Understanding Factors That Influence Stakeholder Trust of Natural Resource Science and Institutions

Understanding Factors That Influence Stakeholder Trustof Natural Resource Science and Institutions

Steven Gray • Rachael Shwom • Rebecca Jordan

Received: 12 August 2010 / Accepted: 7 December 2011 / Published online: 1 January 2012

� Springer Science+Business Media, LLC 2011

Abstract Building trust between resource users and natu-

ral resource institutions is essential when creating conser-

vation policies that rely on stakeholders to be effective. Trust

can enable the public and agencies to engage in cooperative

behaviors toward shared goals and address shared problems.

Despite the increasing attention that trust has received

recently in the environmental management literature, the

influence that individual cognitive and behavioral factors

may play in influencing levels of trust in resource manage-

ment institutions, and their associated scientific assessments,

remains unclear. This paper uses the case of fisheries man-

agement in the northeast to explore the relationships between

an individual’s knowledge of the resource, perceptions of

resource health, and participatory experience on levels of

trust. Using survey data collected from 244 avid recreational

anglers in the Northeast U.S., we test these relationships

using structural equation modeling. Results indicate that

participation in fisheries management is associated with

increased trust across all aspects of fisheries management. In

addition, higher ratings of resource health by anglers are

associated with higher levels of trust of state and regional

institutions, but not federal institutions or scientific methods.

Keywords Cooperation � Fisheries management �Recreational anglers � Scientific assessment � Structural

equation modeling � Trust

Introduction

The benefits of developing and maintaining trust between

natural resource management agencies and natural resource

users have been well documented (Earle and Cvetkovich

1995; Beierle and Cayford 2002; Davenport and others

2007). Fostering trust between managers and those man-

aged has been shown to decrease public perception of risk

(Eiser and others 2007; Needham and Vaske 2008),

increase the likelihood of compliance behaviors (Dickson

and others 2009) increase coordination across diverse

stakeholders (Owen and Virderas 2008) and increase the

overall resilience of coupled social-ecological systems

(Ostrom and others 2008). These positive outcomes are

attributable, in part, to the ability of trust to minimize

social disorder by facilitating cooperation among individ-

uals and groups (Barber 1983).

Past research has highlighted the importance of learning

which enables trust to be built as individuals gather

information about the motives, objectives and behaviors of

others over time (Beratan 2007). Ultimately, the collection

of this information results in the acceptance or rejection of

another’s goals which then determines whether cooperation

is an appropriate response. Trust has also been framed in

terms of mental assessment of costs and benefits of coop-

eration. According to Rousseau and others (1998) trust is

best defined as a willingness to accept vulnerability based

upon positive expectations of the intentions or behaviors of

others. In these terms, individuals are rational actors and

the trust outcomes are carefully considered to determine

S. Gray (&)

Department of Natural Resources and Environmental

Management, University of Hawaii, Manoa, Honolulu,

HI 96822, USA

e-mail: [email protected]

R. Shwom

Department of Human Ecology, Rutgers University, Cook Office

Building, 55 Dudley Road, New Brunswick, NJ 08901, USA

R. Jordan

Department of Ecology, Evolution, and Natural Resources,

Rutgers University, 14 College Farm Road, New Brunswick,

NJ 08901, USA

123

Environmental Management (2012) 49:663–674

DOI 10.1007/s00267-011-9800-7

whether the benefits of cooperation outweigh the potential

costs to vulnerability. These definitions emphasize rational

responses and the role that knowledge, internalizing

information and learning play in influencing trust, which

may result in the decision to work together toward common

goals or to seek out an individually beneficial plan of

action.

However, studies have shown that not all trust judg-

ments are rational and learned appraisals. Rather, in the

absence of information about the motives and objectives of

others, emotions and heuristics seem to fill the knowledge

gap. In an experimental setting, Schwarz and Clore (1983)

found that people use their momentary affective states as

information. Building on this work Dunn and Schweitzer

(2005) determined that these ‘‘affect-as-information’’ heu-

ristics and familiarity with the trustee have a significant

influence on trust judgments. When the trustee is well

known, information is directly processed and evidence or

associations influence the trust decision. However, when

there is little information about the trustee, affect-as-

information and emotional associations may determine the

level of trust afforded to a trustee.

In this paper we investigated the factors that influence

trust in a natural resource management context using rec-

reational marine fisheries as a model. Specifically, we

designed and fielded a survey that sought to understand

how levels of anglers trust was related to familiarity with

management structures and science concepts, perceptions

of resource health, and whether or not anglers participated

in fisheries management. We hypothesized that these

variables would influence anglers’ level of reported trust in

fisheries management institutions and scientific assessment

methods because they are direct measures of self-reported

familiarity with processes and concepts which outline the

intentions of management- or are experiences (e.g., par-

ticipation) where the intentions of institutions are expected

to become better understood. To analyze the factors

influencing levels of trust quantitatively, we used structural

equation modeling (SEM), a type of multivariate analysis

used to determine strength of associations between pre-

dictor and response variables.

Trust in Natural Resource Management

Trust between individuals, groups, and an institution is

central to the effective management of shared natural

resources. In both experimental and field research, trust

appears to be a core variable to explain why participants

across settings tend to cooperate or not (Ostrom and

Walker 2003). This cooperation is especially important in

common-pool resource management because profit maxi-

mizing by harvesters has the capacity to threaten the sus-

tainability of a resource (Ostrom 2007). On short time

scales, harvesting restraint does not benefit the individual

and trust and cooperation are counter-intuitive responses

when the alternative response is to maximize returns

(McAllister and others 2005). Yet, considerable evidence

from past studies have indicated that trust and cooperation

are an appropriate response when the likelihood of

encountering the trustee again is high (Barclay 2004) and

the benefits of cooperation are clearly communicated

(Ostrom and Walker 2003). The mechanism for building

trust in a natural resource context is most often discussed

in terms of fostering shared values between manage-

ment agencies and stakeholder groups (Johnson 1999;

Cvetkovich and Winter 2003; Needham and Vaske 2008).

Aligning values increases cooperation by increasing the

perceived likelihood of accruing benefits. Shared values are

made explicit by reciprocity norms between individuals,

groups and institutions which are socially reinforced

(Ostrom and Walker 2003). These experiences and inter-

actions result in understanding the values of others which

then leads to trust judgments. As Johnson (1999) points

out, however, evidence of shared values is not necessarily

easily identified. Further, values are often complex and

multi-dimensional and are likely to vary across and within

stakeholder groups (Hoppner 2009).

Investigations of trust in a natural resource context have

overwhelmingly examined case studies between local

communities and government agencies. For example, in a

study investigating water-resource decision-making, Leahy

and Anderson (2008) qualitatively described the impor-

tance of trust between local community members and the

U.S. Army Corps of Engineers. In a review of three case

studies from river systems across the United States, Carroll

and Hendrix (1992) outlined trust’s contribution to col-

laborative management when local and affected citizens

were involved. Cvetkovich and Winter (2003) studied trust

between a range of local forest stakeholders and the U.S.

Forest Service management for the protection of endan-

gered species. All studies found that similar values between

local communities and agencies influenced level of trust in

management decisions and describe frameworks for agen-

cies to engage stakeholders through collaboration.

Research has also uncovered impediments to trust

building with local communities. In an investigation of

U.S. Forest Service and stakeholder groups, Davenport

(2007) found that low levels of community engagement,

unclear communication, and a history of adverse relation-

ships between communities and agencies can constrain

collaboration between communities and agencies. This

constraint can make it difficult, if not impossible, to

understand the values of stakeholders and communicate the

goals of the agency. A history of distrust in the federal

government, an agency, or in the management of a par-

ticular public resource can create a negative atmosphere

664 Environmental Management (2012) 49:663–674

123

that is difficult to dispel (Lawrence and others 1997).

Additionally, it has been found that it is much easier to

destroy trust than for trust to be built because negative

events have much greater impact on self-reported trust than

do positive events (Slovic 1993)

Public Participation as a Mechanism for Trust-Building

Although the benefits and outcomes attributed to trust are

somewhat clear, promoting and maintaining trust between

institutions and stakeholders is not an easily accomplished

goal (Johnson 1999). The most common tool available to

most natural resource policy-makers to promote trust is the

inclusion of stakeholders in the decision-making process.

However, this inclusion does not necessarily result in trust.

Measurements for success of these public engagements

vary and general typologies of participation are not easily

developed (Chess and Purcell 1999).

The term ‘‘public participation’’ refers to a diverse set of

activities which engage the public and vary greatly in terms

of who is involved, how early and often in the process, and

who has influence in the final outcome [National Research

Council (NRC) 2008]. Over the years, a set of general

guidelines for public participation have emerged (Rowe

and Frewer 2000). These guidelines emphasize the

importance of early contact and information exchange

between both parties in the management or planning pro-

cess. Researchers have also drawn attention to the impor-

tance of increasing the perception of procedural justice

(Webler and Tuler 2000). For example, public perceptions

regarding the ease of participation and that their input is

valued can be improved by offering multiple opportunities

to participate and engaging the public audiences through a

variety methods. In a natural resource context, Smith and

McDonough (2001) outline six themes which emerged

from focus group conversations regarding a participatory

ecosystem management project and call attention to the

importance of representation, voice, consideration, logic,

and desired outcomes when including the public in eco-

logical decision-making.

Understanding Angler Trust in Recreational Fisheries

Management

This paper uses the case of recreational anglers and fish-

eries management in the northeast to explore the factors

that predict their levels of trust. Using survey data collected

from 244 avid anglers in the Northeast U.S., we tested the

relationships using SEM. Recreational marine fisheries in

the U.S. offer a good opportunity to investigate individual

factors which may influence trust in natural resource

institutions beyond the local community case study

approach because recreational anglers are a loosely-

connected and diverse group of stakeholders. In addition,

marine fisheries management in the U.S. is designed

partially around public participation. Outlined by the

Magnuson Stevens Act (1976), resource managers consult

with stakeholder groups throughout all phases of manage-

ment, from the initial scoping process to selecting harvest

regulations, as a way to align goals and for management to

integrate stakeholder knowledge into the management

process. Stakeholders are even invited to participate in

some aspects of scientific assessment (Kaplan and McCay

2004) which has been shown to increase industry buy-in of

management procedures (Johnson and van Densen 2007).

Further, members of the management bodies (regional

fisheries management councils) are comprised of appoin-

tees who reflect the stakeholder make-up of the fisheries

under management (e.g., commercial fishermen, recrea-

tional anglers, environmental NGO representatives). Fish-

eries management has been designed deliberately this way

and is expected to increase legitimacy in decision-making

and produce more effective regulations by increasing

cooperation in common property management (Wilson and

others 2003).

Despite an extensive and ongoing participatory system

in place, levels of trust between fishery stakeholders and

management are not well understood. This is especially

true of geographically dispersed recreational anglers, for

which there is little data to support both their ecological

and political impact in modern fisheries management

(Arlinghaus 2005). However, because trust judgments are

theorized to be influenced by the varying types of infor-

mation resource users have collected (Beratan 2007) there

are a range of reasonable hypotheses that could be tested.

In our study we predicted that there are three types of

angler knowledge or information that will influence levels

of trust in management institutions and science: (1)

familiarity with fishery science and the management pro-

cesses (2) perceptions of resource health and (3) prior

participation in fisheries management.

To investigate the role of these factors empirically, we

tested the following hypotheses:

H1 High self-ratings of knowledge of fisheries science

influence higher level of trust in state, regional, federal

institutions and recreational and commercial science

If the level and type of knowledge held by anglers is not

consistent with the level and type of knowledge important

to fisheries management (which are scientific concepts)

then a potential disconnect between these two ‘‘knowledge

systems’’ may increase resource conflict (Skogen and

Thrane 2008) and decrease trust. This potential disconnect

between the often quantitative scientific assessment of

resource health and users’ qualitative assessment of

resource health may account for different preferences for

Environmental Management (2012) 49:663–674 665

123

management. Schisms originating from the tension

between scientific knowledge and lay knowledge in pre-

vious studies have been shown to increase resource con-

flicts (Dobbs 2000; Skogen 2001).

H2 High self-ratings of knowledge of fisheries manage-

ment influence higher level of trust in state, regional, fed-

eral institutions and recreational and commercial science

Low levels of knowledge or familiarity with concepts

key to fisheries management may reflect an uncertainty in

the values and goals of management, increasing the diffi-

culty anglers have in making direct and adequate assess-

ments about whether environmental management policies

are designed with angler goals in mind. Thus, the amount

of familiarity and knowledge stakeholders may have about

an institution and its practices are likely to influence the

amount of trust in a decision-making institution (Dietz and

others 2007).

H3 High ratings of fishery resource health influence

higher levels of trust in state, regional, federal institutions

and recreational and commercial science

Evidence of institutional success or failures, whether

accurate or not, is expected to influence trust of natural

resource management institutions. Recreational anglers

spend a great deal of time interacting with nature and their

evaluation of the current resource conditions, or produc-

tivity of these areas, may influence whether they trust the

performance of management bodies. Evidence of ecologi-

cal health may also influence risk perception. In other

wildlife contexts, hunter stakeholder groups who perceived

less ecological degradation reported higher ratings of trust

of management institutions (Needham and Vaske 2008).

H4 Participation in fisheries management will influence

higher level trust in state, regional, federal institutions and

recreational and commercial science

Trust has been repeatedly highlighted as an indicator of

success criteria in participatory decision-making and

planning (Beierle and Konisky 2000; Chess 2000; National

Research Council (NRC) 2008). Participation in manage-

ment processes is thought to increase trust of management

institutions because it allows for a space where shared

values can become better understood and conflicting values

can be addressed as groups work to co-construct a mutually

beneficial plan of action (National Research Council

(NRC) 2008). Further, past research has pointed to the

various types of participation adopted by management

agencies (Arnstein 1969; National Research Council

(NRC) 2008) which range from a top-down hierarchy or

placating system to a fully democratic system. If carried

out correctly, participation is expected to have a strong

positive effect on trust.

Method

In this section we will first describe how we measured the

variables hypothesized to be significant predictors of trust

in our model. We then describe how the survey was

administered and the survey sample. Finally, we discuss

the methods used for analyzing our survey data.

Survey Instrument Development and Administration

A survey instrument was developed to test the relationships

hypothesized above regarding what factors were significant

predictors of levels of trust in fisheries management and

science. Prior to surveying at the saltwater expos, local

interviews were conducted and a pilot questionnaire was

developed and administered to a subsample of 34 recrea-

tional anglers at a recreational fishing club in New Jersey.

Questionnaire items were validated for face validity

through one-on-one informal interviews which took, on

average, about 20 min per interview. The final survey

instrument measured the following exogenous and endog-

enous variables used in the model (see Fig. 1):

Exogenous Variables: Trust in Fisheries Management

and Science

We divide the exogenous variables into five separated

components: three levels of fishery management institu-

tions (state, regional, and federal) and two components of

fisheries science (methods used for assessment of

FisheriesManagementKnowledge

ResourceRating

Participation

FisheryScience

Knowledge

NMFS

RecreationalScience

CommercialScience

State

Regional

Age Class

Sex

Habitat

Size

Catch

Inolved

FMPs

MSA

OverallResources

ParticipateFishery

Bluefish

Participate

Fig. 1 Hypothesized structural model. Rectangles indicate measured

variables and ellipses indicate latent variables


123

recreational stocks and pressure, and those used for com-

mercial stocks and pressure). Although these institutions

and scientific methods are related and rely on one another

to make decisions, they can be considered discrete entities

because management jurisdictions and decisions do vary

by state, region and federal levels. These levels of man-

agement include State Fisheries and Wildlife Departments,

the Regional Fisheries Management Councils, and

NOAA’s National marine fisheries service (NMFS).

Anglers were ask to rate their level of trust in each of these

five management components on a 5 point scale from fully

trust to fully distrust.

Endogenous Variables: Familiarity of Fisheries

Management and Science, Perceptions of Resource Health,

and Participation

To investigate subjective knowledge, anglers were asked to

rate their knowledge on key concepts important to fisheries

management and science on a scale from 1 (very knowl-

edgeable) to 5 (very un-knowledgeable). A subjective scale

was selected instead of an objective measure of knowledge

because past studies have indicated that personal beliefs,

which include self-rated confidence, are central to personal

trust judgments (Castelfranchi and Falcone 2000). To

assess knowledge about fishery science, anglers were asked

‘‘How would you rate your knowledge in the following

areas about the fish species you fish for:’’ in subject areas

fundamental to fisheries science including habitat needs,

determining age class, and determining sex of fish species.

To assess knowledge about civic literacy, anglers were

asked ‘‘How would you rate your knowledge about the

following areas of fisheries management and their pro-

cesses:’’ which included questions about how size and

catch limitations are determined, how to be involved in the

management process, the development of fisheries man-

agement plans, and about the federal mandate which guides

fishery decision-making, the Magnuson Stevens Act.

Anglers were also asked to rate the current state of

marine fishery resources. To assess angler perceptions of

resource health, anglers were asked, using a scale from 1

(excellent) to 5 (very poor), ‘‘How would you rate the

current state of the following resources’’. The resources

rated included several popular recreational fisheries from

the Northeast US, the overall fishery resources in the

region, and the fishery in which they participate most often.

Finally, anglers were asked whether they participated in

fisheries decision-making. In a marine fisheries context,

there are several opportunities in which stakeholders can

participate in fisheries decision-making. These range from

appointment to regional councils and fishery advisory

panels to attendance at fishery management meetings to

membership in organized groups who comment on

proposed fishery actions. Given the wide range of oppor-

tunities and the varied ways in which the idea of partici-

pation can be interpreted by individuals as a meaningful

endeavor, anglers were asked on a dichotomous scale to

indicate whether or not they participated in fisheries man-

agement decision-making. This question was designed to

characterize the number of anglers who believed that they

were involved in the fishery decision-making process in

some manner and not to determine the efficacy or the

extent of different levels of participation.

The final survey instrument was administered in person

(either computer-based or pen and paper) at two north-

eastern fishing expo events. Researchers manned a booth

and anglers were solicited to take part in the study.

Study Sample

Participants in this study were attendees at two northeast-

ern US saltwater fishing expositions in the late winter and

early spring of 2008. These are large trade shows typically

held in convention centers or similar venues with atten-

dance averaging between 5,000 and 10,000. Exhibitor

booths at these events range from recreational fishing ser-

vices and equipment to non-governmental and govern-

mental organizations. They are usually held in late winter

in the northeast US when conditions for fishing are poor.

Study participants were a convenience sample of general

audience attendees over a period of three days. The number

of surveys administered were similar at both 3 day events

in New Jersey (n = 118) and in Rhode Island (n = 122).

Given the nature of the events as a primarily commercial

endeavor and the restrictions placed on survey adminis-

tration by sponsors of these events, a random sample of

attendees was not possible. Although the data collected are

not representative of the population of saltwater expo

attendees, sampling techniques did attempt to solicit par-

ticipation from all general audience members which came

in close proximity of the survey booth. We estimate that

approximately 25 to 30% of individuals who were verbally

or visually invited to participate (e.g., through the banner

which advertised the survey above the booth) filled out the

survey to completion.

Although recreational anglers can be typified by a

variety of characteristics including consumptive behavior

(Fedler and Ditton 1994) and fisheries policy support

(Arlinghaus 2005), recreational anglers on the whole

remain a relatively diverse group. Our sample represents

coastal anglers who were predominantly male (*95%).

Most reported to fish from personal boats in coastal waters

at least 6–12 times per year (*79%) and most (64%)

fished on average more than once a week. According to

recent assessments, the average recreational angler effort is

13 days per year in New Jersey and 11 days per year in


123

Rhode Island (US Fish & Wildlife Service (USFW) 2006).

Thus, the majority of our study participants reported well

above average fishing effort compared to the general

population of recreational anglers in both states and can be

thought of as frequent marine anglers.

Survey Analysis

After survey responses were collected, confirmatory factor

analysis (CFA) was conducted to ensure construct valid-

ity. To validate the variables and questions, we carried

out a reliability analysis using Cronbach’s alpha. Under

the analysis, items which had low factor loadings (\0.6)

or low Cronbach alphas (\0.6) were omitted as suggested

by Kim (2004) and Salisbury and others (2001). Using

this reliability analysis, the final latent variables and

questions which comprise these constructs used in the

final model are shown in Table 1. Ultimately, fisheries

science knowledge included three variables, fisheries

management knowledge included five variables, rating of

the resources included four variables, and participation

included one variable. To test the hypothesized model,

maximum likelihood estimates were obtained using

AMOS 17 software. The covariance matrix used to fit the

proposed model was calculated using survey data col-

lected in both New Jersey and Rhode Island and param-

eter estimates were obtained.

Results

CFA validated question categories and Cronbach’s alpha

for all latent variables indicated reliability between 0.84

and 0.91 (Table 1). Several model fit measures are usually

used to determine goodness of fit. Our model surpassed

the recommended values for all model fit indices, indi-

cating a higher likelihood that our model was valid

(Table 2). The most common goodness of fit test is the

chi square index which should reflect no significant dif-

ference and the model would ideally yield a value of

between 2.0 and 5.0. Our obtained value 2.7 indicated

good fit. The root mean square error of approximation

(RMSEA) provides an indication of the discrepancy

between the observed and model generated covariance.

A RMSEA value [0.05 indicates a good fitting model

where a value between 0.05 and 0.08 indicates a rea-

sonable model fit and [0.08 indicates an ill fitting model.

Our obtained value 0.078 indicated reasonable model fit.

Goodness of fit index (GFI) and Adjusted Goodness of Fit

Index (AGFI) measures fit compared to no model at all

(Joreskog and Sorbom 2001) and both values should be

[0.9 (GFI) and [0.8 (AGFI). Our obtained values 0.9

and 0.83 indicated reasonable fit.

Knowledge of Fisheries Management and Science

Survey responses indicate similarities between New Jersey

and Rhode Island recreational anglers in rating their

familiarity with fishery science and fishery management on

a scale from (1) very knowledgeable to (5) very

unknowledgeable. Here, we report the overall mean value

of responses. Overall, recreational anglers hovered around

the slightly knowledgeable side of the middle designation

(M = 3.0) with determining age class based on size

(M = 2.45), healthy habitat designation for larval and

juveniles fish (M = 2.63) and determining sex of a fish

when caught (M = 2.98). Similarly, recreational anglers

had higher, but similar mid-range ratings of familiarity

with civic aspects of fisheries management such as how

size (M = 2.13) and catch (M = 2.12) limits were deter-

mined with less reported familiarly with how to be

involved in the fisheries management process (M = 2.66)

familiarity in the development of fisheries management

(M = 2.66) plans and fishery legislation (M = 2.60).

Ratings of the Resource

Overall, recreational anglers rate the current state of fish-

eries resources as fair to good in all categories on a scale

from (1) Excellent to (5) Very Poor. The bluefish fishery

was most highly rated (M = 2.05), followed by the fishery

the individual participates in most (M = 2.30), followed by

overall regional (either Aid-Atlantic or New England)

resources (M = 2.67) with similar ratings for the summer

flounder fishery (M = 2.66).

Trust by Management Construct

Survey results indicate that trust in fisheries science and

management varies by scale. Survey responses for amount

of trust for state, regional, federal, recreational science and

commercial science were averaged. Only state (M = 2.82)

and regional (M = 2.90) institutions were more trusted

than not trusted while recreational (M = 3.01) science,

federal level management (M = 3.08) and commercial

science (M = 3.38) all fell on the distrust side of the

spectrum.

Structural Equation Modeling (SEM)

The structural equation model standardized multiple

regression indicated that participation had the most influ-

ence in determining level of trust (Table 3). In fact, par-

ticipation predicted strong positive correlations with state

(r = 0.72), regional, (r = 0.77), federal (r = 0.79), recre-

ational science (r = 0.79) and commercial science

(r = 0.82). These were all found to be statistically


123

significant at 0.05. Ratings of the resources were also found

to have positive correlations and strongly influence trust,

but were only significantly for state (r = 0.26) and regional

(r = 0.26) levels of management and not for the other

constructs.

Discussion

Our results begin to shed light on the complex issue of trust

in natural resource management. First, our findings from

those surveyed indicate that trust in fisheries management

varies across dimensions (in scientific assessments versus

government agencies) and scale (local versus state versus

federal agencies). This has important implications for

efforts to increase trust in fisheries management and

potentially across natural resource management institu-

tions. These findings should also inform future studies on

the role of trust in fisheries management, in particular, its

measurement. Second, we find that trust in fisheries

Table 1 Summary of latent variables, measurement instrument, question loadings, validation and averages to questions responses for NJ and RI

Latent variable NJ mean (SD) RI mean (SD) T mean (SD) Alpha (a)

loadings

Survey questions

Knowledge of fisheries sciencea 0.89

Habitat conditions needed for larval or juvenile fish 2.60(1.3) 2.65(1.3) 2.63(1.3) 0.75

Determine age class based on size 2.37(1.1) 2.54(1.0) 2.45(1.0) 0.71

Determine sex of a fish caught 2.8(1.2) 2.9(1.2) 2.98(1.3) 0.7

Knowledge of fisheries managementa 0.91

How size limitations are determined 2.07(1.1) 2.20(1.1) 2.13(1.1) 0.78

How catch limitations are determined 2.11(1.1) 2.15(1.1) 2.12(1.1) 0.77

How to be involved in the management process 2.66(1.4) 2.66(1.4) 2.66(1.4) 0.74

The development of fishery management plans 2.72(1.4) 2.62(1.4) 2.66(1.4) 0.72

The Magnuson Stevens Act 2.65(1.5) 2.5(1.6) 2.60(1.6) 0.69

Rating of the resourcesb 0.84

Overall fishery resources 2.61(1.1) 2.71(1.1) 2.67(1.1) 0.72

Fisheries in which you participate most 2.25(1.1) 2.35(1.1) 2.30(1.1) 0.67

Summer flounder fishery 2.42(1.3) 2.9(1.2) 2.68(1.3) 0.73

Bluefish fishery 2.04(1.0) 2.07(1.0) 2.05(1.0) 0.66

Institutional trustc 0.9

State fisheries institutions 2.84(1.4) 2.78(1.2) 2.82(1.2) 0.73

Regional fisheries management councils 3.08(1.4) 2.84(1.1) 2.9(1.3) 0.79

Federal fisheries management (NMFS) 3.27(1.5) 2.7(1.4) 3.01(1.5) 0.81

Scientific trustc 0.9

Assessment science used to set seasonal limits for commercial fishermen 3.66(1.4) 3.13(1.3) 3.38(1.4) 0.81

Assessment science used to set seasonal limits recreational fishermen 3.33(1.5) 2.87(1.2) 3.09(1.4) 0.8

Participation

Do you participate in fisheries management in some way? Yes: 48% Yes: 38% Yes: 43% n/a

No: 52% No: 61% No: 57%

a Derived Cronbach’s alpha reliability test and factor loadings derived from principle components analysis (PCA)a Scale of (1) very knowledgeable to (5) very unknowledgeableb Scale (1) excellent to (5) very poorc Scale (1) fully trust to (5) fully distrust (3 is neutral)

Table 2 Summary of model fit indices, recommended, and obtained

values for hypothesized model

Statistic Recommended

value (P)

Obtained

value

X2 – 286.95

df – 116

X2/df [0.05** 2.47

Goodness of fit (GFI) [0.9** 0.9

AGFI [0.8** 0.829

CFI [0.9** 0.932

NFI [0.9** 0.9

RMSEA 0.05–0.08** 0.078

** Surpasses recommended value


123

management is influenced by two factors. Although par-

ticipation in fisheries management was the only variable

that predicted trust across all constructs measured, an

individual’s perception of the health of a fishery was also

correlated with levels of trust. These survey results provide

us preliminary insights into the type of environments in

which trust in natural resource management institutions

might be formed.

Partitioning our survey results to just examine the

reported levels of trust in the concepts measured indicated

lower levels of trust in scientific assessments than gov-

ernmental agencies (with only the federal management

agency reporting as low levels of trust as the scientific

assessments). This lack of trust in scientific assessments is

problematic because fisheries management relies heavily

on the scientific assessment of the resource (population

estimates) and its social pressures (amount and frequency

of fishing activity) to set regulations. These estimates are

often made under conditions of high uncertainty yet have

direct implications for anglers because they are used as a

reference point for setting harvest limits. This uncertainty

is a result of the current limitations of stock assessment

methods to account for the multitude of non-linear rela-

tionships beyond fishing pressure that influence population

abundance. Retrospective analyses have highlighted that

over-estimates of population have led to stock collapses

(Myers and others 1997), while underestimates of popula-

tions have limited fisheries ability to produce social and

economic benefits to society (Hilborn 2002). In a stochastic

environmental system like the ones that characterize fish-

eries, the same assessment methodology may be applied

over a relatively short period of time to the same stock yet

yield oscillating estimates which may be interpreted by

stakeholders as erratic, imprecise, and uncertain. Manage-

ment decisions based on seemingly non-stable behavior

may result in lower levels of trust and compromise the

willingness of an angler to accept vulnerability (e.g.,

limiting harvests) based on these seemingly unstable

estimates.

Another reasonable explanation for low levels of trust in

science maybe reflected in stakeholder conflation between

understanding scientific methodology and the resulting

policy that emerges from these assessments. Low levels of

trust may reflect an issue of management credibility, where

respondents assign low truth value to the claims made by

the assessments (Barber 1983). Interestingly, participants

in our study rated commercial scientific assessments as less

trusted than recreational scientific assessments. This raises

an interesting question for future research on if this dis-

tinction is consistent and why it would be that frequent

recreational anglers perceive a difference in the processes

or outcomes in the two. Do they really think the assess-

ments are not valid? Or are they more concerned with the

policy-making process where outcomes of these assess-

ments are used to justify various policy decisions? In a

fisheries context, scientific assessments where a fish stock

has fallen will be used to justify a decision to limit recre-

ational and commercial harvests. If trust is low in scientific

assessments because the public confuses them with

unpopular policy outcomes (such as limiting harvests) this

may not be addressed by changing the assessment process

or increasing public participation in commercial and rec-

reational scientific assessments. The exact reasons for low

ratings of trust in science are unclear and are an area that

would benefit from further research.

In addition to respondents differentiating between sci-

entific and governmental institutions, trust also appears to

differ with scale of governance. The highest levels of trust

are reported for state fisheries institutions whereas the

federal government has the lowest level of trust. Past

research has indicated that familiarity is thought to affect

trust (Dunn and Schweitzer 2005) and local level gover-

nance is often more trusted than higher levels of organi-

zation (La Porta and others 1997). Although not large

Table 3 Structure equation model coefficients, and standardized multiple regression coefficients

Trust of fishery management institutions Trust of fishery science

State Regional Federal Recreational Commercial

coefficient

(standardized b)

coefficient

(standardized b)

coefficient

(standardized b)

coefficient

(standardized b)

coefficient

(standardized b)

Fishery science

knowledge

0.06(0.04) -0.05(-0.04) -0.03(-0.02) -0.04(-0.03) 0.19(0.12)

Fishery management

knowledge

0.01(0.01) 0.12(0.09) 0.08(0.06) -0.05(-0.03) -0.08(-0.05)

Resource ratings 0.36(0.26)*** 0.37(0.26)*** 0.18(0.11) 0.10(0.07) 0.10(0.06)

Participation 6.99(0.72)*** 7.53(0.77)*** 9.19(0.79)*** 8.56(0.79)*** 9.02(0.82)***

R2 0.58 0.68 0.69 0.64 0.68

*** Indicates statistical significance at the 0.05 level


123

differences in the level of trust, these differences indicate

that our survey participants may discern between levels of

government and have higher levels of trust for local gov-

ernment. Anglers may feel more familiar with their state

agency and may perceive a clearer understanding of their

goals and intentions. Conversely, the goals of management

institutions which are more removed from the local context

may be unknown or difficult to assess. Smaller-scale gov-

ernance which is more familiar to the user may be seen as

more credible which has been shown to improve resource

health. In a comparison of top-down management decisions

based on scientific models which were not credible among

users to management decisions that were user-generated,

Dietz and others (2003) found increases in resource health

as a function of community lead assessments and the

generation of acceptable rules which increased compliance

behaviors.

Along with seeking to understand the varying degrees of

institutional trust, this study also attempted to identify

factors which influence trust. Of the variables investigated,

participation in fisheries management was the only variable

that accounted for higher levels of trust across all con-

structs. This finding may confirm previous studies that have

long supported that trust between parties increases as the

level of participation in decision-making increases (Beierle

and Konisky 2000; Chess 2000; National Research Council

(NRC) 2008). These findings add to this literature by

providing preliminary data that indicate that trust in sci-

entific methods of environmental assessment may also

increase with participation. Science plays a unique role in

environmental decision-making and ideally assessments

are guided by the knowledge, values and preferences of the

affected parties (National Research Council (NRC) 2008).

However our results indicate that stakeholders may also

learn to trust scientific assessments as they become more

familiar with the assessment techniques, questions, and the

institutions charged with conducting them through partic-

ipation. As Beratan (2007) points out, shared understanding

and trust are built through many personal interactions over

time which facilitates learning and participating in deci-

sion-making may allow for opportunities for this type of

learning. As stakeholders gain a better understanding of

institutional values and practices, the goals and intent

of scientific assessments may become more familiar,

therefore increasing trust.

Anglers’ perceptions that a fishery was in good health

was also found to positively influence trust, but only for the

state and regional levels of management. Higher ratings of

fishery resources were found to indicate higher levels of

trust in state and regional management. These ratings

indicate that respondents may attribute resource health to

the decisions and actions of smaller scale management

institutions. However, the inverse may also be true: if

respondents rate resource health as being in poor condition,

they may report lower levels of trust in the managing

institutions. This supports the idea that trust may be a

reflection of a respondent’s confidence in the institution’s

technical competence to carry out their mission (Barber

1983).

There are limitations to our study. First, our sample is

limited in its representativeness of the range of recreational

anglers that exist. It was a convenience sample ascertained

through a limited selection of those attending fisheries

expos in Rhode Island and New Jersey. Surveying a ran-

dom sample of anglers is difficult as there was no publicly

available listing of those who participated in recreational

marine fisheries at the time the survey was undertaken.

This method of recruitment was preferable to random

sampling of the public. As mentioned, the anglers surveyed

reported fishing more frequently than the average angler.

This clearly makes them an important stakeholder group

for the management of the recreational fisheries because

arguably they have more interaction with the managed

resources and therefore their beliefs and behaviors are

important components which need to be considered when

decisions are being made. But it also means that the gen-

eralizability of our results is limited and may vary from

those that spend less time fishing. For example, if the lack

of trust of scientific assessments stems from confusion of

its use in the political process to limit fishing then this

group might be more affected and thus less trusting. In

addition, the sample is limited geographically in the

Northeast. Different states and regions may reflect different

levels of trust with the state and regional fisheries relevant

to them because of historical experiences and a different

knowledge-based by which trust judgments are formed. In

addition, the sample size, although adequate, is relatively

small for the type of analysis performed. Despite these

shortcomings, this sample is useful for exploring relation-

ships between variables using SEM and does provide

empirical evidence to support potential relationships

between cognitive and behavioral factors and trust in nat-

ural resource management constructs.

Second, our survey measurement items are limited. We

define trust as a willingness to accept vulnerability based

upon positive expectations of the intentions or behaviors of

others (Rousseau and others 1998). However, those

answering our survey may have different understandings of

trust such as those based more upon assessments of an

institution’s technical competence and ability to fulfill

responsibilities. In addition, the data we have on past

participation is simply a binary yes or no. This measure

should certainly be refined in future studies to more thor-

oughly investigate the relationship between trust and par-

ticipation. Though the length of a survey is always an

important consideration, given the emergent importance of


123

participation on trust in fisheries management institutions,

it would be helpful to know more about the form, content

and extent of the participation to help disentangle the

relationship. A more detailed analysis of objectively

defined knowledge and fine-scale analysis of type of par-

ticipation and their influence on trust is expected to be the

next step in parsing these variables beyond looking for

general correlations. However, given the dearth of infor-

mation in the literature about predictors of trust, we con-

sider this paper to be an initial inspection by which more

detailed hypotheses can be developed.

In addition, this study measured a respondent’s assess-

ment of trust at a specific point in time. It is important to

note that building trust is a dynamic and evolving process

and these levels are not likely to remain constant. The

causality between trust and participation is not unidirec-

tional. Rather, initial levels of trust may predict participa-

tion and participation may lead to more or less trust. Trust

is fragile, slow to grow and easy to break and as stake-

holders learn about potential changes in management

action, stakeholder trust may change (McKnight and

Chervany 1996; Beratan 2007). However, the process of

building trust is also seen as self-reinforcing where existing

trust may maintain trust in the absence of new events or

information (Blomqvist 1997). Therefore, although we

report the levels of trust at a certain point in time, we can

only expect these levels to remain as constant as new

significant developments in the fishery. Future studies

could be developed to survey individuals over time and

monitor changing trust and participatory experiences.

There are still many questions regarding other reliable

predictors of trust in natural resource management. In our

study, familiarity with scientific concepts used to make

regulations was not a significant predictor in trust judg-

ments in fisheries science. Likewise, familiarity with fish-

ery management processes did not yield significant positive

associations with trust judgments of institutions. These

results indicate that trust is not affected by how knowl-

edgeable individuals rate themselves in areas of abstract

components and processes of environmental management.

Rather, it is much more important that stakeholders have an

opportunity to interact with these components. As Rowe

and Frewer (2000) point out, there is not only a need for

institutions to learn how to effectively communicate

complex ideas to stakeholder groups, but also to create

opportunities where these complex ideas can be discussed

and debated (Hunt and others 1999).

Conclusions

The results of our investigation on trust can be interpreted

as both a learned and affective responses. As many have

pointed out, participation in decision-making by those

affected by the decision is thought to increase trust

(Beierle1998; Beierle and Cayford 2002; Dietz and others

2003; National Research Council (NRC) 2008). This is the

result of institutions and users forming relationships where

the goals of both groups are clearly articulated (Rousseau

and others 1998; Beratan 2007). In these cases, trust is the

result of active information processing and relies less on

emotions and more on a learned experience. However, in

large-scale environments where there few interactions

between institutions and stakeholders, stakeholders will rely

more on evidence generated from their own experiences and

associations to determine the intentions of institutions. Dunn

and Schweitzer (2005) found that when there is little history

or experience between individuals and a trustee, emotions

supplant information to form trust judgments. Thus, higher

ratings of trust can be expected with smaller scale institutions

and when experience indicates to stakeholders that the

resources and the environment are healthy.

It may be more difficult, however, to build trust in the

scientific methods used to make resource management

decisions. Although our study indicates that participation in

environmental decision-making is positively associated

with trust in science, low ratings of trust in science were

reported overall. Previous studies have linked distrust to

science to stakeholder conflation of scientific assessment,

the resulting policy, and the outcome of policy (Haerlin

and Parr 1999). Therefore, natural resource managers

should clearly outline the scientific questions being asked,

the methods and knowledge produced, how this knowledge

informs natural resource management policy and most

importantly how the knowledge created can be integrated

with the goals of stakeholders.

Acknowledgment This research was conducted under NOAA

Award NA07NOS4200129. The authors would like to thank the

Jacques Cousteau National Estuary Research Reserve (JCNERR), the

Rhode Island Saltwater Anglers Association, Brandon Johnson, Caron

Chess, and Josh Kohut.

References

Arlinghaus R (2005) A conceptual framework to identify and

understand conflicts in recreational fisheries systems, with

implications for sustainable management. Aquatic Resources,

Culture and Development 1:145–174

Arnstein SR (1969) A ladder of public participation. Journal of the

American Planning Association 35:216–224

Barber B (1983) The logic and limits of trust. Rutgers University,

New Brunswick 189pp

Barclay P (2004) Trustworthiness and competitive altruism can also

solve the ‘‘tragedy of the commons’’. Evolution and Human

Behavior 25:209–220

Beierle TC (1998) Public participation in environmental decisions: an

evaluation framework using social goals. Resources for the

Future Press, Washington DC, p 31


123

Beierle T, Cayford J (2002) Democracy in practice public: partici-

pation in environmental decisions. Resources for the Future

Press, Washington, DC, p 160

Beierle TC, Konisky DM (2000) Values, conflicts, and trust in

participatory environmental planning. Journal of Policy Analysis

and Management 19:587–602

Beratan KK (2007) A cognition-based view of decision processes in

complex social-ecological systems. Ecology and Society 12(1):27

Blomqvist K (1997) The many faces of trust. Scandinavian Journal of

Management 13:271–286

Carroll MS, Hendrix WG (1992) Federally protected rivers: the need

for effective local involvement. Journal of the American

Planning Association 58:346–353

Castelfranchi C, Falcone R (2000) Trust is much more than subjective

probability: mental components and sources of trust. Proceed-

ings of the 33rd

Chess C (2000) Evaluating environmental public participation:

methodological questions. Journal of Environmental Planning

and Management 43:769–784

Chess C, Purcell K (1999) Public participation and the environment: Do

we know what works? Environmental Science and Technology

33:2685–2692

Cvetkovich G, Winter P (2003) Trust and social representations of the

management of threatened and endangered species. Environ-

mental Behavior 35:286–307

Davenport M, Leahy J, Anderson D, Jakes P (2007) Building trust in

natural resource management within local communities: a case

study of the Midewin National Tallgrass Prairie. Environmental

Management 39:353–368

Dickson E, Gordon S, Huber GA (2009) Enforcement and compliance

in an uncertain world: an experimental investigation. The Journal

of Politics 71:1357–1378

Dietz T, Ostrom E, Stern P (2003) The struggle to govern the

commons. Science 302:908–1912

Dietz T, Dan A, Shwom R (2007) Support for climate change policy:

social psychological and social structural influences. Rural

Sociology 72:185–214

Dobbs D (2000) The great gulf: fishermen, scientists, and the struggle

to revive the world’s greatest Fishery. Island Press, Washington,

DC, p 206

Dunn J, Schweitzer M (2005) Feeling and believing: the influence of

emotion on trust. Journal of Personality and Social Psychology

88:736–748

Earle TC, Cvetkovich GT (1995) Social trust: toward a more

cosmopolitan society. Praeger, Westport, p 228

Eiser JR, Stafford T, Henneberry J, Catney P (2007) Risk perception

and trust in the context of urban brownfields. Environmental

Hazards 7:150–156

Fedler A, Ditton R (1994) Understanding anlger motivations in

fisheries management. Fisheries 19:6–13

Haerlin B, Parr D (1999) How to restore public trust in science.

Nature 400:499

Hilborn R (2002) The darkside of reference points. Bulletin of Marine

Science 70:403–408

Hoppner C (2009) Trust—A monolithic panacea in land use

planning? Land Use Policy 26:1046–1054

Hunt S, Frewer LJ, Shepherd R (1999) Public trust in sources of

information about radiation risks in the UK. Journal of Risk

Research 1:167–181

Johnson B (1999) Exploring dimensionality in the origins of hazard-

related trust. Journal of Risk Research 2:325–354

Johnson TR, van Densen WLT (2007) Benefits and organization of

cooperative research. ICES Journal of Marine Science 64:862

Joreskog KG, Moustaki I (2001) Factor analysis of ordinal variables:

a comparison of three approaches. Multivariate Behavioral

Research 36:347–387

Kaplan IM, McCay BJ (2004) Cooperative research, co-management

and the social dimension of fisheries science and management.

Marine Policy 28:257–258

Kim KS (2004) AMOS analysis of structural equation modeling.

Hannarae, Seoul

Lawrence R, Daniels S, Stankey G (1997) Procedural justice and

public involvement in natural resource decision making. Society

and Natural Resources 10:577–589

Leahy J, Anderson D (2008) Trust meanings in community-water

resource management agency relationships. Landscape and

Urban Planning 87:100–107

McAllister RRJ, Gordon IJ, Janssen MA (2005) Trust and cooperation

in natural resource management: The case of agistment in

rangelands. In: Zerger A, Argent RM

McKnight DH, Chervany NL (1996) The meanings of trust. Tech Rep

MISRC working paper series. Calson School of Management,

University of Minnesota, St. Paul, p 32

Models for mining equipment selection (2005) In: Zerger A, Argent

RM (eds) MODSIM 2005 International congress on modelling

and simulation. Modelling and Simulation Society of Australia

and New Zealand, pp. 170–176

Myers R, Hutchings J, Barrowman NJ (1997) Why do fish stocks

collapse? The example of cod in Atlantic Canada. Ecological

Applications 7:91–106

National Research Council (NRC) (2008) The effects of public

participation. In: Dietz T, Stern PC (eds) Public participation in

environmental assessment and decision-making. National

Research Council Press, Washington, DC, p 322

Needham MD, Vaske JJ (2008) Hunters’ perceptions of similarity and

trust in wildlife agencies and personal risk associated with

chronic wasting disease. Society and Natural Resources 21(3):

197–214

Ostrom E (2007) Collective action and local development processes.

Sociologica 3:1–32

Ostrom E, Walker J (2003) Trust and reciprocity: interdisciplinary

lessons for experimental research. In: Ostrom E, Walker J (eds)

The Russell sage foundation series on trust. Russell Sage

Foundation, New York, p 409

Ostrom E, Norberg J, Wilson J, Walker B (2008) Diversity and

resilience of social-ecological systems. In: Norberg J, Cumming

G (eds) Complexity theory for a sustainable future. Columbia

University Press, New York, pp 105–136

Owen A, Virderas J (2008) Trust, cooperation, and implementation of

sustainability programs: the case of Local Agenda 21. Ecological

Economics 68:259–272

Rousseau DM, Sitkin SB, Burt RS, Camerer C (1998) Not so different

after all: a cross-discipline view of trust. Academy of Manage-

ment Review 23:393–404

Rowe G, Frewer JL (2000) Public participation methods: a framework

for evaluation. Science Technology and Human Values 25:

13–29

Salisbury WD, Pearson RA, Pearson AW, Miller DW (2001)

Perceived security and World Wide Web purchase intention.

Industrial Management and Data Systems 101:65–176

Schwarz N, Clore GL (1983) Mood, misattribution, and judgments

of well-being: Informative and directive functions of affective

states. Journal of Personality and Social Psychology 45:513–

523

Skogen K (2001) Who’s afraid of the big, bad wolf? Young peoples’

responses to the conflicts over large carnivores in eastern

Norway. Rural Sociology 66:203–226

Skogen K, Thrane C (2008) Wolves in context: using survey data to

situate attitudes within a wider cultural framework. Society and

Natural Resources 21:17–33

Slovic P (1993) Perceived risk, trust and democracy. Risk Analysis

13:675–682


123

Smith P, McDonough M (2001) Beyond public participation: fairness

in natural resource decision making. Society and Natural

Resources 14:239–249

US Fish & Wildlife Service (USFW) (2006) National Survey

of Fishing, Hunting, and Wildlife-Associated Recreation.

Washington, DC

Webler T, Tuler S (2000) Fairness and competence in citizen

participation: reflections from a case study. Administration and

Society 32:56–595

Wilson DC, Nielsen JR, Degnbol P (eds) (2003) The fisheries

co-management experience: accomplishments, challenges and

prospects. Kluwer Academic, Dordrecht, p 324


123

Understanding Factors That Influence Stakeholder Trust of Natural Resource Science and Institutions

Documents

Transcript of Understanding Factors That Influence Stakeholder Trust of Natural Resource Science and Institutions