This article was downloaded by: [UQ Library]On: 15 November 2014, At: 10:40Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Human Dimensions of Wildlife: AnInternational JournalPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/uhdw20

Perceived Risks from Disease andManagement Policies: An Expansionand Testing of a Zoonotic Disease RiskPerception ModelHeather A. Triezenberga, Meredith L. Goreab, Shawn J. Rileya &Maria K. Lapinskicd

a Fisheries and Wildlife, Michigan State University, East Lansing,Michigan, USAb School of Criminal Justice, Michigan State University, East Lansing,Michigan, USAc Communication, Michigan State University, East Lansing, Michigan,USAd Michigan AgBioResearch, Michigan State University, East Lansing,Michigan, USAPublished online: 24 Mar 2014.

To cite this article: Heather A. Triezenberg, Meredith L. Gore, Shawn J. Riley & Maria K. Lapinski(2014) Perceived Risks from Disease and Management Policies: An Expansion and Testing of a ZoonoticDisease Risk Perception Model, Human Dimensions of Wildlife: An International Journal, 19:2,123-138, DOI: 10.1080/10871209.2014.844288

To link to this article: http://dx.doi.org/10.1080/10871209.2014.844288

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

Human Dimensions of Wildlife, 19:123–138, 2014Copyright © Taylor & Francis Group, LLCISSN: 1087-1209 print / 1533-158X onlineDOI: 10.1080/10871209.2014.844288

Perceived Risks from Disease and ManagementPolicies: An Expansion and Testing of a Zoonotic

Disease Risk Perception Model

HEATHER A. TRIEZENBERG,1 MEREDITH L. GORE,1,2

SHAWN J. RILEY,1 AND MARIA K. LAPINSKI3,4

1Fisheries and Wildlife, Michigan State University, East Lansing, Michigan, USA2School of Criminal Justice, Michigan State University, East Lansing, Michigan,USA3Communication, Michigan State University, East Lansing, Michigan, USA4Michigan AgBioResearch, Michigan State University, East Lansing, Michigan,USA

Human dimensions information about wildlife disease risk management is an importantcomponent in developing successful policies because policy implementation depends onstakeholder involvement and public support. Understanding how stakeholders perceiverisks is fundamental to successful disease management, yet a clear systematic theo-ry-based framework is lacking. We expanded and tested a portion of the zoonotic diseaserisk information seeking and processing (ZDRISP) framework using a tailored designmethod survey with a sample (n= 4,000) of Michigan deer hunters. Our study revealedrespondents perceived risks from wildlife disease management policies, which are influ-enced by disease risk perceptions and subjective and descriptive norms. These resultsadvance a zoonotic disease risk perception theoretical framework that can be applied tounderstand stakeholder perceptions of different wildlife diseases having varying levelsof prevalence, susceptibility, or severity. Integrating insights about human perceptionsof the disease and its management policies will enhance likelihood of success of wildlifedisease management.

Keywords bovine tuberculosis, risk perceptions, white-tailed deer, wildlife disease,zoonotic disease risk information seeking and processing

Introduction

Around the world, successful wildlife disease–related management policies may benefitfrom incorporating an understanding of how stakeholders engage with risk informationbecause management strategies depend on public support and implementation (Clarke,2009; O’Brien et al., 2001; O’Brien, Schmitt, Rudolph, & Nugent, 2011; Zinsstag,2008). Effectiveness of each of three management strategies commonly used to minimizeimpacts of wildlife diseases—creating socially acceptable control measures, implement-ing public policy to address diseases, and creating sufficient public awareness (Wobeser,

Address correspondence to Heather A. Triezenberg, Michigan State University, Fisheries andWildlife, 480 Wilson Road, 13 Natural Resources Building, East Lansing, MI 48824, USA. E-mail:vanden64@msu.edu

123

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

124 H. A. Triezenberg et al.

2002)—benefit from knowledge of stakeholders’ beliefs and attitudes toward wildlifediseases, and how stakeholders are influenced by societal norms. The likely actions stake-holders may take in response to wildlife diseases are based on their risk perceptions and riskinformation they encounter, whether from a management agency, mass media, or informalnetworks, and can be either helpful or unhelpful at managing the disease (Clarke, 2009;Maddux & Rogers, 1983; Witte & Allen, 2000). A human dimensions research frameworkcan enable understanding of how and why wildlife-associated disease risk perceptions varyacross stakeholders and disease types (Clarke, 2009; Decker et al., 2010; Vaske, Shelby, &Needham, 2009). To accelerate progress toward closing the gap on systematic and theory-based human dimensions of wildlife-disease studies, our project expanded and tested thecognitive hierarchy portion of the zoonotic disease risk information seeking and processing(ZDRISP) model. A theory-based and empirically tested zoonotic disease risk percep-tion model can enhance diagnostic capacity and protocols for managing wildlife disease(Zinsstag, 2008).

We conducted our study within the context of antlerless deer hunters’ perceptionsrelated to bovine tuberculosis (TB) in northeast Michigan. Currently, TB prevalence rate inwhite-tailed deer (Odocoileus virginianus) is estimated at less than 2%, yet progress towardachieving the eradication of the disease has slowed in recent years (O’Brien, Schmitt,Fitzgerald, & Berry, 2011). Public resentment, especially among deer hunters, to controlmeasures such as liberal deer harvest policies and baiting restrictions has increased just aspopulation reduction is beginning to reach the management objective (O’Brien, Schmitt,Fitzgerald, et al., 2011).

Zoonotic Disease Risk Perception Model and Testing

The ZDRISP model (Clarke, 2009) extends the risk information seeking and processingmodel (RISP) (Griffin, Dunwood, & Neuwirth, 1999) by incorporating wildlife value ori-entations and personal values (Fulton, Manfredo, & Lipscomb, 1996), perceived “role” asan opinion leader (Clarke, 2009), and specifies perceived hazard characteristics and offersa framework for how individuals engage (e.g., search, collect, deal with, act upon) withinformation related to zoonotic diseases. One challenge with the ZDRISP model is it iscomplex, making it difficult for researchers to empirically test the entire model and forwildlife managers to fully incorporate this diagnostic resource into wildlife disease man-agement. Our research addresses this challenge by focusing on a portion of ZDRISP model(Clarke, 2009).

The relationship between personal values, wildlife value orientations, and perceivedhazard characteristics in the ZDRISP model is similar in structure to the cognitive hierarchymodel, which links a limited number of fundamental values, value orientations, attitudes,and higher order attitudes and norms (Clarke, 2009; Fulton et al., 1996). Risk perceptions,attitudes, and perceived hazard characteristics reflect higher-order judgments. The ZDRISPmodel does not include normative constructs in its values, wildlife value orientations, andattitudes portion of the framework, despite norms being important predictors of higherorder attitudes in the cognitive hierarchy model (Fulton et al., 1996). Social norms areimportant drivers of human behavior (Lapinski & Rimal, 2005). Consistent with the cog-nitive hierarchy model, we believe that social norms are a necessary factor for predictinghazard perceptions in the ZDRISP model. We included dimensions of social norms aboutperceptions of what other people are doing (i.e., descriptive norms) and perceptions aboutwhat other people think one should be doing (i.e., subjective norms) relating to harvestingantlerless deer and reducing TB prevalence.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

Perceived Risks from Disease and Management Policies 125

The most substantial adaptation we made to the ZDRISP model was to expand the per-ceived hazard characteristics to not only include perceived disease risks (Clarke, 2009) butalso to include perceived risks associated with the disease management policies (expandeddimensions in italics in Figure 1). Perceptions of risks may differ when one considers thedisease itself as opposed to the policies implemented to manage the disease; understandingthese differences may be important for determining how stakeholders engage with informa-tion related to wildlife disease. If stakeholders’ perceptions reveal distinct levels of concernabout disease risks and disease management risks, then diagnostic tools and policy devel-opment processes can benefit from incorporating how stakeholders assess risks of objects(e.g., wildlife diseases) versus situations (e.g., wildlife disease management policies).

Our goal was to further expand and test the cognitive hierarchy portion of the ZDRISPmodel (Clarke, 2009; Fulton et al., 1996). We specifically wanted to determine and assessfactors contributing (health, economic, recreation, culture of deer hunting) to (a) per-ceived hazard characteristics (disease risks and management of disease risks) and (b) therelationships among the latent factors cognitive hierarchy aspects of the ZDRISP model.

Hypotheses

We developed the following hypotheses (Figure 2).

Fundamental Values

Self-enhancement will have a: direct negative influence on wildlife use value orientation(H1a) and direct positive influence on wildlife protection value orientation (H1e). Self-transcendence will have a: direct positive influence on wildlife use value orientation (H1b)

Affective ResponseWorry

Perceived"Role" as

an OpinionLeader

Information sufficiencySufficiency thresholdCurrent knowledge

Information Subjective Norms

Channel BeliefsMass media

Perceived InformationGathering Capacity

InformationSeeking

AvoidanceSeeking

InformationProcessing

HeuristicSystematic

Perceived Hazard CharacteristicsPerceived disease risks

Health and safetyEconomic concernHuntingCulture of deer hunting

Perceived disease management risksHealth and safetyEconomic concernHuntingCulture of deer hunting

NormsSubjective normsHarvest antlerless deerReduce TB

Descriptive normsOthers are taking action to reduceTB

Wildlife Value OrientationsWildlife useWildlife protection

Personal ValuesSelf-enhancementSelf-transcendenceOpenness to changeConservation

Figure 1. Expanded (in italics) zoonotic disease risk perception model, modified from Clarke’s(2009) zoonotic disease risk information seeking and processing model. This research expands andtests the cognitive hierarchy components, outlined with the dotted line.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

126 H. A. Triezenberg et al.

Figure 2. Conceptual model illustrating theorized relationships among latent factors and structuralregression for zoonotic disease risk perception model, resident deer hunter respondents, northeastMichigan, 2012. The sign on each causal pathway indicates the expected direction of the relationship.

and direct positive influence on wildlife protection value orientation (H1f). Openness tochange will have a: direct negative influence on wildlife use value orientation (H1c) anddirect positive influence on wildlife protection value orientation (H1g). Conservation willhave a: direct positive influence on wildlife use value orientation (H1d) and direct negativeinfluence on wildlife protection value orientation (H1h). In our study context of TB, self-enhancement, self-transcendence, and openness-to-change will negatively influence thewildlife use value orientation because the values are about oneself (Clarke, 2009; Schwartz& Boehnke, 2004) whereas the wildlife use value orientation is about wildlife (Fultonet al., 1996). Similarly, these value concepts will positively influence the wildlife protec-tion value orientation because one could achieve self-enhancement, self-transcendence, oropenness-to-change through protecting wildlife. In contrast, the conservation value conceptwill positively influence the wildlife use value orientation because of the desire to con-tinue wildlife use traditions (Clarke, 2009; Schwartz & Boehnke, 2004), especially those

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

Perceived Risks from Disease and Management Policies 127

prevalent at hunt camps or clubs common in the study area. The conservation value con-cept, with its focus on relationships with other people and social order, will negativelyinfluence the wildlife protection value orientation, which focuses on human relationshipswith wildlife (Fulton et al., 1996).

Wildlife Value Orientations

Wildlife use value orientation will have a: direct positive influence on subjective norms forharvesting antlerless deer (H2a); direct positive influence on subjective norms for reducingTB (H2c); and direct positive influence on descriptive norms that others are doing thingsto reduce TB (H2e). Wildlife protection value orientation will have a: direct positive influ-ence on subjective norms for harvesting antlerless deer (H2b); direct positive influence onsubjective norms for reducing TB (H2d); and direct positive influence on descriptive normsthat others are doing things to reduce TB (H2f). Because wildlife use value orientationsfocus on how wildlife should be used for human benefit (Fulton et al., 1996) they willpositively influence normative perceptions that antlerless deer should be harvested, peopleshould take steps to reduce TB, and other people are taking action to reduce TB. Similarly,wildlife protection value orientation, with its focus on human responsibility toward ani-mals (Fulton et al., 1996), will positively influence the normative perceptions of harvestingantlerless deer, reducing TB, and actions other people are taking to reduce TB.

Social Norms

Subjective norms for harvesting antlerless deer will have a: direct negative influence onperceptions of disease risks attitudes (H3a) and direct negative influence on perceptions ofmanagement of disease risks attitudes (H4a). Subjective norms about reducing TB will havea: direct positive influence on perceptions of disease risks attitudes (H3b) and direct nega-tive influence on perceptions of management of disease risks attitudes (H4b). Descriptivenorms that others are doing things to reduce TB will have a: direct positive influence onperceptions of disease risks attitudes (H3c) and direct negative influence on perceptionsof management of disease risks attitudes (H4c). Normative perceptions related to harvest-ing antlerless deer will negatively influence perceptions of disease risks and perceptions ofmanagement of disease risks because (1) group norms minimizing antlerless deer harvesthave prevailed for decades and (2) wildlife managers have not communicated high threatand high efficacy message frames (e.g., the link between antlerless deer harvest rates andTB prevalence) to hunters (Muter, Gore, Riley, & Lapinski, 2013). Hunters’ perceptionsof subjective norms related to reducing TB and descriptive norms that others are reducingTB will positively influence perceptions that TB is a salient issue, and negatively influenceperceptions of the management of disease risks. In other words, if hunters perceive thatothers want them to do something about TB and are taking action themselves, then they aremore likely to perceive that TB is actually an issue and be less concerned about the risksfrom the TB management.

Attitudes

Disease risk perceptions will have a direct positive influence on perceptions of managementof disease risks attitudes (H5). When hunters’ perceive risks from TB, it will positivelyinfluence their perceptions of risks from TB management. In other words, if they think TBactually is an issue then they will perceive that management policies exist to address theissue. In contrast, low perceptions that TB is an issue will negatively influence perceptions

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

128 H. A. Triezenberg et al.

of management policies intended to manage a disease, because after all why have TBmanagement policies if in their view TB is not an issue?

Case Study

Bovine tuberculosis (TB), principally a disease of cattle caused by the bacteriaMycobacterium bovis, has become endemic in free-ranging white-tailed deer in northeastMichigan, USA (O’Brien et al., 2002). Citizens, livestock, and wildlife may be impacted bysocial, economic, health, and ecological risks from the presence of TB (Muter et al., 2013).In 1997, a statewide committee of TB-stakeholders (e.g., scientists, livestock producers,wildlife professionals, hunters, citizens) recommended eliminating TB in deer in Michigan(Carstensen, O’Brien, & Schmitt, 2011). A 2011 study revealed the TB prevalence rate indeer in the core TB endemic area (∼1,500 km2) is approximately 2% (Carstensen et al.,2011; O’Brien, Schmitt, Fitzgerald, et al., 2011). Progress toward achieving TB manage-ment objectives (e.g., eradication) has slowed and public resentment, especially among deerhunters, of control measures has increased in recent years (O’Brien, Schmitt, Fitzgerald,et al., 2011). Deer hunters’ willingness to sustain aggressive deer harvests has declinedas they see fewer deer (O’Brien, Schmitt, Fitzgerald, et al., 2011). Achieving TB eradica-tion will require stakeholder support (Muter et al., 2013). Understanding how deer hunters,especially those interested in harvesting antlerless deer in the core TB endemic area, engagewith information about TB risks may be a determinant of successful wildlife managementpolicies (O’Brien, Schmitt, Rudolph, et al., 2011).

Methods

Sampling Design, Sampling, and Data Collection

The sample frame for this research was drawn from residents who purchased antlerless deerlicenses for the TB area in northeast Michigan anytime during the past five hunting seasons(2007–2011) or resident hunters who reported deer hunting in the region in recent huntingseasons (2007–2009) on the annual harvest surveys. The Michigan Department of NaturalResources (MDNR) provided contact information of hunters meeting our sample framecriteria. We randomly selected 3,000 antlerless license holders, 1,500 each for frequent(>3 years out of the 5-year period) and infrequent (<2 years out of the 5-year period)buyers, and 1,000 hunters who had reported hunting in this region. Our final sample (n =4,000; aged 18 years or older) was robust enough to test our theoretical models.

Data for this study were collected April–May 2012. We used a modified tailoreddesign-method (Dillman, Smyth, & Christian, 2009) to conduct this research. A non-respondent telephone survey, consisting of a subset of questionnaire items, was admin-istered to a sample of non-respondents (n = 20) after a 2-week waiting period followingthe final postcard. The University Committee on Research Involving Human Subjects atMichigan State University (x12-104e) reviewed and approved methods used in this researchon March 8, 2012.

Questionnaire Development

Latent theoretical factors were measured by multi-item closed-ended (i.e., Likert-type)questions designed to form scales (Dillman et al., 2009) (Table 1). Sociodemographicswere measured by single-item questions. Wildlife value-orientations and attitudes about

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

Perceived Risks from Disease and Management Policies 129

Table 1Latent variables, measurement items, and measurement model, resident deer hunter

respondents, northeast Michigan, 2012

Latentfactors Measurement item

Standardizedfactor loading SE

Personal valuesSelf-enhancement (Cronbach’s α = .70)

Having social power .52 .04Maintaining authority removedPreserving public image .64 .03Being successful .64 .03Being influential .69 .03Being capable removedBeing self-indulgent .40 .04Enjoying life .40 .05

Self-transcendence (Cronbach’s α = .88)Protecting the environment .62 .03Being broadminded removedLoyalty .77 .02Honesty .88 .01Being helpful .85 .01Being responsible .92 .01

Openness to change (Cronbach’s α = .81)Having a varied life .38 .04Exercising freedom .91 .01Exercising independence .88 .01Choosing own goals .82 .02

Conservation (Cronbach’s α = .76)Being respectful .91 .01Exercising self-discipline .76 .02Honoring parents and elders .77 .02Being obedient removedMaintaining social order .43 .04

Wildlife Value Orientations (WVOs)Wildlife Use (Cronbach’s α = .51)

Humans should manage wildlifepopulations so that humans benefit.

.55 .06

The needs of humans should takepriority over fish and wildlifeconservation.

.73 .07

It is acceptable for people to harvestwildlife if they think it poses a threatto their property.

.39 .05

People who want to hunt should beprovided the opportunity to do so.

.07 .05

(continued)

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

130 H. A. Triezenberg et al.

Table 1(Continued)

Latentfactors Measurement item

Standardizedfactor loading SE

Wildlife Protection (Cronbach’s α = .66)I view all living things as part of one big

family..47 .04

Animals should have rights similar to therights of humans.

.37 .04

I take great comfort in the relationships Ihave with animals.

.74 .03

I feel a strong emotional bond withanimals.

.74 .03

Norms (Cronbach’s α = .90)Subjective (hunt antlerless deer)Most [insert group] who I hunt with think I should harvestantlerless deer in northeast MI.

. . . family members . . . .81 .02

. . . friends . . . .92 .01

. . . hunt camp or hunt club members . . . .84 .02Subjective (reduce TB) (Cronbach’s α = .96)Most [insert group] who I hunt with think I should try toreduce bovine TB in northeast MI.

. . . family members . . . .92 .01

. . . friends . . . .99 .00

. . . hunt camp or hunt club members . . . .93 .01Descriptive (perceptions of what others are doing)(Cronbach’s α = .72)Most northeast Michigan deer hunters are . . .

. . . harvesting antlerless deer to managebovine tuberculosis.

.80 .03

. . . taking action to reduce bovinetuberculosis.

.79 .03

. . . doing their part to harvest antlerlessdeer.

.44 .04

. . . doing their part to reduce bovinetuberculosis.

removed

AttitudesDisease Risk Perceptions (Cronbach’s α = .81)I believe that bovine tuberculosis poses a threat . . .

. . . to my health. .44 .03

. . . to my economic livelihood. .64 .03

. . . to hunting in northeast Michigan. .92 .02

. . . to the culture of deer hunting innortheast Michigan.

.85 .02

(continued)

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

Perceived Risks from Disease and Management Policies 131

Table 1(Continued)

Latentfactors Measurement item

Standardizedfactor loading SE

Management of Disease Risk Perceptions (Cronbach’sα = .79)

I believe that bovine tuberculosis management poses athreat . . .

. . . to my health. .30 .04

. . . to my economic livelihood. .66 .02

. . . to hunting in northeast Michigan. .97 .01

. . . to the culture of deer hunting innortheast Michigan.

.90 .01

X2 = 2,102, df = 843, p < .01, X2/df = 2.49, RMSEA = .05, 90% CI RMSEA .00- N.A., CFI =.91.

TB and TB management policies were measured by asking agreement or disagreementwith each statement using a five-point scale. Normative questions were asked in an iden-tical way along with a “don’t know” response option. Respondents were asked to indicateusing a five-point Likert-type scale how very unimportant or very important each of thepersonal value questions were to them. Items from the wildlife-value orientations (WVOs)scale for wildlife use and wildlife protection (Fulton et al., 1996; Teel & Manfredo, 2009)were included in the questionnaire. Personal value items were adapted from Schwartz andBoehnke (2004) and Struch, Schwartz, and van der Kloot (2002).

Attitude questions about perceived risks from TB and its management policies weredeveloped for this questionnaire. We included one risk perception measure for each of therisk object sub-dimensions (e.g., health, economic livelihood, recreation, culture) of theattitudes toward perceived risks from the TB and the management policies of TB basedon Riley, Gore, and Muter (2010). We developed a series of subjective norm items thatreflected the desired behaviors (e.g., harvest antlerless deer, reduce TB) by the social groupsour sample would mostly likely engage with during hunting activities based on Lapinskiand Rimal (2005). We also developed descriptive norm items related to perceptions of othernortheast Michigan deer hunters engaging in the desired behaviors, also based on Lapinskiand Rimal (2005). We pilot tested the questionnaire for face validity and question relevance(Trochim, 2001) and revised accordingly.

Statistical Analysis

We used STATA version 12 (STATA, 2011) for the analyses presented in this article.We used a two-step approach for structural equation modeling, where step one is to testthe measurement models (i.e., confirmatory factor analysis) with the observed variablesloading on latent factors. The second step tests the structural relationship among latentfactors on the parsimonious measurement model (i.e., structural regression) (Kline, 2005).We examined standardized coefficients to identify each indicator’s estimated factor load-ing. The goodness of fit indicators to judge the overall model fit to the data as acceptablewere: Root mean square error of approximation (RMSEA) <.08, 90% confidence intervalfor RMSEA .00 -.10 (or N.A. for upper limit, if lower limit includes .00), and comparativefit index (CFI) >.90 (Kline, 2005; Zajac, Bruskotter, Wilson, & Prange, 2012).

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

132 H. A. Triezenberg et al.

Results

Response Rates and Respondent Characteristics

Our usable response rate for this analysis was 31% (n = 1,228) after adjusting for duplicatesand opt-outs, when a respondent returned their questionnaire but indicated they did notwant to participate in the study (overall response rate was 38%). Comparison of completedquestionnaire and non-respondent telephone survey responses revealed non-respondentshad lower perceived risks of TB management threatening the culture of deer hunting thanrespondents, X2 (1, N = 1228) = 7.62, p < .01.

The majority of our respondents were male (94%) and the mean age was 56 years.Thirty-seven percent of respondents indicated they had some college or technical school,while another 27% indicated they had completed at least an undergraduate degree.Approximately 48% of respondents indicated their household income, before taxes, in2011 was <$59,999 and another 29% indicated their income was $60,000 - $99,999.

Measurement Model

The measurement model fit the data well (CFI = .91, RMSEA = .05, 90% confidenceinterval RMSEA .00 – N.A., X2 = 2202, df = 843, X2 /df = 2.49, p < .01) (Table 1).

Structural Model

The hypothesized conceptual model fit the data well (CFI = .90, RMSEA = .05, 90% confi-dence interval RMSEA .00 – N.A., X2 = 2253, df = 871, p < .01, X2/df = 2.49); however,path coefficients for personal values latent factors were not significant (p < .05), discon-firming H1a -1h, so they were removed to test a parsimonious model (Table 2; Figure 3).The modified model achieved an acceptable fit with our data (CFI = .94, RMSEA = .06,90% confidence interval RMSEA .00 – N.A., X2 = 828, df = 257, p < .01, X2/df =3.22); (Table 2; Figure 4). Risk perceptions of disease management policies were positivelyinfluenced by perceptions of disease risk (in support of H5) and negatively influenced byall three normative factors (in support of H4a-4c). Normative factors of others wanting(subjective) or actually taking action (descriptive) to reduce TB were positively related todisease risk perceptions (in support of H3b-3c); subjective norms of harvest antlerless deerwithout reference to TB were negatively related to disease risk perceptions (in support ofH3a). Wildlife Value orientations positively influenced subjective and descriptive norms (insupport of H2a-2f).

Table 2Structural model goodness-of-fit measures, resident deer hunter respondents, northeast

Michigan, 2012

Model X2 df X2/df p RMSEA 90% CI RMSEA CFI

Hypothesized 2,253 871 2.59 <.01 .05 .00-N.A. .90Modifieda 828 257 3.22 <.01 .06 .00-N.A. .94

aDeleted paths from personal values (Self-enhancement, Self-transcendence, Openness to change,and Conservation) to wildlife value orientations (Wildlife use and Wildlife protection).

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

Perceived Risks from Disease and Management Policies 133

Figure 3. Hypothesized zoonotic disease risk perception model structural results, resident deerhunter respondents, northeast Michigan, 2012. Dashed lines are non-significant pathways at p < .05,solid lines represent significant pathways at p < .05, X2 = 2,253, df = 871, X2/df = 2.59, p < .01,RMSEA = .05, 90% confidence interval RMSEA .00 – N.A., CFI = .90.

Discussion

Zoonotic and emerging infectious diseases threaten ecosystems, economies, cultures, andhuman health worldwide (Friend, 2006; Levi, Kilpatrick, Mangel, & Wilmers, 2012). Thisstudy tests and expands the ZDRISP model to create more systematic approaches to humandimensions research on wildlife-associated disease (Decker et al., 2010). By exploring fac-tors contributing to perceived hazard characteristics and relationships among latent factorsinfluencing disease-related risk perceptions, we improve the ability of human dimensionsinquiry to provide robust and resilient tools to predict and mitigate wildlife disease-relatedcrises, especially when coupled with epidemiological insight.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

134 H. A. Triezenberg et al.

Figure 4. Modified zoonotic disease risk perception model, resident deer hunter respondents, north-east Michigan, 2012. Path coefficients adjacent to causal pathway, z-statistic in parenthesis, ∗p <

.01, ˆ = p < .05, X2 = 828, df = 257, X2/df = 3.22, p < .01, RMSEA = .06, 90% confidenceinterval RMSEA .00 – N.A., CFI = .94.

Our research recognizes and characterizes the duality of attitudes toward objects (i.e.,disease) and attitudes toward situations (i.e., management of disease). This duality iswell known (Rokeach, 1968), but not well attended to in wildlife disease management.Attending to both categories of attitudes (e.g., disease and disease management) will informmore precise development, implementation, and evaluation of wildlife disease–relatedinterventions, and provide more reliable data on which policy alternatives are considered.Explicitly considering stakeholders’ perceptions of disease risks, as well as perceptions ofdisease management risks, can help improve probabilities of achieving desired outcomesbecause managing zoonotic disease and associated societal consequences of disease man-agement necessitates an understanding of human attitudes (Decker et al., 2006; Otupiri,Adam, Laing, & Akanmori, 2000). Incorporation of stakeholders’ risk perceptions of a dis-ease and its management policies may be particularly relevant in places in the world wherecapacity for surveillance, control, and prevention of TB is limited by other social concerns,diseases, or suboptimal communication between scientists and policy makers (Zinsstag,2008).

Perceptions of disease management risks are influenced by perceptions of disease risks,perceptions of what others are doing (i.e., descriptive norms), and perceptions of whatreferent others think an individual should be doing (i.e., subjective norms). Norms serveas indicators of the prevailing conduct, influencing an individual’s perceptions, attitudes,

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

Perceived Risks from Disease and Management Policies 135

and behaviors (Lapinski & Rimal, 2005; Manfredo, 2008). The cognitive hierarchy modelincorporates norms as a predictor of higher order attitudes (Fulton et al., 1996), which inour model was the disease risk perceptions and disease management risk perceptions (i.e.,perceived hazard characteristics). More specifically, norms can lead to cooperation amongmembers of groups (Manfredo, 2008) and influence behavior (Lapinski & Rimal, 2005).Achievement of >60% decrease in TB-prevalence in the core endemic TB area in northeastMichigan since management efforts began as a result of deer hunter cooperation throughantlerless harvests to decrease deer densities and deer populations, but as hunters encounterfewer deer, their willingness to cooperate has declined (O’Brien, Schmitt, Fitzgerald, et al.,2011). Disease management that relies on population reduction, implemented by coopera-tive behavior by volunteer hunters, is likely to encounter similar phenomena as observed inour TB case study. The governance of game species in the United States can effectively reg-ulate hunters to harvest fewer animals but not more. Motivating hunters to harvest more deerlikely will require informative and persuasive communication strategies (O’Brien, Schmitt,Rudolph, et al., 2011; Shanahan, Gore, & Decker, 2012).

From a theoretical perspective, in this study, descriptive and subjective normsabout disease risks were negatively related to management of disease risk perceptions.Perceptions of disease management risks are low when hunters perceive that others aretaking action or want hunters to take action. And, when hunters perceive that others aretaking action to reduce the disease (i.e., descriptive norm) and that others want them toreduce the disease (i.e., subjective norm), disease risk perceptions increase. This is evi-dence of horizontal norm transmission (Manfredo, 2008), colloquially known as “go withthe flow” or “follow the crowd.” In contrast, hunters’ perceptions of others wanting them toharvest antlerless deer were negatively related to perceptions of disease risks and manage-ment of disease risks, signaling that norms related to antlerless deer harvest may supersedeperceptions related to the disease and its management. This may evidence a form of verticalnorm transmission from elders to younger family members (Manfredo, 2008)) or suggesthunters do not consider liberal antlerless deer harvest a management strategy aimed atreducing disease risks. Because norms are significant for disease risk and managementperceptions, linking them to appropriate referent groups, high efficacy actions (Muter et al.,2013), as well as the disease and its management may be an essential component of effec-tive wildlife disease management. It is also possible that for norms to be relevant in zoonoticdisease risk perception models, they need to be specific high efficacy actions (Muter et al.,2013) for each disease and that in our case harvesting antlerless deer action was too general.

Communication programs can be a valuable tool for managers addressing wildlife dis-ease management (Muter et al., 2013), but only if its content addresses risk perceptioninformation needs (Shanahan et al., 2012). Explicitly incorporating perceptions of man-agement policies and norms into activities may be necessary if agencies seek to developand implement successful wildlife-disease management programs, improve public supportfor eradication policies, and maximize communication efforts about management policies.Outreach efforts aimed at creating or solidifying perceptions that others—family, closefriends, or fellow hunters—think one should engage in activities to reduce TB (i.e., sub-jective norms) or perceptions that others are taking action to reduce TB (i.e., descriptivenorms) consistent with disease management policies may help achieve goals. Considerationof audience characteristics is important when selecting mode (e.g., mass media, direct mail,e-mail) for risk communication messages.

Our study revealed perceived risks of disease and disease management are most heav-ily influenced by beliefs about threats to hunting and culture of deer hunting. The endemicTB area has a long history of deer hunting, including hunt camps and clubs dating back

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

136 H. A. Triezenberg et al.

to statehood, likely contributing to hunters’ risk perceptions of TB and TB managementpolicies to hunting and culture of deer hunting. An analysis of outreach materials used inTB management revealed communication efforts focused on TB risks to wildlife, livestock,ecosystems, human health, and economy (Muter et al., 2013; O’Brien, Schmitt, Fitzgerald,et al., 2011). A lack of any mention of risks to hunting and culture of deer hunting, andpossibly explaining why deer hunters’ resentment of management policies have increasedbecause current outreach efforts have not sufficiently addressed the salient issues of risks tohunting and deer hunting culture. Improving hunter cooperation to achieve TB eradicationgoals may depend on directly addressing perceived threats to hunting and culture of deerhunting in the area. Future outreach efforts, including informative or persuasive communi-cation, that directly address perceived risks of a wildlife disease or its management policiesto hunting and culture of deer hunting may influence deer hunters’ understanding of theissue and willingness to engage in liberal deer harvest needed to eliminate this wildlifedisease. These findings may be useful for other areas worldwide, such as wildlife diseasemanagement in Africa, where success of management may depend on engagement withhunters who engage in subsistence harvest or tourism-related trophy hunting (Manfredo,2008; Zinsstag, 2008).

Limitations and Future Research

Our empirical data supporting the proposed zoonotic disease risk perception model is cross-sectional, so we are not able to determine causal inferences among the factors presentedherein. Human dimensions researchers and wildlife disease managers could benefit fromfuture research that seeks to determine causal inferences among factors by being moreinformed in developing their management response. Statewide, approximately 90% of deerhunters in Michigan are male with an average age of 43 years (Frawley, 2012). However,10% of the approximately 700,000 people who purchased a license to hunt deer in Michiganare under 17 years, including 2% under age of 12 years (Frawley, 2012). Our survey onlyaddressed hunters 18 years or older (due to Institutional Review Board [IRB] restrictions),but we consider our respondent demographics comparable to demographics of people whopurchased deer licenses statewide. Public trust responsibilities with wildlife disease man-agement extend beyond those who purchase deer hunting licenses and the bovine TB issue,therefore additional research using the zoonotic disease risk perception framework to avariety of zoonotic-related diseases and with a variety of stakeholders could benefit wildlifemanagers as well as further test this conceptual model. Finally, future research that evalu-ates policies or practices (e.g., communication or outreach) to deer hunters that address riskperceptions and social norms related to wildlife disease management may demonstrate theefficacy of strategies that are informed by human dimensions risk information.

Acknowledgments

This project was supported by the U.S. Department of Agriculture, Michigan Departmentof Natural Resources through the Partnership for Ecosystem Research and Management,and Michigan State University AgBioResearch.

References

Carstensen, M., O’Brien, D. J., & Schmitt, S. M. (2011). Public acceptance as a determinant ofwildlife management strategies for bovine tuberculosis in free-ranging U.S. wildlife. VeterinaryMicrobiology, 151, 200–204. doi:10.1016/j.vetmic.2011.02.046

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

Perceived Risks from Disease and Management Policies 137

Clarke, C. (2009). Seeking and processing information about zoonotic disease risk: A proposedframework. Human Dimensions of Wildlife, 14, 314–325. doi:10.1080/10871200903096155

Decker, D. J., Evensen, D. T., Siemer, W. F., Leong, K. M., Riley, S. J., Wild, M. A., . . . Higgins,C. L. (2010). Understanding risk perceptions to enhance communication about human-wildlifeinteractions and the impacts of zoonotic disease. Institute for Laboratory and Animal Research,51, 255–261.

Decker, D. J., Wild, M. A., Riley, S. J., Siemer, W. F., Miller, M. M., Leong, K. M., .. Rhyan, J. C.(2006). Wildlife disease management: A manager’s model. Human Dimensions of Wildlife, 11,151–158. doi:10.1080/10871200600669908

Dillman, D. A., Smyth, J. D., & Christian, L. M. (2009). Internet, mail, and mixed-mode surveys: Thetailored design method (3rd ed.). Hoboken, NJ: John Wiley & Sons.

Frawley, B. J. (2012). Michigan deer harvest survey report. Lansing, MI: Michigan Department ofNatural Resources.

Friend, M. (2006). Disease emergence and resurgence: The wildlife-human connection. Reston, VA:U.S. Geological Survey.

Fulton, D. C., Manfredo, M. J., & Lipscomb, J. (1996). Wildlife value orientations: A conceptual andmeasurement approach. Human Dimensions of Wildlife, 1, 24–47.

Griffin, R. J., Dunwood, S., & Neuwirth, K. (1999). Proposed model of the relationship of riskinformation seeking and processing to the development of preventative behaviors. EnvironmentalResearch, 80, S230–S245.

Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New York,NY: The Guilford Press.

Lapinski, M. K., & Rimal, R. N. (2005). An explication of social norms. Communication Theory, 15,127–147.

Levi, T., Kilpatrick, A. M., Mangel, M., & Wilmers, C. C. (2012). Deer, predators, and the emer-gence of Lyme disease. Proceedings of the National Academy of Sciences in the United States ofAmerica, 109, 10942–10947.

Maddux, J. E., & Rogers, R. W. (1983). Protection motivation and self-efficacy: A revised theory offear appeals and attitude change. Journal of Experimental Social Psychology, 19, 469–479.

Manfredo, M. J. (2008). Who cares about wildlife? New York, NY: Springer Science + BusinessMedia.

Muter, B. A., Gore, M. L., Riley, S. J., & Lapinski, M. K. (2013). Evaluating bovine tuberculosisrisk communication materials in Michigan and Minnesota for severity, susceptibility and efficacymessages. Wildlife Society Bulletin, 37, 115–121. doi:10.1002/wsb.238

O’Brien, D. J., Fitzgerald, S. D., Lyon, T. J., Butler, K. L., Fierke, J. S., Clarke, K. R., .. Derry,D. E. (2001). Tuberculosis lesions in free-ranging white-tailed deer in Michigan. Journal ofWildlife Disease, 37, 608–613.

O’Brien, D. J., Schmitt, S., Fierke, J. S., Hogle, S. A., Winterstein, S. R., Cooley, T. M., .. Kaneene,J. B. (2002). Epidemiology of Mycobacterium bovius in free-ranging white-tailed deer, Michigan,USA, 1995–2000. Preventative Veterinary Medicine, 54, 47–63.

O’Brien, D. J., Schmitt, S., Fitzgerald, S. D., & Berry, D. E. (2011). Management of bovine tubercu-losis in Michigan wildlife: Current status and near term prospects. Veterinary Microbiology, 151,179–187. doi:10.1016/j.vetmic.2011.02.042

O’Brien, D. J., Schmitt, S., Rudolph, B., & Nugent, G. (2011). Recent advances in the man-agement of bovine tuberculosis in free-ranging wildlife. Veterinary Microbiology, 151, 23–33.doi:10.1016/j.vetmic.2011.02.022

Otupiri, E., Adam, M., Laing, E., & Akanmori, D. B. (2000). Detection and management of zoonoticdisease at the Kumasi slaughterhouse in Ghana. Acta Tropica, 76, 15–19.

Riley, S. J., Gore, M. L., & Muter, B. A. (2010). Expert perspectives on bovine tuberculosis manage-ment policies in Michigan and Minnesota. East Lansing, MI: Michigan Agricultural ExperimentStation.

Rokeach, M. (1968). Beliefs, attitudes and values: A theory of organization and change. SanFrancisco, CA: Jossey-Bass.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014

138 H. A. Triezenberg et al.

Schwartz, S. H., & Boehnke, K. (2004). Evaluating the structure of human values with confir-matory factor analysis. Journal of Research in Personality, 38, 230–255. doi:10.1016/S0092-6566(03)00069-2

Shanahan, J. E., Gore, M. L., & Decker, D. J. (2012). Communication for effective wildlife man-agement. In D. J. Decker, S. J. Riley, & W. F. Siemer (Eds.), Human dimensions of wildlifemanagement (2nd ed.) (pp. 157–173). Baltimore, MD: Johns Hopkins University Press.

STATA. (2011). STATA (Version 12) [Computer software]. College Station, TX: StataCorp.Struch, N., Schwartz, S. H., & van der Kloot, W. A. (2002). Meanings of basic values for women

and men: A cross-cultural analysis. Personality and social psychology bulletin, 28, 16–28.doi:10.1177/0146167202281002

Teel, T. A., & Manfredo, M. J. (2009). Understanding the diversity of public interests in wildlifeconservation. Conservation Biology, 24, 128–139. doi:10.1111/j.1523-1739.2009.01374.x

Trochim, W. M. K. (2001). The research methods knowledge base. Mason, OH: Atomic DogPublishing.

Vaske, J. J., Shelby, L. B., & Needham, M. D. (2009). Preparing for the next disease: The human-wildlife connection. In M. J. Manfredo, J. J. Vaske, P. J. Brown, D. J. Decker, & E. A. Duke(Eds.), Wildlife and society: The science of human dimensions (pp. 244–261). Washington, DC:Island Press.

Witte, K., & Allen, M. (2000). A meta-analysis of fear appeals: Implications for effective publichealth campaigns. Health Education & Behavior, 27, 591–615.

Wobeser, G. (2002). Disease management strategies for wildlife. Revue Scientifique Et Technique DeL Office International Des Epizooties, 21, 159–178.

Zajac, R. M., Bruskotter, J. T., Wilson, R. S., & Prange, S. (2012). Learning to live with black bears:A psychological model of acceptance. The Journal of Wildlife Management, 76, 1331–1340.doi:10.1002/jwmg.398

Zinsstag, J. (2008). Towards effective control of bovine tuberculosis in Africa: A case for publicengagement in science. Ethiopian Journal of Health Development, 22, 117–118.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

10:

40 1

5 N

ovem

ber

2014