interview instrument for resi-

dents is available (in Portuguese) in Nance (2004, app. A).

a. Affirmative answers were further explained by residents to

reduce ambiguity in the meaning of terms.

Community Participation in Urban Sanitation 䉳291

and maintenance, respectively. Residents from all seven cases

participated in construction of their individual household

connections, but they participated in different ways.

Participation by residents in constructing household connec-

tions varied widely and took the following forms: doing the

work by themselves, providing materials and having the sewer

agency do the work, and paying plumbers to construct the

connections. Residents participated in construction differ-

ently depending on several factors, including skill level,

household income, time availability, and the prerogative of

the implementing agency. Residents in all seven cases made

formal commitments to participate in maintenance, but their

actual participation varied depending on household income

and the level of maintenance support that each received from

the agency responsible for sewer maintenance. Even when

residents tried to do maintenance, they often lacked the skills

or tools to succeed and needed professional support from the

responsible sanitation agency. Even though residents had

previously agreed to handle maintenance on their own, many

refused to do so because they believed that maintenance was

the government’s responsibility.

Participation activities linked to mobilizing, decision mak-

ing, construction, and maintenance were measured along

three dimensions: (1) the scope of the activities, (2) the num-

ber of participants involved, and (3) the overall level of

authority of the residents to initiate and control their partici-

pation activities. We measured participation along these

three dimensions because we felt that important features of

participation would be missed by using any one dimension

alone. In some cases, for example, many participants showed

up for project meetings, but the scope of how they could par-

ticipate was limited. In this instance, participation in terms of

number of participants scores high, but participation mea-

sured as scope of activities scores low. The third dimension

(level of authority) contributes by recognizing that initiating

action and taking control are the strongest manifestations of

participation, regardless of the scope of the action or the

number of residents involved. Without the use of multiple

dimensions of participation, it would not have been possible

to account for these types of distinctions.

We used responses from interviews with agency participa-

tion staff to determine the scope of the agency’s participation

program for each of our case study projects (see Table 3). We

employed responses from interviews with project area resi-

dents to estimate the number of participants who had

engaged in each form of participation (see Table 4). We

adapted an existing participation scale developed by Paul

(1987) to assign scores to the level of authority exercised by

participants in each form of participation (see Table 5). As a

final step, we used Equation 1 to compute an index for each

form of participation, expressed on an ordinal scale from 0 to

100; 0 indicated no participation, and 100 indicated the high-

est degree of participation possible with the selected set of

participation measures.

Participation Index =A·wA+P·wP+S·wS(1)

where: A =level of user authority (assigned using Paul’s

scale)

P =number of participants (responses from residents)

S =scope of participation (responses from staff)

w =weighting factor

wA+wP +wS =1

Participation index results reported here are based on cal-

culations with the three weights in Equation 1 equal to each

other. We used equal weights because we had no basis for

valuing one dimension of participation more than any other.

Table 5.

Levels of authority of condominial sewer users.

Form of Participation Level of User Authority a

Mobilizing 0. Little or no mobilization of users.

1. Agency mobilizes users.

2. Agency mobilizes users with

assistance from community leaders.

3. Users mobilize themselves to respond

to the agency.

4. Users mobilize themselves without

seeking response from the agency.

Decision making 0. Agency selects service level and does

not inform users.

1. Agency selects service level and

informs users.

2. Agency provides service options to

users.

3. Users select service level and seek

agency approval.

4. Users select service level and do not

seek agency approval.

Construction 0. Little or no construction of house

work connections.

1. Agency constructs and informs users.

2. Agency constructs with user input.

3. Users construct with agency

assistance.

4. Users construct without agency

assistance.

Maintenance 0. Little or no maintenance of

work condominial sewer.

1. Agency maintains and informs users.

2. Agency maintains with user input.

3. Users maintain with agency

assistance.

4. Users maintain without agency

assistance.

a. Participation scores (i.e., 0=0, 1=25, 2=50, 3=75, 4=100) were

assigned based on the level of user authority determined from

detailed case study descriptions presented in Nance (2004, apps.

E and F). This scoring methodology was adapted from Paul’s

(1987) 4-point scale of participation intensity.

䉴Results

Participation Form and Project

Performance

Performance scores for each case are

listed in the last column of Table 2. Case R2

received a low sewer performance index

(SPI) score of 12. Cases R1, N2, and N3

received mid-range SPI scores of 67, 66, and

55, respectively. Cases R3, N1, and N4

received high SPI scores of 71, 68, and 72,

respectively. However, because the mid-

range and high scores were clustered within

a relatively small range (i.e., 55 to 72), the

differences among them were not statisti-

cally significant (at ρ≤0.05) based on a

nonparametric statistical test for ordinal

data. Consequently, the seven cases were

placed into two performance categories.

Case R2 made up the poor performance

category and the remaining six cases made

up the good performance category.

The case study performance outcomes

that we computed based on our detailed

studies came as a surprise. At the outset of

our study, we asked staff members to identify

projects that performed particularly well and

particularly poorly. Based on their views, we

anticipated that at least two of the seven pro-

jects would rate as poorly performing pro-

jects, one each in Recife and Natal. In

retrospect, our strategy did not account for

two significant factors. One was the overall

difference in condominial sewer perfor-

mance in Recife and Natal. As explained

below, Natal had a good citywide record of

performance with condominial sewer pro-

jects; Recife’s record was mediocre at best.

The second factor, which is detailed in

Nance (2005), is that initial appraisals of per-

formance by agency staffs we interviewed for

purposes of case study selection did not accu-

rately reflect how projects actually per-

formed. This situation highlights the

difficulties involved in selecting representa-

tive projects based only on the views of sani-

tation agency staffs. As a consequence of

these two factors, we had only one poorly

performing case among the seven selected.

On the basis of previously cited literature on community

participation, we expected to find relatively high levels of par-

ticipation associated with the cases in the good performance

category and a low level of participation for the case in the

poor performance category. But the overall participation

results for these seven cases, illustrated in Figure 2-A, are not

292 Nance & Ortolano

100

R1 R2 R3 N1 N2 N3 N4

Cases

Participation

100

R1 R2 R3 N1 N2 N3 N4

Participation in

Mobilizing & Decisions

Cases

100

R1 R2 R3 N1 N2 N3 N4

Participation in

Construction & Maintenance

Cases

Figure 2. Levels of participation used in the seven cases.

Community Participation in Urban Sanitation 䉳293

consistent with this expectation. As shown in that figure, all

seven cases received mid-range scores for overall participa-

tion, ranging from 45 to 65 (with an average score of 52),

despite the much wider spread in levels of performance

among the seven cases.

A different view emerges if we disaggregate the concept of

participation by breaking it into two parts: participation in

mobilizing and decision making, and participation in con-

struction and maintenance work (see Figures 2-B and 2-C). In

Figure 2-B, the poorly performing case had a considerably

lower level of participation in mobilizing and decision making

compared to the cases with good performance. In Figure 2-C,

both the best and worst performing cases (cases N4 and

R2, respectively) had considerably higher levels of participa-

tion in construction and maintenance work compared to the

other cases.

For these seven cases, participation in construction and

maintenance work was not associated with condominial sewer

performance, but participation in mobilizing and decision

making was associated with performance.17 Our original

hypothesis concerning the influence of participation might be

restated as follows with regard to condominial sewer projects:

higher levels of participation in mobilizing and decision making are

associated with improved performance as measured by the condominial

sewer performance index.18

Significant Differences between Recife and Natal

At the time of our initial fieldwork (1994 to 1995), about 24

percent of Recife’s population was connected to the public

sewer system, which included both conventional and condo-

minial sewers.19 The majority of the population discharged

sewage to on-site systems, storm drains, street gutters, or open

ditches. A small portion of the collected sewage was treated,

but most was released untreated and eventually flowed into

Recife’s many canals and rivers. At that time, the city of Recife

had a poor overall record of condominial sewer project per-

formance, largely because of the ad hoc nature of project

implementation by four different agencies.20 More than half of

the forty-three condominial sewer projects implemented in

Recife from 1980 to 1995 had less than fifty household con-

nections each (Nance 2004). Many condominial sewer projects

initiated during that time period were never completed, a

large percentage of projects barely functioned, and most pro-

jects were not designed with a suitable destination for the col-

lected wastewater. Although a few projects were successful,

sanitation agency staff we interviewed unanimously character-

ized Recife’s overall performance in wastewater collection by

condominial sewers as fair when choosing among the cate-

gories bad, fair, good, very good, and excellent.21 As one of our

interviewees put it, “Sanitation is bad here in Recife.”22 It is

notable that the two cases in Recife that exhibited good per-

formance based on their SPI scores (i.e., cases R1 and R3) were

uncommon outcomes.

Additional aspects of the institutional setting in Recife

affected the process of implementing condominial sewer pro-

jects. We have already mentioned the coordination difficulties

associated with having four different agencies involved in pro-

ject implementation, particularly when those agencies chose

not to cooperate. These problems were compounded because

technical staffs within the agencies often differed in the degree

to which they believed condominial sewer systems were a suit-

able technology for use in Recife. Differences also existed

among agencies regarding the desirability of community par-

ticipation in project implementation. Another important fac-

tor concerned the lack of alignment that sometimes existed

between the goals of elected officials in Recife and the state of

Pernambuco. During some years, officials at one level of gov-

ernment promoted the use of condominial sewers, but officials

at the other level felt that condominial sewer technology was

unsuitable. Under the circumstances, changes in elected offi-

cials at the state or city level could lead to funding disruptions

in ongoing projects. For example, funding for a project

promised by a mayor to a particular neighborhood might be

cut off abruptly when a new mayor is elected. The logic of

patronage and clientelism that is common in Brazil can easily

preclude continuity from one regime to the next, and Recife

experienced these disruptions in continuity for the period cov-

ered by our study (Nance 2004, chap. 5).

The circumstances in Natal were quite different from

those in Recife. In 1995, approximately 20 to 25 percent of

Natal’s population was connected to either conventional or

condominial sewers, and about 90 percent of the condominial

sewer connections served low-income and poor residents.23 As

of 1995, Natal had a good overall record of condominial sewer

project performance, which can be attributed to the efforts of

a single sewer agency that was, on the whole, committed to the

technology. The majority of projects initiated in the 1980 to

1995 period were either completed or in the process of being

completed as of 1995 (Nance 2004). Most of these projects

functioned adequately and released collected sewage to a suit-

able destination (e.g., a trunk sewer or a community septic

tank). However, the sewage treatment plant, which was a sin-

gle stabilization pond, was barely maintained. Agency staff we

interviewed characterized Natal’s overall performance as

good when choosing among the categories bad, fair, good,

very good, and excellent.24 This characterization is consistent

with results obtained by Watson (1995). The four cases in

Natal that exhibited good performance based on their SPI

scores were not extraordinary. Many condominial sewer

projects in Natal performed well.25

Two characteristics of the strategies followed by the sani-

tation agency in Natal are particularly notable in the context

294 Nance & Ortolano

of our study. First, the agency had an overall plan for extend-

ing the trunk sewer network into Natal’s various sewer basins.

Second, the agency had a standardized procedure for imple-

menting community participation. This is reflected in Figures

2-B and 2-C, where, except for case N4, which is a high-

income neighborhood, all the participation scores are nearly

the same across the cases. This contrasts with the high vari-

ability in scores for Recife.

Another aspect of the situation in Natal relevant to our

study is that for long periods during the 1980s and 90s, the

mayor of Natal, the governor of the state of Rio Grande do

Norte, and leaders of the state sanitation agency were aligned

in their support of condominial sewer technology. This

meant there were few disruptions caused by disputes between

the city and the state over the desirability of condominial sew-

ers and relatively few disruptions after new officials were

elected (Nance 2004, chap. 6). Consequently, the state sani-

tation agency was able to make progress in implementing its

plans for extending the trunk sewer network.

In contrast to Natal, most of the condominial sewer projects

in Recife performed poorly. The fact that the few well-

functioning condominial sewers in Recife were located in

neighborhoods having relatively high levels of influence on

political officials and agency staffs led us to examine how com-

munities sometimes influenced project outcomes by taking

actions outside of an agency’s formal participation program.

䉴Other Forms of Community Participation

Participation Outside the Context of Agency

Participation Programs

One of the advantages of our qualitative, small-sample

study is the ability to identify new variables that influence out-

comes based on a detailed understanding of a case. In prob-

ing more fully the influence of participation on outcomes, we

noted during our interviews that some communities had

exerted significant influence outside the context of the for-

mal participation programs run by sanitation agencies. Our

findings in this regard (presented below) suggest that our

conceptualization of participation as being linked to the four

stages of the project cycle was excessively narrow. A more

expansive view of participation would include ways in which a

community participated beyond its involvement in a partici-

pation program. Although we present measures of these

other forms of participation, the measures are simplistic and

preliminary because we did not set out to investigate this

issue in our original research design and field study.

One form of participation outside of the participation pro-

grams run by sanitation agencies involved residents exerting

influence by exploiting their relationships with influential

actors. Community residents with personal, business, political,

or familial relationships with public officials, agency staffs, or

politicians often have the potential to use these relationships to

exert significant influence. As is commonly known in Brazil,

“getting a job as a janitor in a public building, as a garbage col-

lector, a teacher . . . frequently depends on political con-

tacts.” (Mainwaring 1999, 182). The benefits of influence

created by personal relationships and other ties can spread to

an entire neighborhood when the ties are used for projects

that serve that neighborhood.

A second way in which communities can exert influence

outside of agency participation programs is based on their abil-

ity to organize. Communities that are well organized have the

potential to bargain effectively with elected and appointed offi-

cials, for example, by assembling blocs of votes for a candidate

promising urban services. Effective neighborhood organiza-

tions and grassroots popular movements became important in

the 1980s in Brazil (during and following the democratic tran-

sition), when neighborhoods first mobilized against the

authoritarian military regime and then organized to press for

citizenship, which included the right to receive basic urban ser-

vices (Hochstetler 2000; Mainwaring 1989). Residents can use

Table 6.

Estimates of the levels of community influence in the seven case studies.

Good Performance Poor Performance

Form of Influence R1 R3 N1 N2 N3 N4 R2

Organizationa++ −− −+−

Relationshipsb++ −−−+−

Source: Nance (2004).

a. Statements by residents were used to determine the existence of a neighborhood association and a representative community leader(s)

at the time the project was implemented. This characteristic was judged positive if a neighborhood association and leader were present.

b. Statements by residents were used to determine the presence or absence of relationships with officials. This characteristic was judged

positive if relationships were present.

Community Participation in Urban Sanitation 䉳295

the influence created by organizing to bring condominial sew-

ers and other urban services into their neighborhoods or to

improve the quality of service provided by public agencies.

Table 6 shows our estimates of the ability of each of the

seven communities involved in our study to exert influence by

taking advantage of connections to officials or by using their

ability to organize. As indicated in the notes to Table 6, we

relied on interview results in making the judgments reflected in

the table. In gauging organization in the table, we relied only

on whether a neighborhood association and a leader for the

association were present. Population size is linked to the ability

of a community to deliver votes in exchange for projects in the

sense that larger communities would potentially have more

votes to deliver. In the opinion of staffs at the sanitation agen-

cies, three neighborhoods in our study would be considered to

have relatively large populations: cases R3, N2, and N3.26 We

use this information on population size below in interpreting

how influence was exerted in some communities. Although the

socioeconomic characteristics of a neighborhood may have an

implicit bearing on whether residents have relationships with

elected and appointed officials, we relied on our interviews to

judge whether any such relationships were in fact exploited in

the context of the condominial sewer projects.

Community Influence and Project Performance

In presenting our results from examining two additional

forms of participation that we identified above as community

influence, we first consider the four case studies in Natal (see

Table 6). Three of these four projects were located in neigh-

borhoods where communities neither exerted influence via

their personal relationships nor by using their ability to orga-

nize and thereby pressure local officials. These types of influ-

ence were only present in case N4. However, there was no

significant variation in the performance of these four condo-

minial sewer systems (at ρ≤0.05). This suggests that there

was little need for a community to exert influence outside the

sanitation agency’s participation program to obtain a good

project in Natal. This is consistent with the state sanitation

agency’s systematic use of community participation. From the

perspective of investigating community influence, Recife’s sit-

uation is more interesting because there is more variation

among the cases and because the four sanitation agencies

operating in Recife utilized ad hoc approaches in imple-

menting condominial sewer projects.

Our analysis of the cases in Recife demonstrates that in cases

R1 and R3, residents took action beyond the sanitation agen-

cies’ community participation programs in an effort to obtain

better project outcomes. In case R1, which involved a middle-

income community, residents received relatively high levels of

maintenance service by the sewer agency.27 In interviews con-

ducted with residents in the project area, we learned that sev-

eral residents had political ties to a local official who had served

as mayor and governor; the residents also had ties to his politi-

cal party. In addition, some residents reported having connec-

tions with staff of COMPESA, the agency that implemented the

project. Moreover, some COMPESA staff lived in the case R1

project area. During fieldwork in the case R1 neighborhood, we

observed that this condominial sewer project received unusu-

ally consistent maintenance attention from COMPESA (com-

pared to other project areas in Recife). On several instances, we

observed COMPESA staff fixing problems inside homes, which

was highly unusual because residents were responsible for their

in-home plumbing, fixtures, and house connections. Forty-one

percent of residents interviewed for case R1 reported that

COMPESA fixed their household condominial sewer whenever

there was a problem. Although it is difficult to trace the causal

process, these outcomes for case R1 suggest the possibility that

the existence of these ties to government officials and COM-

PESA may have affected the delivery of maintenance services,

with consequent effects on project performance.

Case R3 demonstrates a clear linkage between condominial

sewers and community influence prior to project implementa-

tion. In this case, a large population of poor residents had ini-

tiated an occupation of vacant land (known locally as an

invasion, or invasao) for the purpose of establishing a new com-

munity with urban services. This occupation was coordinated

with the campaign to reelect a populist governor for the state

of Rio Grande do Norte. During this governor’s first term in

the early 1960s, there had been a large invasion of poor citi-

zens into a different neighborhood (Green Garden, a pseudo-

nym). The governor had provided Green Garden citizens with

legal tenure and infrastructure services. Residents of the case

R3 occupation created a neighborhood organization called the

Comissao de Ocupacao (Occupation Commission) to represent

their interests in negotiating for service similar to that received

by the community that had invaded the Green Garden area.

The commission’s tactics for exerting influence included

assembling the votes of the community and negotiating with

the state government for an urbanization project. The new gov-

ernor, who was elected in 1986 and took office in 1987,

included the case R3 area in his statewide favela (shantytown)

urbanization program, and this is the program under which

the case R3 condominial sewer was installed.

Residents we interviewed reported that politicians always

visited the case R3 area just before election time, and they con-

tinued to do so years after the urbanization project was fully

implemented. The Occupation Commission was disbanded in

1990 after the project was completed but, as of 1995, the com-

munity had not stopped exerting influence on officials. During

our interviews with the president of the case R3 neighborhood

association, he reported that he continued to meet every

month with officials of the state sewer agency to complain and

call attention to his community’s ongoing water and sewer

maintenance needs. We observed state sewer agency staff fix-

ing sewer problems in the case R3 area on several occasions

during our fieldwork.

In contrast with case R3, residents of the case R2 project

area exerted no influence. Indeed, they had no potential

influence to exert (see Table 6). Even though case R2 resi-

dents participated in construction and maintenance of the

condominial sewer system, their project was essentially

ignored by the sewer agencies. Project funding was insuffi-

cient, the project design was impractical, and project changes

made during construction were poorly controlled. In short,

problems with the case R2 condominial sewer were discov-

ered soon after the project was installed, and nothing was

done to correct them. Instead, the maintenance agency aban-

doned the project. Although residents complained, they

lacked the ability to exert much influence because they had

no personal relationships to exploit. Moreover, even if the

community had tried to organize votes (as in the case of R3),

the population of the neighborhood was so small that this

might not have had much effect.

The two cases of success in Recife were exceptions to that

city’s overall record of poor performance and, therefore, these

cases offer insights into how good performance was achieved in

that city. In contrast to the situation in Natal, it appears that

communities in Recife had to organize effectively and use ties

to influential political actors to gain leverage on the agency that

delivered sewer services. In Recife, the two communities with

good condominial sewer performance obtained this good per-

formance as a result of exerting influence to gain leverage.28

䉴Discussion

How well does participation explain variations in the per-

formance of condominial sewer projects in Recife and Natal?

We found that this question cannot be addressed without dis-

tinguishing among several different forms of participation.

Different forms of participation within stages of the project

cycle do not appear to be equally associated with perfor-

mance, and the ability of a community to exert influence out-

side of an agency’s formal participation program is a form of

participation that was significant in Recife.

Consider the four forms of participation in the agencies’

community participation programs. Two cases with signifi-

cantly different outcomes, case N4 (good performance) and

case R2 (poor performance), each had comparably high lev-

els of participation in construction and maintenance work.

The remaining cases, all with good performance, achieved

mid-range levels of participation in construction and mainte-

nance. This finding does not support the view that increased

participation in construction and maintenance work is asso-

ciated with enhanced performance. Furthermore, cases N4

and R2 had significantly different levels of participation in

mobilizing and decision making. For these two forms of par-

ticipation, the case with good performance (N4) had mid-

range participation, and the case with poor performance

(R2) had very low participation.29 Mid-range levels of partici-

pation in mobilizing and decision making were similarly asso-

ciated with good performance for the remaining cases. This

finding led us to hypothesize that participation in mobilizing

and decision making is associated with enhanced condo-

minial sewer project performance.

Of course, these associations are only suggestive.

Generalized causal relationships between variables cannot be

established with the small sample size used in this study. The

above observation about the significance of mobilization and

decision making forms of participation is, however, consistent

with overall findings in the literature on demand-responsive

projects (e.g., Sara and Katz 1997). This consistency is based

on the view that a high level of involvement in mobilization

signals a strong demand for a project.

We also found that in some localities, the ability of commu-

nities to use their connections to agency staffs and political fig-

ures and their ability to organize (e.g., to demonstrate in large

numbers or to deliver votes as leverage) was at least as impor-

tant as engagement in participation programs run by sanita-

tion agencies. This was observed in Recife, which had a

relatively poor overall performance record with condominial

sewer projects. The two cases with uncommonly good perfor-

mance (cases R1 and R3) involved communities that exerted

influence, which contributed to good sewer service.

Conversely, the case with poor performance (case R2) involved

a community that lacked connections to officials or the ability

to organize effectively to press its case for good sewer services.

Our results raise questions about the validity of general

claims that more community participation will lead to better

project outcomes. From the perspective of affecting perfor-

mance of a condominial sewer project, when and how com-

munity participation occurs are significant. Arguing that

community participation must be viewed in a nuanced fash-

ion is, of course, nothing new. This was a position taken more

than three decades ago in a widely cited article by Arnstein

(1969), in which she presented a ladder of participation, a

framework that was later adapted by Paul (1987) for projects

in developing countries. The arguments herein reinforce

Arnstein’s position by demonstrating the relative importance

of different forms of community participation.

Our findings concerning differences between community

involvement in an agency-run program and participation that

exerts influence on political officials and agency staffs indicate

that our initial conceptualization of participation was too restric-

tive. When we began the study, we were concerned only with par-

ticipation as it occurred through programs designed by

agencies. At that point, we failed to give sufficient attention to

how communities, by exploiting various ties to influential actors

or by organizing effectively, could shape the performance of

sewer services.

In expanding our view of participation to include the way

communities utilize relationships to officials and agency staffs

296 Nance & Ortolano

Community Participation in Urban Sanitation 䉳297

and their ability to organize effectively, our explanations of

case study results became richer. This broadened view of par-

ticipation, together with an appreciation of the different

institutional settings in Recife and Natal, helps place the role

of participation in agency programs in context and provides

a more complete understanding of the role of community

participation in affecting project performance.

䉴Conclusions

Our results for Recife and Natal demonstrate that differ-

ences between the highest and lowest levels of condominial

sewer project performance can be small in one city and large

in another and that both project-level and city-level informa-

tion is needed to understand and explain these differences.30

Also, after distinguishing among different forms of community

participation, we find that not all forms of participation are

equally influential in delivering successful condominial sewer

service. In particular, for agency-organized participation based

on phases of the project cycle, we find that only two forms

of participation—mobilization and decision making—are

positively associated with project outcomes. This assertion is a

tentative hypothesis that can be subject to further investigation.

We also find that people’s involvement in agency-organized

forms of participation is not the only type of participation rel-

evant to explaining outcomes. Communities can affect out-

comes by exerting influence through their relationships with

community officials and by organizing to make a stronger case

for good service with agency staffs and elected officials. A com-

munity’s ability to organize effectively may also allow it to

assemble votes or to attract media attention through demon-

strating, thereby gaining leverage with officials. Our findings

suggest that participation in agency-run participation pro-

grams is not always a significant driver of success and that

forms of community participation external to the project can

have more significant effects on project outcomes.

The successful co-production of condominial sewers

requires an engaged sanitation agency as well as involved resi-

dents, and where an appropriate balance is not achieved, such

as in Recife, communities with potential to exert influence will

act to obtain better service. Explicit attention to alternative

forms of participation, including participation outside of

agency-run participation programs, will help planning practi-

tioners and development scholars better understand the range

of venues needed for communities, agencies, and officials to

move forward in creating successful sanitation projects.

In summary, our study provides the following: (1) a recon-

ceptualization of the relevance of participation as an aspect of

the co-production of urban infrastructure, (2) a technique for

measuring participation in the context of the project cycle, (3)

empirical evidence suggesting a limited impact of the con-

struction and maintenance forms of participation (i.e., partici-

pation vis-à-vis contributions) on sewer system performance,

(4) empirical evidence of a potentially significant impact of the

mobilization and decision-making forms of participation (i.e.,

participation vis-à-vis involvement in decision making) on sys-

tem performance, and (5) insights into how the organizational

and relational forms of participation (i.e., participation vis-à-vis

exertion of influence) can affect sewer system outcomes.31

Authors’ Note: The authors gratefully acknowledge the comments we

received on early drafts of the article by Jennifer Davis, Alnoor Ebrahim,

and Max Stephenson.This research was funded by the following grants:

Stanford School of Engineering John K. Vennard Civil Engineering

Scholarship, Stanford Institute for International Studies Dissertation

Writing Grant, Buck Foundation Environmental Fund Dissertation

Fieldwork Grant, and Stanford Institute for International Studies

Dissertation Travel Grant.

䉴Notes

1. For examples of the literature on the influence of participa-

tion, see Bryant and White (1982); Cabrera (1995); Finsterbusch

and Van Wicklin (1987); Isham, Narayan, and Pritchett (1995);

Korten (1980), Korten and Alfonso (1985), Narayan (1995),

Ostrom, Schroeder, and Wynne (1993), Paul (1987), Prokopy (2002;

2005), Schubeler (1993), World Bank (1994), Sara and Katz (1997),

and Wright (1997).

2. The terms co-production and participation (shorthand for

community participation) are used interchangeably in this article

and in the literature (Joshi and Moore 2004; Ostrom 1996;

Whitaker 1980; Brudney and England 1983; Rich 1981).

3. Sara and Katz (1997) provide an overview of the literature

on demand-responsive water and sanitation projects.

4. We were also guided by Ostrom’s (1996) analysis of co-

production in condominial sewers and Watson and Jagannathan’s

(1995) analysis of participation in the water and sanitation sector.

5. Differences in institutional inputs and approaches to

implementation beyond the neighborhood level are more fully

addressed in separate publications (see Nance 2004, 2006). In

these publications, Nance argues that project performance is pri-

marily a function of the strength and stability of a project coali-

tion composed of the relevant agencies, officials, and

beneficiaries associated with an individual project. Projects per-

form better in a context that supports the creation of strong, sta-

ble project coalitions. Decentralization of Brazil’s sanitation

sector during the 1980 to 1995 period tended to destabilize and

weaken the condominial sewer project coalitions that developed,

even as democratization resulted in increased participation,

enhanced community influence, and more innovation by imple-

menting agencies. Recife quickly decentralized its implementa-

tion of participatory sanitation projects, but its ad hoc approach

was associated with weak project coalitions that did not achieve

good performance in the majority of the projects that were

implemented during the time period. Natal, however, main-

tained a systematic, centralized implementation approach for a

longer period. Strong and stable project coalitions formed more

readily around this centralized participatory model, initially

resulting in a better overall performance record than Recife.

6. Although some recent observations are provided (see the

notes to Table 1), changes to the case study neighborhoods over

time have no bearing on the analysis presented in this article,

which covers the implementation periods of selected case study

projects and measures the performance of these projects approx-

imately five years (on average) after implementation.

7. Our study does not address condominial sewers in small-

to medium-sized cities, towns, or rural areas.

298 Nance & Ortolano

8. Examples of sources from this literature are Glaser and

Strauss (1967), Yin (1994), Ragin (2000), Patton (2002), and

George and Bennet (2005).

9. Based on the provisional judgments reflected in Table 1,

there is no obvious association between participation and perfor-

mance. A slight majority of projects had good participation in the

opinion of knowledgeable practitioners, but performance was

equally distributed among the categories. These trends in per-

ception appear to be similar for both cities based on the infor-

mation in Table 1.

10. Our reasoning has a resemblance to the “indirect method

of differences” elucidated by John Stuart Mill ([1843] 1967). The

difference is that reasoning used by Mill involved cases selected

in sequence. To see the linkage to Mill’s description, consider

that the “effect” is a good condominial sewer project and that the

“cause” is good community participation in an agency-run partic-

ipation program. Using the indirect method of difference, the

researcher first examines “instances of the effect to see if they all

agree in displaying the same antecedent condition . . .” The

researcher then examines “instances of the cause to see if they all

agree in displaying the effect . . .”

11. On this point, note the views of Glaser and Strauss (1967):

By maximizing or minimizing difference among comparative

groups, the sociologist can control the theoretical relevance of

his data collection (p. 55) . . . [T]he sociologist cannot cite the

number and types of groups from which he collected data until

[emphasis in original] the research is completed (p. 50).

12. Researchers of large-N quantitative studies believe that

sample selection on both outcome and explanatory variables

simultaneously can lead to incorrect inferences; however, this

assumes that samples are the same as cases, an assumption that is

not necessarily true in small-n qualitative studies such as ours. Yin

(1994) expressed this point well as follows: “Cases are not ‘sam-

pling units’ and should not be chosen for this reason. . . . The

cases should serve in a manner similar to multiple experiments,

with similar results . . . or contrasting results . . . predicted

explicitly at the outset of the investigation” (pp. 31, 51). In our

study, projects were more like experiments rather than normally

distributed homogeneous sampling units. Recife had many small

projects with a higher level of performance failure and inconsis-

tent participation, and Natal had fewer large projects with a lower

level of performance failure and more consistent participation.

Knowing this about our population of projects ruled out the pos-

sibility of meeting the assumptions of a quantitative sampling

logic. Consequently, we used qualitative case study logic to select

projects based on “explicit predictions” of participation and per-

formance. Further discussion about the rationale for selecting

cases in small-sample qualitative research designs is available in

Collier and Mahoney (1996) and Ragin (1997).

13. From the resulting sample sizes, the corresponding confi-

dence intervals ranged from ±10.5 percent to ±13.5 percent per

case study (at ρ≤0.10, where ρis the level of statistical signifi-

cance).

14. All of the interviews for this research were conducted (in

Portuguese) by the senior author and her Brazilian research

assistants.

15. For more information about this methodology for mea-

suring participation, see Nance (2004, app. D).

16. See Nance (2004: appendices E and F) for case-by-case

descriptions of how people were involved in the various forms of

participation.

17. Our study did not examine the possibility that participa-

tion in construction and maintenance work was associated with

other measures of performance such as cost recovery.

18. We began our study with the following hypothesis:

higher levels of participation are associated with improved performance as

measured by the condominial sewer performance index. This initial

hypothesis did not disaggregate participation into four forms.

19. Interview with informant 29, COMPESA operations and

maintenance coordinator; interviewed by Nance, December 6,

1994, Recife, transcript.

20. Our characterization is supported by the findings of a

World Bank-commissioned survey of condominial sewer experi-

ence conducted by Watson (1995). Her study investigated and

compared the condominial sewer experiences of seven Brazilian

cities: Brasília, Recife, Natal, Cuiabá, Joinville, Petrolina, and

Itapissuma. The sample of cities chosen for the investigation

included two-thirds of all known condominial sewer connections at

that time (52,550 out of 75,000 total connections). The seven cities

represented sixty neighborhoods; however, cities were the unit of

analysis, so there was no evaluation of individual neighborhoods.

In the study, the overall record of condominial sewer performance

in Recife was characterized as “less than satisfactory,” based on a

three-category scale of good, average, and less than satisfactory.

Watson found that some projects were operating well even in the

badly performing cities of Recife and Cuiabá, and that all cities had

some problems.

21. This result is based on the average response of eight

agency staff members to the question, “How would you evaluate

the overall performance of condominial sewers in Recife?”

Possible answers were based on the following 5-point scale: 1-

Poor, 2-Fair, 3-Good, 4-Very Good, 5-Excellent. The interviews

were conducted in Recife in March 1995.

22. Interview with informant 100, president of an engineering

company; interviewed by Nance, March 19, 1995, Recife, transcript.

23. Interview with informant 1, engineering consultant and for-

mer city official (Recife) and former state official (Pernambuco);

interviewed by Nance, December 1, 1994, Recife, transcript.

24. This result is based on the average response of five agency

staff members to the question, “How would you evaluate the over-

all performance of condominial sewers in Natal?” Possible

answers were based on the following 5-point scale: 1-Poor, 2-Fair,

3-Good, 4-Very Good, 5-Excellent. The interviews were con-

ducted in Natal in 1995.

25. For additional evidence and more in-depth discussion

about the differences in condominial sewer implementation in

Recife and Natal during this time period, see Nance (2004).

26. The size of the project population was estimated as the

number of households in the project area times the average num-

ber of people per household. We left it to local condominial

sewer practitioners to gauge whether a particular population was

considered large.

27. In case R1, the existence of good topography, supporting

infrastructure (e.g., piped water, pavement, and drainage infra-

structure), and lower density (e.g., fewer houses per block and

fewer residents per house) also contributed positively to the func-

tioning of the sewer system.

28. See Nance (2004) for additional evidence and detailed

explanations of why community influence was more important in

Recife than in Natal for the condominial sewer sector.

29. At first glance, one might imagine that the residents of

case R2 predicted the future failure of their project or antici-

pated its abandonment by the sewer agency and therefore

refused to participate early on in project mobilizing and decision

making. But interviews with residents and with the design engi-

neer show that this is not the case and suggest that residents

would have participated if given the opportunity. Most residents

reported that there were no opportunities for involvement in the

project. The engineer reported that his original scope of work

did not include resident involvement or work with the commu-

nity and did not include a budget for hiring participation staff. In

the field, residents were only spoken to in a limited way every now

Community Participation in Urban Sanitation 䉳299

and then regarding the location and layout of the system, which

was being implemented in their neighborhood without their

involvement by a city agency that did not embrace participation

in its implementation approach. At the time, a nonparticipatory

implementation approach was the norm. After project operation

began, residents mobilized over an extended period of time to

complain to city and state agencies about poor system perfor-

mance, including attending community meetings, making verbal

complaints, making written complaints, and collecting signa-

tures. Both residents and city staff expressed surprise and outrage

that the state agency responsible for maintenance had aban-

doned the project.

30. The study on which this article is based covered the period

1980 to 1995.

31. These results contribute to the emerging literature on the

limits of participation (e.g., Khwaja 2002, 2004; Cooke and

Kothari 2001)

䉴References

Arnstein, Sherry R. 1969. A ladder of citizen participation. Journal

of the American Institute of Planners 35 (4): 216-24.

Bakalian, Alexander, and N. Vijay Jagannathan. 1991. Institutional

aspects of the condominial sewer system. In Water and Sanitation.

No. SW-6: World Bank Infrastructure Notes.

Brudney, J. L., and R. E. England. 1983. Toward a definition of the

co-production concept. Public Administration Review 43 (1): 59-65.

Bryant, Coralie, and Louise G. White. 1982. Managing development

in the Third World. Boulder, CO: Westview.

Cabrera, Ricardo. 1995. Review of “Rural water and sanitation

project” evaluations in Latin America. Report to CARE.

Collier, David, and James Mahoney. 1996. Insights and pitfalls:

Selection bias in qualitative research. World Politics 49 (1): 56-91.

Cooke, Bill, and Uma Kothari, eds. 2001. Participation: The new

tyranny? London and New York: Zed Books.

Finsterbusch, Kurt, and Warren A. Van Wicklin. 1987. The con-

tribution of beneficiary participation to development project

effectiveness. Public Administration and Development 7:1-23.

George, A., and A. Bennett. 2005. Case studies and theory develop-

ment in the social sciences. Cambridge, MA: MIT Press.

Glaser, Barney G., and Anselm L. Strauss. 1967. The discovery of

grounded theory: Strategies for qualitative research. Chicago: Aldine.

Hochstetler, Kathryn. 2000. Democratizing pressures from

below? Social movements in the new Brazilian democracy. In

Democratic Brazil: Actors, institutions, and processes, edited by

Peter R. Kingstone and Timothy J. Power, 162-82. Pittsburgh,

PA: University of Pittsburgh Press.

Isham, Jonathan, and Satu Kahkonen. 1999. What determines

the effectiveness of community-based water projects?

Evidence from Central Java, Indonesia on demand, respon-

siveness, service rules, and social capital. Social capital initiative

working paper no. 14. Washington, DC: World Bank.

Isham, Jonathan, Deepa Narayan, and Lant Pritchett. 1995. Does

participation improve performance? Establishing causality with

subjective data. World Bank Economic Review 9 (2): 175-200.

Joshi, Anuradha, and Mick Moore. 2004. Institutionalized co-

production: Unorthodox public service delivery in challeng-

ing environments. Journal of Development Studies 40 (4): 31-49.

Khwaja, Asim I. 2002. Can good projects succeed in bad commu-

nities? Collective action in public good provision. John F.

Kennedy School of Government Faculty Research Working Papers,

Cambridge, MA: Harvard University.

———. 2004. Is increasing community participation always a good

thing? Journal of the European Economic Association 2 (2-3): 427-36.

Korten, David C. 1980. Community organization and rural devel-

opment: A learning process approach. Public Administration

Review 40:480-511.

Korten, David C., and Felipe B. Alfonso. 1985. Bureaucracy and the

poor: Closing the gap. West Hartford: Kumarian.

Levine, C. H. 1984. Citizenship and service delivery: The promise

of co-production. Public Administration Review 44:178-87.

Mainwaring, Scott. 1989. Grassroots popular movements and the

struggle for democracy: Nova Iguacu. In Democratizing Brazil:

Problems of transition and consolidation, edited by Alfred Stepan,

168-204. New York and Oxford: Oxford University Press.

———. 1999. Rethinking party systems in the third wave of

democratization: The case of Brazil. Stanford, CA: Stanford

University Press.

Mara, Duncan. 1996. Low-cost urban sanitation, Chichester, UK:

Wiley.

Menendez, Aurelio. 1991. Access to basic infrastructure by the

urban poor. Economic development institute policy seminar report

no. 28. Washington, DC: World Bank.

Mill, John Stuart. [1843] 1967. A system of logic: Ratiocinative and

inductive. Toronto, ON: University of Toronto Press.

Nance, Earthea B. 2004. Putting participation in context: An eval-

uation of urban sanitation in Brazil. PhD diss., Stanford

University, California.

———. 2005. Multistakeholder evaluation of condominial sewer

services. American Journal of Evaluation 26 (4): 480-500.

———. 2006. Brazil’s transition to participatory sanitation, 1980-

1995. Paper presented at the annual meeting of the Latin

American Studies Association, San Juan, Puerto Rico.

Narayan, Deepa. 1995. The contribution of people’s participa-

tion: Evidence from 121 rural water supply projects.

Environmentally sustainable development occasional paper series

no. 1. Washington, DC: World Bank.

Ostrom, Elinor. 1992. Crafting institutions for self-governing irri-

gation systems. San Francisco, CA: ICS Press.

———. 1996. Crossing the great divide: Co-production, synergy,

and development. World Development 24 (6):1073-88

Ostrom, Elinor, Roger B. Parks, and Gordon Whitaker. 1978.

Patterns of metropolitan policing. Cambridge, MA: Ballinger.

Ostrom, Elinor, Larry Schroeder, and Susan Wynne. 1993.

Institutional incentives and sustainable development: Infrastructure

policies in perspective. Boulder, CO: Westview.

Parks, Roger, Paula Baker, Larry Kiser, Ronald Oakerson, Elinor

Ostrom, Vincent Ostrom, Stephen Percy, Martha Vandivort,

Gordon Whitaker, and Rick Wilson. 1982. Co-production of

public services. In Analyzing urban service distributions, edited

by Richard C. Rich, 185-99. Lexington, MA: Lexington Books.

Patton, Michael Quinn. 2002. Qualitative research and evaluation

methods. Thousand Oaks, CA: Sage.

Paul, Samuel. 1987. Community participation in development

projects: The World Bank experience. In World bank discussion

paper no. 6. Washington, DC: World Bank.

Percy, S. L. 1984. Citizen participation in the co-production of

urban services. Urban Affairs Review 19 (4): 431-46.

Prokopy, Linda S. 2002. The relationship between participation

and project outcomes: A study of rural drinking water

projects in India. PhD diss., University of North Carolina,

Chapel Hill.

———. 2005. The relationship between participation and project

outcomes: Evidence from rural water supply projects in India.

World Development 33 (11): 1801-19.

Ragin, Charles C. 1997. Turning the tables: How case-oriented

research challenges variable-oriented research. Comparative

Social Research 16:27-42.

———. 2000. Fuzzy-set social science. Chicago: University of

Chicago Press.

300 Nance & Ortolano

Rich, R. C. 1981. Municipal service and the interaction of the vol-

untary and governmental sectors. Administration and Society 13

(1): 59-76.

Sara, Jennifer, and Travis Katz. 1997. Making rural water supply sus-

tainable: Report on the impact of project rules. UNDP-World Bank

Water and Sanitation Program.

Schubeler, Peter. 1993, March. Participative strategies for improv-

ing urban infrastructure and services in developing countries.

Draft. In Workshop for Architecture and Planning, Ltd., UNCHS

Habitat. Nairobi, Kenya.

Sinnatamby, Gehan S. 1983. Low cost sanitation systems for

urban peripheral areas in northeast Brazil. PhD dissertation,

University of Leeds, UK.

Sundeen, R. A. 1985. Co-production and communities: Implica-

tions for local administrators. Administration and Society 16 (4):

387-402.

Wade, R. 1988. The management of irrigation systems: How to

evoke trust and avoid the prisoners’ dilemma. World

Development 16 (4): 489-500.

Watson, Gabrielle. 1995. Good sewers cheap? Agency-customer

interactions in low-cost urban sanitation in Brazil. World Bank

Water and Sanitation Division. Washington, DC: World Bank.

Watson, Gabrielle, and N. Vijay Jagannathan. 1995. Participation

in water and sanitation. Environment Department papers:

Participation series, paper no. ENV-002. Washington, DC:

World Bank.

Whitaker, Gordon P., 1980. Coproduction: Citizen participation

in service delivery. Public Administration Review 40 (3): 240-46.

World Bank. 1994. World development report 1994: Infrastructure for

development. New York: Oxford University Press.

———. 2004. World development report 2004: Making services work for

poor people. New York: Oxford University Press.

Wright, Albert M. 1997, November. Toward a strategic sanitation

approach: Improving the sustainability of urban sanitation in

developing countries. In UNDP-World Bank Water and Sanitation

Program (internal working document). Washington, DC: World Bank.

Yin, Robert K. 1994. Case Study Research: Design and Methods.

Thousand Oaks, CA: Sage.

THESIS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY Unpacking Sanitation Planning Comparing Theory and Practice

Book

Full-text available

Dec 2010

Jennifer R Mcconville

Lack of proper sanitation is linked to significant negative impacts on environmental and public health, economy, and human dignity. Despite the efforts made to reach the United Nations Millennium Development Goals, the world will miss its target of halving the percentage of people without access to improved sanitation by 2015; and there is general agreement within the field of sanitation that the sector has failed to deliver substantial improvements to the most needy. This global challenge of providing sanitation services to the underserved highlights the need to critically evaluate and change the way in which sanitation planning and service provision is approached. The overall objective of this thesis is to better understand the planning processes used in the field of sanitation and their importance for the sustainability of sanitation efforts. To achieve this, it attempts to bridge the professions of urban planners and sanitation engineers. Specifically it explores how sanitation planning processes are structured, to what extent participation plays a role in sanitation planning, and to what extent different perspectives of criteria for sustainable sanitation appear in the process. In order to unpack the planning process into these different elements this thesis develops an analytical framework (the SanPlan Scan) based on a mixture of theory and practice from planning and sanitary engineering. The performance of this framework is subsequently tested for its ability to identify interesting trends in participation levels, procedural planning modes, and criteria for sustainability in a number of case studies from sanitation projects in West Africa and popular sanitation planning guidelines. The resulting analyses identify critical differences between sanitation planning guidelines and practice in the field. For example, the guidelines consistently recommend more collaborative and participatory planning styles, especially including users, than was seen in the cases studied. The results also show that the process of designing sanitation options, and to some degree the selection process, remained dominated by expert-led planning styles, despite the abundance of rhetoric regarding the need for participation. The main conclusion that can be drawn from the multiple studies within this thesis is that more attention is needed to how the planning process itself is designed and conducted. This thesis supports the development of systematically adapted sanitation planning processes, by providing a starting point for discussing and understanding the practice of sanitation planning.

Simplified sewerage to prevent urban leptospirosis transmission: a cluster non- randomised controlled trial protocol in disadvantaged urban communities of Salvador, Brazil

Article

Full-text available

May 2023

Introduction: Leptospirosis is a globally distributed zoonotic and environmentally mediated disease that has emerged as a major health problem in urban slums in developing countries. Its aetiological agent is bacteria of the genus Leptospira, which are mainly spread in the urine of infected rodents, especially in an environment where adequate sanitation facilities are lacking, and it is known that open sewers are key transmission sources of the disease. Therefore, we aim to evaluate the effectiveness of a simplified sewerage intervention in reducing the risk of exposure to contaminated environments and Leptospira infection and to characterise the transmission mechanisms involved. Methods and analysis: This matched quasi-experimental study design using non-randomised intervention and control clusters was designed to assess the effectiveness of an urban simplified sewerage intervention in the low-income communities of Salvador, Brazil. The intervention consists of household-level piped sewerage connections and community engagement and public involvement activities. A cohort of 1400 adult participants will be recruited and grouped into eight clusters consisting of four matched intervention-control pairs with approximately 175 individuals in each cluster in baseline. The primary outcome is the seroincidence of Leptospira infection assessed through five serological measurements: one preintervention (baseline) and four postintervention. As a secondary outcome, we will assess Leptospira load in soil, before and after the intervention. We will also assess Leptospira exposures before and after the intervention, through transmission modelling, accounting for residents' movement, contact with flooding, contaminated soil and water, and rat infestation, to examine whether and how routes of exposure for Leptospira change following the introduction of sanitation. Ethics and dissemination: This study protocol has been reviewed and approved by the ethics boards at the Federal University of Bahia and the Brazilian National Research Ethics Committee. Results will be disseminated through peer-reviewed publications and presentations to implementers, researchers and participating communities. Trial registration number: Brazilian Clinical Trials Registry (RBR-8cjjpgm).

Smart Engagement in Small Cities: Exploring Minority Participation in Planning

Article

Full-text available

Apr 2023

Smart engagement approaches are now widely applied in community planning processes. However, there continues to be a lack of representation from marginalized groups such as racial/ethnic minorities in planning processes. In this study, we explore what smart community engagement methods are being applied by small cities in the U.S., and how minority communities are participating in the planning process with those engagement methods. We analyzed planning documents and public engagement data from five small cities located in different regions of the U.S. with varying levels of minority populations. We evaluated the planning processes of the study cities, specifically comprehensive planning, and what smart community engagement tools they have applied. Our study shows that smart engagements are performed primarily through community surveys and online outreach initiatives. Despite adopting these approaches, most cities received lower participation from minority populations compared to non-Hispanic Whites. Cities with higher participation rates provided more engagement opportunities and conducted targeted community events and surveys to reach out to minority and low-income communities. From this study, we conclude that cities should apply varied methods for community engagement and should not rely solely on smart approaches to engage with minority communities. For cities to increase their overall civic participation, including those underrepresented, smart engagement approaches should be supported by targeted public events and outreach activities.

Institutional change for the development of urban sanitation in the Global South: A social science review

Article

Jun 2024
Util Pol

Defining Co-Production: A Review of the Planning Literature

Article

Full-text available

Dec 2023
J PLAN LIT

Co-production is a concept that is becoming increasingly popular across various fields including planning. This article reviews planning literature on co-production and reveals that the term has not been well defined. The existing definitions are inconsistent and ambiguous, requiring more conceptual clarity to avoid contention. Based on the systematic literature review, and aided by bibliometric analysis, the article identifies seven dimensions within the current definitions of co-production: (1) actor, (2) reason, (3) input, (4) output, (5) phase, (6) means, and (7) context. This article concludes by proposing a conceptual and analytical framework for defining co-production in planning theory and practice.

Sanitation challenges in Dar es Salaam: the potential of Simplified Sewerage Systems

Article

Full-text available

Feb 2023

In the context of growing urbanization, sanitation in many cities isin acute crisis with severe social and environmental consequences. The Sustainable Development Goal (SDG) of sanitation for all by 2030 is increasingly elusive. Municipalities have been experimenting with a range of lower-cost sanitation solutions. Simplified Sewerage Systems (SSS) have emerged in different cities asone response, but with mixed results. This paper evaluates an SSS project in an informal settlement in Dar es Salaam, Tanzania. Drawing on a combination of a survey and focus group discussions, the paper examines the social and economic impacts of the SSS and identifies a set of key concerns for future urban sanitation interventions: affordability, maintenance and governance. We conclude by considering the implications for future research and practice on urban sanitation, including the limits of technology-based approaches and the necessity to focus on the diverse needs of residents in place.

K.A.P.S.: A Disaster Training Approach for High-Risk Communities

Article

Jan 2023

Communities along the US Gulf Coast are at high risk of natural and human-caused hazards. We developed a disaster training designed to increase Knowledge, Attitude, Preparedness, and Skills (K.A.P.S). We held a series of six identical training sessions in Geismar, Louisiana, a community that faces multiple hazards. Residents (n = 34) were trained using a community-tailored approach that combined constructivist (hands-on) and traditional (lecture) methods of instruction. Pre-test and post-test surveys demonstrated that the instructional content was effective (p < .01), and that individual preparedness knowledge increased significantly because of the constructivist teaching approach (p < .05). The results indicate that this high-hazard setting called for more extensive instructional content, constructivist teaching methods, and the inclusion of residents at all education levels.

Addressing Citizen-Participation for Sustainable Water & Sanitation Services in Kenya's Low-Income Areas

Article

Jan 2022

Inadequate water utility services in the low-income areas (LIAs) of Kenya remains the greatest water sector challenge for the coming decades, especially with a fast-growing low-income population. Land tenure issues, vandalism, unlicensed water service providers, irregular payment of fees, and poor urban planning are cited as the major limitations for the utilities. Citizen participation in the overall water governance framework is, however, also limited. Through a literature review approach, this article explores the inherent gaps in citizen participation in water governance in the LIAs and recommends measures for its systematic integration for enhanced water and sanitation access in Kenya’s LIAs.

Getting By and Getting Out: How Residents of Louisiana’s Frontline Communities Are Adapting to Environmental Change

Article

Jun 2021

Scholars argue that U.S. programs and policy designed to help households adapt to or move away from environmental risk were not designed to address climate change. Others demonstrate that disaster response upholds and produces structural inequality. This article examines how existing mitigation and adaptation policies fail to respond to lived conditions of residents and communities on the front lines of environmental change and perpetuate inequality. Based on interviews with residents in the lower bayou communities of Terrebonne Parish, Louisiana, and professionals working in the study area, we identified three factors that influence the outcomes of mitigation and relocation initiatives. First, we found that adaptation is a dynamic, ongoing process which can lead to the need for multiple types of assistance for a given property or household over time. Second, program timing and how residents make decisions about whether and how to rebuild or relocate are misaligned. Third, current programs deny resources to frontline communities by creating participation barriers for low- and moderate-income households. The findings affirm the need for more flexible policy guidelines if assistance programs are to transform communities in ways that respond to resident priorities and the realities of environmental change.

Activation of Co-Production: General Context and Antecedents

Chapter

Dec 2020

This chapter focuses on the first two components of the framework proposed in the book: the general context and the antecedents of co-production. While the general context includes the characteristics of the environment in which co-production takes place, the antecedents of co-production consist of those aspects that are necessary, according to the main co-production research, for co-production to happen. The antecedents to co-production are categorised by considering the extent to which they are related to lay actors, the regular service provider, and the nature of the co-produced service.

Rethinking Party Systems in the Third Wave of Democratization: The Case of Brazil

Article

Full-text available

Jan 1999
FOREIGN AFF

Among the many countries that underwent transitions to democracy in recent decades, only Russia is as important to the United States and the world as Brazil. The fifth-largest country and population in the world, with nearly one-half the inhabitants of Latin America, Brazil has the worldâs ninth-richest economy. Given the nationâs size and influence, its capacity to achieve stable democracy and economic growth will have global impact. Understanding democracy in Brazil is therefore a crucial task, one which this book undertakes. Theoretically, the author argues that most party systems in the third wave of democratization, after 1974, have distinctive features that require us to reformulate theories about party systems generally; previous works have paid scant attention to the importance of variance in the degree of institutionalization of party systems. The author also argues that many third-wave cases underscore the need to focus on the capacity of the state and political elites to structure and restructure party systems from below. Empirically, the author studies the Brazilian party system and democratization, with particular reference to the 1979-96 period. He underscores the weakness of the party system and the resulting problems of democratization. He argues that the party system is poorly institutionalized, explores the reasons for the difficulties of party building, and addresses the consequences of weak institutionalization, which leads him to reaffirm the central significance of parties in the face of widespread skepticism about their importance.

Democratizing Pressures from Below?: Social Movements in the New Brazilian Democracy

Chapter

Feb 2000

Kathryn Hochstetler

Turning the Tables: How Case-Oriented Research Challenges Variable-Oriented Research

Article

Jan 1997
Comp Soc Res

C.C. Ragin

Toward a Definition of the Coproduction Concept

Article

Jan 1983

This article attempts to define "coproduction" in a manner useful to policy makers and to academics concerned with measuring the concept. Coproduction is considered the critical mix of activities that service agents and citizens contribute to the provision of public services. The involvement of the former consists of their work as professionals, or "regular producers," in the service process. Citizen coproductive activities, or "consumer production," are voluntary efforts of individuals or groups to enhance the quality and/or quantity of services they receive. Based on this definition, three types of coproduction are distinguished according to the nature of the benefits achieved: individual, group, and collective.

Citizenship and Service Delivery: The Promise of Coproduction

Article

Mar 1984

Bureaucracy and the Poor: Closing the Gap.

Article

Mar 1982

Patterns of Metropolitan Policing

Article

Jun 1979

Community Participation in Development Projects: The World Bank Experience. World Bank, Washington, D.C

Article

Jan 1987

Samuel Paul

Citizen Participation in the Co-Production of Urban Services

Article

Jun 1984

Stephen L. Percy

Citizen coproduction is the productive involvement urban residents can supply to the provision of city services. This article introduces the symposium on citizen coproduction of urban services by tracing the concepts and arguments in the coproduction literature. Definitions of coproduction are reviewed, along with propositions that relate citizen coproduction to outcomes such as effectiveness, efficiency, and responsiveness. Also considered are impediments to the implementation of coproduction programs related to urban service delivery. This article also introduces the other articles in this symposium.

Coproduction and Communities Implications for Local Administrators

Article

Feb 1985

Richard A Sundeen

While coproduction has been the topic of a growing number of scholarly articles, this literature has neglected the relationship between coproduction and community. The concept of coproduction is rooted in the older and broader theoretical framework of community; because community contexts differ, so does the local capacity for coproduction. A typology of "alternative auspices" under which various local functions are carried out suggests that coproduction may be performed by individuals, informal groups, and formal nonprofit organizations. There are implications in this for the local administrator's three general role sets: extra community, local community, and intra-agency roles. Particularly, there may be different local roles-facilitator, broker, and organizer-to be played, depending upon the community's capacity for coproduction.

Community Participation in Urban Sanitation: Experiences in Northeastern Brazil

Abstract and Figures

Recommended publications

Brazil's health councils: The challenge of building participatory political institutions

Improving Municipal Urban Design Decision-Making by Using Digital Tools

The making of environmental marginality: Sanitation and risk perception in a Brazilian neighborhood.

As redes técnicas de saneamento