Human Perspectives and Conservation of Grizzly Bears in Banff National Park, Canada

Conservation Practice and Policy

Human Perspectives and Conservation of GrizzlyBears in Banff National Park, CanadaEMILY C. CHAMBERLAIN,∗ MURRAY B. RUTHERFORD,∗‡ AND MICHAEL L. GIBEAU†∗School of Resource and Environmental Management, Simon Fraser University, 8888 University Drive, Burnaby, BC V5A 1S6, Canada†Parks Canada, Mountain National Parks, Box 213, Lake Louise, Alberta T0L 1E0, Canada

Abstract: Some conservation initiatives provoke intense conflict among stakeholders. The need for action,the nature of the conservation measures, and the effects of these measures on human interests may be disputed.Tools are needed to depolarize such situations, foster understanding of the perspectives of people involved,and find common ground. We used Q methodology to explore stakeholders’ perspectives on conservation andmanagement of grizzly bears (Ursus arctos horribilis) in Banff National Park and the Bow River watershedof Alberta, Canada. Twenty-nine stakeholders participated in the study, including local residents, scientists,agency employees, and representatives of nongovernmental conservation organizations and other interestgroups. Participants rank ordered a set of statements to express their opinions on the problems of grizzlybear management (I–IV) and a second set of statements on possible solutions to the problems (A–C). Factoranalysis revealed that participants held 4 distinct views of the problems: individuals associated with factor Iemphasized deficiencies in goals and plans; those associated with factor II believed that problems had beenexaggerated; those associated with factor III blamed institutional flaws such as disjointed management andinadequate resources; and individuals associated with factor IV blamed politicized decision making. Therewere 3 distinct views about the best solutions to the problems: individuals associated with factor A called forincreased conservation efforts; those associated with factor B wanted reforms in decision-making processes;and individuals associated with factor C supported active landscape management. We connected people’sdefinitions of the problem with their preferred solutions to form 5 overall problem narratives espoused bygroups in the study: the problem is deficient goals and plans, the solution is to prioritize conservation efforts(planning-oriented conservation advocates); the problem is flawed institutions, the solution is to prioritizeconservation efforts (institutionally-oriented conservation advocates); the problems have been exaggerated,but there is a need to improve decision-making processes (optimistic decision-process reformers); the problemshave been exaggerated, but managers should more actively manage the landscape (optimistic landscapemanagers); and the problem is politicized decision making, solutions vary (democratizers). Although these 5groups differed on many issues, they agreed that the population of grizzly bears is vulnerable to extirpation,human use of the area should be designed around ecological constraints, and more inclusive decision-makingprocesses are needed. We used our results to inform a series of workshops in which stakeholders developed andagreed on new management strategies that were implemented by Parks Canada. Our research demonstratesthe usefulness of Q method to illuminate people’s perspectives and identify common ground in settings whereconservation is contested.

Keywords: Banff National Park, conservation policy, human perspectives, Q methodology, Ursus arctos

Perspectivas Humanas y la Conservacion de Osos Grizzli en el Parque Nacional Banff, Canada

Resumen: Algunas iniciativas de conservacion provocan conflictos intensos entre los diferentes actores. Lanecesidad de actuar, la naturaleza de las medidas de conservacion y los efectos de estas medidas sobre losintereses humanos pueden estar en disputa. Se requieren herramientas para despolarizar tales situaciones,

‡email [email protected] submitted May 13, 2011; revised manuscript accepted October 26, 2011

420Conservation Biology, Volume 26, No. 3, 420–431C©2012 Society for Conservation BiologyDOI: 10.1111/j.1523-1739.2012.01856.x

Chamberlain et al. 421

promover el entendimiento de las perspectivas de la gente involucrada y encontrar terreno comun. Utilizamosla metodologıa Q para explorar las perspectivas de diferentes actores sobre la conservacion y manejo deosos grizzli (Ursus arctos horribilis) en el Parque Nacional Banff y en la cuenca del Rıo Bow en Alberta,Canada. Veintinueve actores participaron en el estudio, incluyendo residentes locales, cientıficos, empleadosde agencias y representantes de organizaciones de conservacion no gubernamentales y otros grupos de interes.Los participantes ordenaron una serie de afirmaciones para expresar sus opiniones sobre los problemas delmanejo de osos grizzli (I–IV) y un segundo conjunto de afirmaciones sobre posibles soluciones a los problemas(A–C). El analisis de factores revelo que los participantes mantenıan 4 visiones distintas de los problemas:individuos asociados con el factor I enfatizaron las deficiencias de las metas y planes; aquellos asociados con elfactor II consideraron que los problemas habıan sido exagerados; aquellos asociados con el factor III culparona los defectos institucionales como el manejos descoordinado y recursos inadecuados; todos los individuosasociados con el factor IV culparon a la toma de decisiones politizada. Hubo 3 visiones diferentes respectoa las mejores soluciones a los problemas: individuos asociados al factor A solicitaron mayores esfuerzos deconservacion, aquellos asociados con el factor B querıan reformas en los procesos de toma de decisiones; ylos individuos asociados con el factor C apoyaron el manejo activo del paisaje. Conectamos las definicionesdel problema con las soluciones preferidas para formar 5 narrativas generales de problemas adoptadaspor los grupos del estudio: el problema es metas y planes deficientes, la solucion es priorizar los esfuerzosde conservacion (partidarios de la conservacion orientada por planificacion); el problema es institucionesdefectuosas, la solucion es priorizar los esfuerzos de conservacion (partidarios de la conservacion orientadainstitucionalmente); los problemas han sido exagerados, pero existe la necesidad de mejorar los procesos detoma de decisiones (reformadores de procesos de decision optimistas); los problemas han sido exagerados, perolos gestores deben manejar el paisaje mas activamente (manejadores de paisaje optimistas); y el problemaes la toma de decisiones politizada, las soluciones varıan (democratizadores). Aunque estos cinco gruposdifirieron en muchos aspectos, estuvieron de acuerdo en que la poblacion de osos grizzli es vulnerable a laextirpacion, que se debe disenar el uso humano del area tomando en cuenta limitaciones ecologicas y que serequieren procesos de toma de decisiones mas incluyentes. Utilizamos nuestros resultados para informar unaserie de talleres en los que los actores desarrollaron y acordaron nuevas estrategias de manejo que fueronimplementados por Parks Canada. Nuestro estudio demuestra la utilidad del metodo Q para iluminar lasperspectivas de la gente e identificar terreno comun en escenarios donde la conservacion es impugnada.

Palabras Clave: metodologıa Q, Parque Nacional Banff, perspectivas humanas, polıticas de conservacion, Ursusarctos

Introduction

Efforts to conserve species and ecosystems often takeplace in highly politicized settings, where participantswith differing values, attitudes, and beliefs clash over poli-cies and where existing institutions lack the capacity toresolve conflicts. The species, ecosystem, or policy maybecome the symbolic focus for disputes that are rootedin other political and social disagreements (Clark et al.2005). To develop effective and enduring conservationprograms, better tools are needed to depolarize such situ-ations, foster understanding of the different perspectivesof the people involved, and find common ground.

We used Q methodology to engage stakeholders inthe exploration and discussion of their perspectives onthe controversial problem of conserving and managinggrizzly bears (Ursus arctos horribilis) in Banff NationalPark and the Bow River Watershed of Alberta, Canada.Q methodology, which allows people to map their sub-jective perspectives on a topic, is being used increas-ingly by environmental policy scholars (e.g., Addams &Proops 2000; Cuppen et al. 2010; Sandbrook et al. 2011).We used Q methodology to illuminate people’s diver-gent understandings of a conservation problem and toidentify areas of potential agreement among stakehold-

ers. We used our results to inform a series of multistake-holder, interdisciplinary problem-solving workshops inwhich participants developed and agreed on several griz-zly bear conservation policies that had been contestedpreviously (Rutherford et al. 2009).

Grizzly Bears in the Banff-Bow Valley

The Banff-Bow Valley is in the southeastern portion of theCentral Rocky Mountains ecosystem, a contiguous area ofabout 40,000 km2 in southwestern Alberta and southeast-ern British Columbia (Gibeau 2000; Parks Canada 2007).The number of grizzly bears in the Central Rockies ecosys-tem is estimated at between 400 and 500 (Herrero et al.2000), and about 60 of these bears occur in Banff Na-tional Park (Gibeau et al. 1996). The total number ofgrizzly bears in Alberta has decreased from an estimatedpre-European 6000 (Eastern Slopes Grizzly Bear Project1998) to a recent estimated <700 (ASRD & ACA 2010).

There is a substantial amount of biological data on griz-zly bears in the Banff-Bow Valley due largely to the workof the Eastern Slopes Grizzly Bear Project, a major re-search program that began in 1994 to study grizzly bearsin the Central Rockies (Herrero 2005). This project in-cluded a 9-year study (1994–2002) of grizzly bear de-mography in the Bow River watershed (Garshelis et al.

Conservation BiologyVolume 26, No. 3, 2012

422 Conservation of Grizzly Bears

2005a). The results of the research show that despitehuman development, bear abundance likely increasedslightly over the study period, although a slight decreasein abundance is also within the 95% CI. The long-termviability of the grizzly bear population is vulnerable tostochastic events and decreases in the number of repro-ductive females (Garshelis et al. 2005a, 2005b).

Although much is known about the grizzly bears ofthe Banff-Bow Valley, management policies have beencontroversial. Policy makers have struggled to reconcileconflicting demands for bear conservation, commercialdevelopment, and recreational use. The Banff-Bow Valleyis one of the most developed areas where grizzly bearsoccur, and humans have been the primary cause of mor-tality of grizzly bears in this region since 1980. The rail-way and highway were the largest sources of grizzly bearmortality between 1990 and 2009 (Gibeau 2000; Benn &Herrero 2005; Bertch & Gibeau 2010). Some stakehold-ers believe the Eastern Slopes Grizzly Bear Project pro-duced sufficient scientific evidence to conclude that griz-zly bears will not persist in the valley unless conditionsare changed substantially to reduce bear mortality (e.g.,Bow Valley Grizzly Bear Alliance 2002). Others argue thatthe scientific research was flawed (e.g., D. Leighton, un-published), the bear population is stable and not likelyto be extirpated, and human use and enjoyment of Cana-dian national parks are being unduly limited (Cooper etal. 2002). We explored these and other perspectives.

Methods

In Q methodology, participants “map” their perspectivesby rank ordering a set of statements that express a rangeof views on the subject being investigated (referred to as aQ sort). Typically statements are sorted from “most agree”to “most disagree” or “most like my view” to “most unlikemy view.” These Q sorts are correlated and analyzed toidentify groups of individuals who sort the statements insimilar ways, indicating a shared viewpoint. The analysesalso highlight areas of agreement and disagreement. Fora thorough introduction to Q methodology, see Brown(1980), McKeown and Thomas (1988), and Brown et al.(1999). For previous applications of this method to inves-tigate perceptions about conservation of large carnivores,see Byrd (2002) and Mattson et al. (2006).

Twenty-nine individuals took part in our study, includ-ing wildlife biologists, local business operators, tourismrepresentatives, environmental activists, local residents,industry representatives, employees of nongovernmentalorganizations, and staff of federal and provincial agencies(Table 1). We selected participants with targeted non-random sampling (Brown 1980). The goal in selectingparticipants for a typical Q study is not to select a statis-tically representative sample of a population, but insteadto ensure that a comprehensive range of views about the

subject is represented and perhaps to include influentialvoices. It is not possible with Q methodology alone todetermine what percentage of the population holds eachview, but the method can provide a rich understandingof the subjective views of the participants involved. Themethod is especially useful when participants are keyindividuals involved in a particular community matter.

We developed 2 separate sets of statements. The firstset dealt with the problems associated with grizzly bearmanagement in the Banff-Bow Valley and the seconddealt with potential solutions. We drew statements fromsemistructured interviews conducted with participantsseveral weeks before the Q-sorting exercise. Using theprinciples of variance design (Fisher 1960; Brown 1970),we selected 38 statements that captured the range ofidentified problems and 30 statements that captured therange of identified solutions (Tables 2 & 3). Each state-ment was printed on a separate index card.

Participants were instructed to rank order the state-ments into 9 columns ranging from “most unlike mypoint of view” (−4) to “most like my point of view”(+4), thereby forming an inverted quasi-normal distribu-tion in which few statements were placed in the extremecategories (−4, −3, +4, +3) and many statements wereconcentrated in the middle categories (−2 to +2) (Sup-porting Information). We also conducted a follow-up in-terview with each participant to explore their rationalefor sorting the statements as they did.

We used the software PQMethod (Schmolck & Atkin-son 2002) to calculate correlations among the Q sorts,conduct a principal components analysis on the correla-tion matrices, and rotate the factors (varimax rotation).The software also constructs a model Q sort, or factor ar-ray, for each factor, which is a weighted representation ofthe sorts of those participants who are highly associatedwith (“loaded on”) that factor and not with other factors.We adopted an alpha of 0.01 to determine the statisticalsignificance of factor loadings and factor scores.

Following Mattson et al. (2006), we defined statementsof positive virtual consensus among factor groups as stan-dardized scores of ≥0 for all factors and at least onestandardized score >+2 and negative virtual consensusas standardized scores of ≤0 for all factors and at leastone standardized score <−2. Virtual consensus state-ments represent potential common ground, where thereis some agreement across views and at least one factorgroup feels strongly about the statement.

One month after the Q sorts were completed, we helda workshop with participants to discuss the preliminaryresults. All participants in the Q study were invited to theworkshop, and 15 attended, including at least one rep-resentative of each view identified in the Q study. Theworkshop discussions and our interviews with respon-dents after the Q sorts helped us to understand why par-ticipants sorted the statements as they did and informedour final interpretation of their views.



Table 1. Study participants (denoted by affiliation where consent allowed) and their factor loadingsa for the Q sorts related to problems andsolutions for grizzly bear management in Banff-Bow Valley, Alberta, Canada.

Problems factors Solutions factors

Participant affiliationb I II III IV A B C

Participants associated with solutions factor A(conservation advocates)

Government of Alberta 76∗# −76 23 25 64

∗# −54 05year of the Great Bear (NGO) 75

∗# −75 40 −01 72∗# 18 −29

Parks Canada 78∗# −78 28 29 38 03 −20

environmental organization 72∗ −72 51

∗07 85

∗# −17 −03environmental organization 79

∗ −79 44∗ −13 82

∗# −11 −36Government of British Columbia 55

∗ −55 55∗

03 77∗# −09 −01

Canadian Parks and Wilderness Society 64∗ −64 55

∗03 80

∗# −04 00Yellowstone to Yukon Conservation Initiative 76

∗ −76 48∗

01 72∗# −38 −24

Parks Canada 72∗ −72 49

∗06 70

∗# 13 −18Arc Wildlife Services Ltd. 28 −28 75

∗# 07 80∗# −04 25

anonymous 10 −10 70∗# 42 58

∗# 44 11University of Calgary 17 −17 83

∗# −07 82∗# −04 −16

Government of British Columbia 25 −25 70∗# −08 82

∗# −02 −20Parks Canada 11 −11 73

∗# 14 76∗# 31 03

Government of Alberta 38 −38 26 45∗# 82

∗# −11 −23Tourism Canmore 12 −12 20 39 56

∗# 13 16commercial business 41 −41 40 55

∗# 47∗

47∗ −09

Participants associated with solutions factor B(decision-process reformers)

Shell Canada 63∗# −63 24 18 36 63

∗# −07Alberta Beef Producers −46 46

∗# −18 40 −08 78∗# 05

National Park Ski Areas Association −76 76∗# 15 18 −35 63

∗# 29commercial business −47 47

∗53

∗31 −01 72

∗# 28anonymous −62 62

∗ −16 46∗ −49 64

∗# −06commercial business −64 64

∗ −02 42∗ −34 46 13

Parks Canada 33 −33 79∗# 20 39 48

∗# −24Parks Canada −15 15 50

∗56

∗01 71

∗# 07community resident −05 05 05 39 17 57

∗# 10

Participants associated with solutions factor C(landscape managers)

anonymous −63 63∗# −40 39 −50 45 54

∗#

commercial business −78 78∗# 22 11 −13 04 90

∗#

anonymous −30 30 −44 78∗# 01 22 74

∗#

aFactor loadings indicate the extent to which a participant’s Q sort is similar or dissimilar to the model Q sort (or “factor array”), which isthe weighted representation of the sorts of those participants who are highly associated with that factor and not with other factors. Loadingsof participants significantly associated with a factor (p < 0.01) are marked with a ∗. Pure factor representations (participants significantlyassociated with only one factor) are marked with #. Decimals to two places are omitted.bParticipants are grouped according to their loadings on the solutions factors.

Results

We identified 4 main views on problems with grizzly bearmanagement (problems factors I−IV) and 3 main viewson solutions to the problems (solutions factors A−C).Each factor represents a group of participants who sortedthe statements similarly as a function of their similar un-derstanding of problems or solutions. Factor loadings (Ta-ble 1) indicate the extent to which an individual’s Q sortis similar to the model Q sort constructed to representthe factor (the factor array). Factor scores show howeach statement is ranked in the model factor array foreach factor (Tables 2, 3, & Supporting Information).

Problems Sorts

For the problems sorts we initially extracted 3 factors,one of which was bipolar. A bipolar factor includes 2directly opposing views that are represented at oppositeends of the same factor. We separated individuals withhigh negative loadings on this factor from those with highpositive loadings to distinguish problems factors I and II.The remaining factors became factors III and IV.

PROBLEMS FACTOR I, DEFICIENT GOALS AND PLANS

Participants associated with factor I were concernedabout the status of grizzly bears in the Banff-Bow Val-



Table 2. Problems Q sample statements and factor scoresa for factors I (deficient goals and plans), II (exaggerated problems), III (institutionalflaws), and IV (politicized decision making) in a Q-method examination of problems and solutions for grizzly bear management in Banff-BowValley, Alberta, Canada.

Factorb

Statement I II III IV

P1. There is a false crisis mentality spurred by interest groupswho have more in line than the health and welfare of grizzlybears.

−2 +2 −2 +3

P2. There is a lack of an overall conservation strategy forgrizzly bears, lack of clear goals, targets, and a bigger vision.

+3 −3 −1 0

P3. There is not enough funding to implement what we knowneeds to be done for grizzly bear management.

0 −1 +3 −4

P4. The grizzly bear population is vulnerable. +1 0 +4 +2P5. The grizzly bear population of the Banff-Bow Valley is the

healthiest it has been in 25 years.0 +3 +1 +1

P6. People management in Banff Park has been successful andhas led to us cultivating bears not wiping them out.

−3 +3 0 −1

P7. The current management of grizzly bears is somewhatdisjointed between several different responsible agencies.Techniques to manage bears are not consistent andcommunication is not as good as it could be between theseagencies.

+2 −3 +3 0

P8. An unrelenting tide of humanity has descended on a placethat has a finite capacity to accommodate human pressure.

+4 −4 0 +4

P9. The grizzly bear population is at an equilibrium, it’sneither dropping nor increasing. Management is doing agood job with what they’re working with.

−2 +2 +1 −1

P10. We tend to get caught up in the chicken littlesyndrome—thinking that the sky is falling and we need tofix everything—without recognizing Parks Canada’ssuccesses in grizzly bear management.

−1 +4 +1 +1

P11. The squeaky wheel wins in grizzly bear management.Organizations that speak loudly and are connected to themedia have their views incorporated into policy.

0 0 −2 +1

P12. Although human use in Banff Park has increased, that useis more concentrated and people are better educated, sopeople are having less of an impact on grizzly bears.

−1 +3 −1 −2

P13. Increasing human use of grizzly bear habitat, throughrecreational use, residential use, and tourism development,both inside and outside of the Park has resulted in increasedmortality rates of grizzly bears.

+2 −3 +4 −2

P14. There will be more challenges for residents with bearactivity intruding in communities in the future.

+4 +1 +1 +3

P15. The population status of grizzly bears is not sustainable inthe long term. If we sit back today and call it acceptable, wewon’t make the improvements that need to be made tomaintain the position we’re in now.

+3 −1 +3 +1

P16. The Bow Valley is an important linkage for the regionalgrizzly bear population. If we lose the connections andopportunities in this area, then there is a high risk of thepopulation being placed in jeopardy.

+4 0 +4 −2

P17. The precautionary principle doesn’t hold water in grizzlybear management. The onus of proof is still on thosedefending wildlife instead of on developers.

+2 −1 +2 −3

P18. Management is fragmented by jurisdiction. There are nosystem wide specific objectives that Parks Canada and theprovincial agencies are trying to manage for.

+2 −4 +2 −1

P19. There is no well organized or visionary plan in place thatoutlines when success is achieved in management andwhen we’ve achieved a healthy population.

+3 −2 −2 +3

P20. Banff Park doesn’t have room for more bears because theecosystem in the Park is at carrying capacity.

−1 −1 −4 −1

continued



Table 2. (continued.)

Factorb

Statement I II III IV

P21. We’re taking our local situation with bears andextrapolating. In the regional context, grizzly bearpopulations are healthy.

−4 +1 −3 +1

P22. Elsewhere in Alberta, grizzly bear populations areshrinking. The Bow Valley needs to be a source of bears toincrease the regional population of bears.

0 −2 +1 −4

P23. Decisions are made with urban perceptions and bywildlife groups, with less consideration given to agriculture.Livestock producers have generally borne the costs ofgrizzly bear protection and do not get adequatecompensation for losses incurred by bears.

−1 0 −1 −3

P24. We have unnecessarily sacrificed human activities inBanff National Park for grizzly bear protection.

−4 0 −3 −2

P25. We are on a trend to having way too many bears in thearea, which means we’ll be bound to have more problemsbetween bears and people and a huge proportion ofhabituated bears.

−1 +2 −4 0

P26. The population status of grizzly bears is acceptable as itis. We’ll never achieve zero mortality of bears given thecircumstances we’re in and that’s fine.

−3 +2 0 +2

P27. Politics and special interest pleading have interfered withthe essential scientific understanding of the fundamentalestablished mandates of conservation organizations. We’reno longer talking about science, we’re talking about whocan speak the loudest and who can get the most mediacoverage.

+1 −3 −1 +2

P28. It is not the role of a National Park to be a bear factoryand produce bears, but instead to have the right amount ofbears for the Park itself.

−2 +4 0 +3

P29. Management is largely reactive, it’s based on the politicalbureaucratic mood of the day and is not entirely sciencebased.

+3 −2 0 +2

P30. A disproportionate amount of resources are going intosaving bears when they’re shot just outside Park borders.

−2 0 −3 −4

P31. Human use issues receive greater priority in Parksmanagement to the point where grizzly bears have beenjeopardized.

+1 −4 0 −3

P32. Grizzly bears have been over managed. The trend ofclosing each area with a female grizzly in it is leading us toclose Banff.

−3 +1 −4 −1

P33. Political pressure lets people get what they want.Decision making is politicized.

+1 +1 +2 +4

P34. Most of the discourse associated with policy making hasbeen high-jacked by people whose views are short term anddo not take into account the larger interests, sensibilities orhistory of this country.

0 −2 −3 0

P35. If something will impact recreational opportunities, theburden of proof is always on the bear, their habitat, and thepeople who defend their habitat, to show that harm is beingdone. This is wrong.

+1 −1 +2 −3

P36. Instead of celebrating our achievements in grizzly bearmanagement, we continue to talk about our challenges.

0 +4 +3 +2

P37. Grizzlies are managed from the perspective that they’rean endangered species when they’re not. The Banff-BowValley is not the last stand of the grizzly bear.

−3 +3 −1 0

P38. The grizzly bear population is doing very well, describingthe population as just “stable” is the crisis version of what ishappening.

−4 +1 −2 0

aFactor scores show how each statement was ranked (+4 to –4) in the factor array (model Q sort) for each factor.bThe problems factors represent a group of participants who sorted the statements similarly as a function of their similar understanding of theproblem of grizzly bear management in the Banff-Bow Valley.



Table 3. Solutions Q sample statements and factor scoresa for factors A (conservation advocates), B (decision-process reformers), and C(landscape managers) in a Q-method examination of problems and solutions for grizzly bear management in Banff-Bow Valley, Alberta, Canada.

Factorb

Statement A B C

S1. Restricting human use doesn’t have to be the answer. Human use has already been restricted in the areasmost important for grizzly bears and we don’t need more restrictions.

−4 +3 −3

S2. Tighten the integration of scientific management and research. Management actions should be directlycoupled to the outcomes of research.

0 +2 −4

S3. Limit growth on provincial lands adjacent to the Park. +2 −3 0S4. Make bears a higher priority in provincial management. In Alberta, create bold, legally accountable legislation

that makes government manage for the needs of grizzly bears.+3 −2 −1

S5. Use less invasive research on grizzly bears and strictly monitor the population. Research is not a mandate forNational Parks, and parks are not a lab.

−4 0 +1

S6. Use science more to guide policy decisions. +2 +3 −1S7. National Parks are not game preserves, they should be managed for people to come here to see and learn

things.−3 0 −2

S8. Increase participation and communication with park residents. −1 +1 +1S9. When management closes one area of the Park for grizzly bear management, they have to open another area

for recreational opportunity.−4 −1 +4

S10. Create bear habitat in wilderness areas in the backcountry, outside of communities and development areas,to keep bears and people separate.

−1 +4 +3

S11. Develop specific objectives for each habitat area. Figure out how many bear deaths can be tolerated in eacharea (demographic target), and how much habitat change is acceptable.

+1 −2 +3

S12. Build an appreciation for grizzly bears among recreational users. The issue of management comes down tomanaging people.

+2 +1 +3

S13. Keep collaring and drugging bears to a minimum because these techniques completely change a bear’sbehavior and then you’re no longer studying wild bears. This is the bear’s National Park too.

−3 −1 +2

S14. Restrict human use in the Park, create areas where bears can live on the landscape and meet their yearround needs.

+3 −4 −2

S15. Develop a more formal process between Parks Canada and the provincial agencies for managing bears bydeveloping a multiagency group to deal with grizzly bear management that has some power to influencedecisions.

+4 +4 −4

S16. We need to keep in mind the historical context for ecological integrity. People think that Banff NationalPark is Eden, but in fact Banff history was for tourism.

−3 −1 +1

S17. Scientists and decision makers should be clearer about what the science indicates is in the interest of bears. 0 +2 −1S18. Focus on monitoring trends of the grizzly bear population in scientific research, and finding less intrusive

ways to do so.−1 +2 +2

S19. Engage landowners in decisions. Get more input from people out on the land who are actually seeing thewildlife on a more regular basis.

0 0 +1

S20. Parks Canada must take a stronger stance toward prioritizing ecological integrity in Banff National Park. +3 −3 0S21. We need a more concerted management effort between the province, Parks Canada, industry, and people

who do things on the land.+4 +4 −2

S22. We need to change our value system and value other things besides profit if we want bears on thelandscape. We are compromising our long term well-being for short term material gains of wealth and power.

+1 −4 −3

S23. Find ways so that humans and grizzly bears can co-habitate in the same ecosystem by minimizing bearhabituation. Our biggest mistake in management has been to designate separate spaces for bears and humans.

−2 0 0

S24. Managers should say outright that the function of a National Park is a conservation function. Someone needsto say no to the next round of development expansion.

+1 −4 −4

S25. Design human use around ecological constraints. +4 +1 +3S26. Increase habitat in the Park for bears so that less bears move onto the plains and come into conflict with

agricultural operations.−2 −1 −3

S27. Improve the communication structure between various parties that have a role to play in grizzly bearprotection. Develop a standardized protocol for information sharing between organizations.

+1 +1 0

S28. Change the configuration of habitat to reduce the potential for conflict between humans and bears, such asgetting rid of high-quality bear habitat near human development.

−2 −2 +3

S29. Adjust values and attitudes so that people value a live bear so highly that they wouldn’t cause thecircumstances of that bear’s death.

−1 −3 −1

S30. Find a more effective way of including interests, not just those who are loud, but where prudence andunderstanding drive the logic and argument, not just passion.

0 +3 +2

aFactor scores show how each statement was ranked (+4 to –4) in the factor array (model Q sort) for each factor.bThe problems factors represent a group of participants who sorted the statements similarly as a function of their views about possible solutionsto the problem of grizzly bear management in the Banff-Bow Valley.



ley and emphasized that the population was not sustain-able (statements P15, P26, P37, P38). These individu-als believed inadequate planning and poor goal settingwere the roots of the problem, including reactive man-agement that was not based on science (P29) and lackof an overall conservation strategy or visionary plan andclear goals and targets (P2, P19). They rejected the claimthat management of human use had been successful (P6)and instead considered increased human use as problem-atic (P8). They also expected problems with grizzly bearactivity in human communities to increase (P14).

PROBLEMS FACTOR II, EXAGGERATED PROBLEMS

Participants associated with factor II believed that thegrizzly bear population was healthy and management ofbears had been successful, but that problems with the sta-tus of bears had been overemphasized (P37) and achieve-ments had not been recognized or celebrated sufficiently(P10, P36). They also believed human use in Banff Na-tional Park was well managed (P6, P12). They stronglydisagreed with criticisms of management, such as thatmanagement was fragmented across agencies and juris-dictions (P7, P18), or that there was a lack of adequateconservation objectives (P2), or that unrelenting humanuse had increased mortality rates of bears (P8, P31).

PROBLEMS FACTOR III, INSTITUTIONAL FLAWS

Individuals associated with factor III shared many of theconcerns of those associated with factor I about the sta-tus of grizzly bears and the effects of human use ongrizzly bears and their habitat (P4, P13, P15, P16, P21,P24). However, they did not identify poor planning andgoal setting as the most important aspects of the prob-lem (P2, P19, P29). Instead, these participants blamedinadequate funding (P3) and other flaws in institutionalarrangements, such as disjointed management and poorcommunication (P7). Their concern about the status ofthe grizzly bear population was accompanied by a beliefthat the park could support more bears (P20, P25). De-spite their concerns about grizzly bears and institutions,they agreed that achievements had not been celebrated(P36).

PROBLEMS FACTOR IV, POLITICIZED DECISION MAKING

Participants associated with factor IV emphasized thepoliticization of grizzly bear management. They believedthat decisions were unduly influenced by political pres-sure from interest groups (P33), that these groups hadfalsely promoted a crisis mentality on the basis of objec-tives other than the welfare of bears (P1), and that man-agement lacked a visionary plan and criteria for assessingsuccess (P19). But they rejected claims that human-useissues were assigned too high a priority (P31) and that

the precautionary principle was not being applied in griz-zly bear management (P17). These participants were alsoconcerned about increased human use of a place with fi-nite capacity to accommodate human pressure (P8) andincreasing challenges for residents due to grizzly bear ac-tivity in human communities (P14). Their concerns wereassociated with a belief that the Banff-Bow Valley shouldnot be managed as a source of grizzly bears to increasethe abundance of the regional population (P22, P28).They also strongly rejected statements that identifiedthe amount of funding and resources for managementas problems (P3, P30).

VIRTUAL CONSENSUS ABOUT PROBLEMS

For the problems sort, there was virtual consensus thatthe grizzly bear population was vulnerable, but health-ier than in the past (P4, P5); there would be contin-uing problems with bears in human communities, butthe park could support more bears (P14, P20); decisionmaking was politicized (P33); and past achievementshad not been celebrated sufficiently (P36) (SupportingInformation).

Solutions Sorts

SOLUTIONS FACTOR A, CONSERVATION ADVOCATES

We called individuals associated with factor A conserva-tion advocates because their preferred solutions empha-sized conservation of grizzly bears. This group supportedrestricting human use in the area, designing human usearound ecological constraints, reducing recreational op-portunities if necessary for bear management, and cre-ating areas where bears can live on the landscape andmeet their needs (S1, S25, S9, S14). Conversely, they op-posed the idea that management should prioritize hu-man uses (S7, S16) and supported giving higher priorityto grizzly bears and ecological integrity (S20, S4). Theythought there was an opportunity for increasing the levelof collaboration among management agencies and othergroups to support bear conservation (S21, S15). Unlikeother perspectives in our Q study, these conservationadvocates strongly disagreed that there was a need tochange bear-research techniques (e.g., by eliminating ra-dio collaring of bears) (S5, S13).

SOLUTIONS FACTOR B, DECISION-PROCESS REFORMERS

Individuals associated with factor B, the decision-processreformers, advocated changes to decision-making pro-cesses. They called for decision making that effectivelyincludes interests and encourages reasoned argument(S30), is guided by scientific research (S2, S6), and in-volves greater collaboration among government agen-cies and other interests (S15, S21). Unlike the conser-vation advocates, the decision-process reformers did not



believe ecological integrity and conservation should begiven higher priority than development and other objec-tives (S20, S24) and did not want human use of the arearestricted further (S14, S1). The decision-process reform-ers did not believe human values and attitudes neededto be adjusted for the grizzly bear population to be main-tained (S22, S29). This group also supported creating bearhabitat outside developed areas (S10).

SOLUTIONS FACTOR C, LANDSCAPE MANAGERS

We called individuals associated with factor C landscapemanagers because they recommended active manage-ment of the landscape to keep grizzly bears and peo-ple separate, such as reducing the extent of bear habitatnear human development (S28), increasing the extent ofbear habitat in the backcountry (S10), opening new ar-eas for recreation when other areas are closed for bearprotection (S9), and designing human use around eco-logical constraints (S25). The landscape managers alsosupported developing area-specific targets for acceptablenumbers of bear deaths and habitat change (S11). Theyrejected the claim that further restrictions on human usewere not needed (S1). However, similar to the decision-process reformers, the landscape managers strongly dis-agreed that managers should focus on the conservationfunction of the national park and curtail development(S24) and that human values and attitudes needed to beadjusted for the bear population to be maintained (S22).The landscape managers also did not support increasingthe extent of bear habitat in the park so that fewer bearswould travel outside the park and come into conflictwith agricultural activities (S26). Finally, the landscapemanagers were not in favor of linking management ac-tions more directly to the outcomes of scientific research(S2).

VIRTUAL CONSENSUS ABOUT SOLUTIONS

Areas of virtual consensus about solutions included de-signing human use around ecological constraints (S25),improving appreciation for grizzly bears (S12), and im-proving decision making by making it more inclusive anddeliberative (S30) (Supporting Information).

Discussion

Linking Problem Definitions with Preferred Solutions

The participants in this study all had access to the samescientific research on grizzly bears (results from the East-ern Slopes Grizzly Bear Project were widely dissemi-nated), yet these individuals held very different beliefsabout grizzly bears, parks, and humans in the Banff-BowValley and had constructed very different definitions of

the problem of grizzly bear management. There wereclear links between the ways participants defined theproblem and the solutions they preferred (Dery 1984;Weiss 1989; Ebbin 2010). We summarized these linkednarratives about problems and solutions in terms ofthe “intellectual tasks” for problem analysis describedin the policy sciences literature: clarify goals; describehistorical and current trends in relation to those goals;identify and analyze the factors (conditions) causing orcontributing to trends; make projections about the fu-ture on the basis of trends and conditions; and de-velop and evaluate preferred alternatives to address theproblem (Table 4) (Clark et al. 1996; Lasswell 1971;Clark 2002).

Those who defined the problem of grizzly bear man-agement in terms of deficient goals and plans (factorI) or institutional flaws (factor III) tended to be con-servation advocates (factor A): 13 of the 18 individualssignificantly associated with problems factors I or IIIwere also significantly associated with solutions factorA. Those who believed the problems of grizzly bear man-agement were exaggerated (factor II) tended to be eitherdecision-process reformers (factor B) or landscape man-agers (factor C): 4 of the 7 individuals significantly asso-ciated with factor II were significantly associated withsolutions factor B and 2 others were significantly asso-ciated with solutions factor C. Individuals who definedthe problem as politicized decision making (factor IV)diverged in their choices of solutions. Thus, there were 5main problem narratives espoused by groups in the study:planning-oriented conservation advocates (problems fac-tor I and solutions factor A); institutionally-oriented con-servation advocates (problems factor III and solutionsfactor A); optimistic decision-process reformers (prob-lems factor II and solutions factor B); optimistic land-scape managers (problems factor II and solutions fac-tor C); and democratizers (problems factor IV, mixedsolutions).

Views about Science and Grizzly Bear Policy

Participants’ views about the appropriate role of sciencein policy making varied. The optimistic decision-processreformers strongly supported using science more directlyto guide policy and management, whereas the optimisticlandscape managers tended not to support this approach,and the conservation advocates were neutral or slightly infavor of using science to guide policy and management.These views contrast with those found by Mattson et al.(2006) in a previous Q study of perspectives on large car-nivore conservation in the northern Rocky Mountains ofthe United States. In their study, the group that was mostconcerned about the status of large carnivores stronglyendorsed science as the driver of policy. Mattson et al.(2006:402) speculate that the group they labeled “car-nivore advocates” supported science because “they con-



Table 4. Summaries of the 5 problem-orientation narratives about conservation of grizzly bears in the Banff-Bow Valley, Alberta, Canada.

Institutionally- OptimisticProblem- Planning-oriented oriented decision- Optimisticorientation conservation conservation process landscapetask advocates advocates reformers managers Democratizers

∗

Clarify goals prioritize ecological integrity; conservegrizzly bears and grizzly bear habitat

maintain the right amount of bears forthe park; conservation is not theonly goal of national parks

maintain the rightamount of bearsfor park

Describe trends grizzly bearpopulationstatus is notacceptable;increased humanuse

bearpopulation isvulnerable

grizzly bear population is healthy;problems are overemphasized

grizzly bearpopulation ishealthy; problemsareoveremphasized;human pressurein Banff-BowValley isincreasing

Identifyconditions

deficient goals andplans for grizzlybearconservation;inadequatehuman-usemanagement

deficientinstitutions;inadequatehuman-usemanagement

human-use management and grizzlybear management have beensuccessful

management ispoliticized andnot sciencebased; deficientcriteria formeasuringsuccess; fundingis adequate

Make projections grizzly bearpopulation is notsustainable;increasedbear-humanconflict

grizzly bearpopulation isnotsustainable;park couldsupportmore bears

too many grizzly bears in theBanff-Bow Valley

increased grizzlybear activity inhumancommunities

Invent, evaluate,and selectalternatives

limit human use and development;improve collaboration betweenagencies and interest groups

effectivelyincludeinterests;improve col-laboration;use scienceto guidepolicy; donot prioritizeconservationor furtherlimit humanuse

effectively includeinterests;actively managegrizzly bearhabitat; changescientificresearchmethods; do notprioritizeconservationover otherobjectives

∗The democratizers diverge in the solutions that they prefer.

flated scientific results with value and policy preferences”and believed science would help them achieve theirpreferences.

Unlike Mattson et al. (2006), we found that conserva-tion advocates did not highly rank statements about us-ing science to guide management. They did not opposefurther scientific research, but they ranked scientific re-search as less important than alternatives such as priori-tizing ecological integrity and coordinating managementefforts. Perhaps they believed that enough knowledgeon vulnerability of the grizzly bear population already ex-isted to support prioritizing conservation and restrictinghuman use. In our study the decision-process reformerscalled for science-guided policy, perhaps because they

interpreted the most comprehensive scientific researchas showing marginal growth in grizzly bear abundance(Garshelis et al. 2005a), which reinforced their beliefsthat management had been successful and that conserva-tion and ecological integrity should not be given greaterpriority. The optimistic landscape managers also believedthe grizzly bear population was healthy, which supportedtheir view that more research on the status of that popula-tion was unwarranted, especially if the research involvedscientific practices that they perceived as intrusive (suchas radio collaring bears).

As is often the case in politicized conservation settings,it appears that science was being selectively understoodand used by the different groups to support their own



value-based demands (Kellert et al. 1996; Sarewitz 2004;Kahan et al. 2011).

Finding Common Ground in Politicized Settings

Durning (2006) and Durning and Brown (2007) identified3 main ways Q methodology can be used to resolve con-flicts and develop solutions to contentious policy prob-lems. In our research, the Q study contributed directlyin 2 of the ways these authors identified and perhapsindirectly in the third. First, the Q results offered par-ticipants insight into their own views and the views ofothers, revealing that perspectives about grizzly bear con-servation and human behavior in the Banff-Bow Valleywere complex and nuanced rather than being simply foror against conservation or human use. Also, the initialworkshop brought people together in a noncontentiousatmosphere to discuss the preliminary research resultsand provided an opportunity to improve communicationand understanding of different perspectives.

Second, our results showed participants the distinctproblem narratives underlying their arguments about pre-ferred solutions. Parks Canada followed up on the Q studyby hosting a series of workshops with many of thesestakeholders (and others) in which an interdisciplinaryproblem-solving framework (Lasswell 1971; Clark 2002)was used to expand on the potential common groundidentified in the Q study and work toward a more inclu-sive shared understanding of the problem (Rutherford etal. 2009). We presented results of the Q study at theseworkshops to illustrate how different groups were defin-ing the problem and how these definitions informed pref-erences about potential solutions.

The third way Q method can help to resolve conflictsis to “forge a new solution that has not been advocated”(Durning 2006:608). In the Banff-Bow Valley, workshopparticipants and a successor interdisciplinary problem-solving group developed and agreed on several newmanagement strategies that were implemented by ParksCanada, including seasonal restrictions on recreationalaccess in areas with a history of conflicts between grizzlybears and humans, and a new management target for ac-ceptable annual human-caused mortality of grizzly bears(Rutherford et al. 2009). These issues had been highlycontentious in the past. Although the Q study did notdirectly forge these strategies, it identified the commonground on which these strategies were built, includingthe consensus on improving decision making by makingit more inclusive and deliberative.

Policy scholars assert that problem definition stronglyaffects policy making, by determining whether a prob-lem gets on the policy agenda for consideration andby influencing “which kinds of evidence bear on theproblem, which solutions are considered effective andfeasible, who participates in the decision process, howpolicies are implemented, and by which criteria policies

are assessed” (Weiss 1989:97−98; Pal 2005). Simply con-ducting more scientific research related to a perceivedproblem may not change people’s problem definitionsor move them toward agreement because they may ac-cept only the evidence that supports their existing beliefsand discount contrary evidence (Sarewitz 2004; Kahaneet al. 2011). Thus, developing better understanding ofdiverse problem narratives, as Q methodology facilitates,may be critical to improving conservation programs incontested settings.

Acknowledgments

We thank Parks Canada, the President’s Research GrantsFund of Simon Fraser University, J. Roulet, S. Clark, S.Brown, 3 anonymous reviewers, and the participants inthe Q study.

Supporting Information

Sorting templates for the problems Q sorts (AppendixS1), additional tables showing the statements that re-ceived the highest scores (positive or negative) for eachfactor (Appendix S2), and tables showing the consensusstatements (Appendix S3) are available online. The au-thors are responsible for the content and functionality ofthese materials. Queries (other than absence of the mate-rial) should be directed to the corresponding author.

Literature Cited

Addams, H., and J. Proops, editors. 2000. Social discourse and environ-mental policy. Edward Elgar, Northampton, Massachusetts.

ASRD (Alberta Sustainable Resource Development) and ACA (AlbertaConservation Association). 2010. Status of the grizzly bear (Ursusarctos) in Alberta: update 2010. Wildlife status report 37. Publica-tion T/223. Alberta Sustainable Resource Development, Edmonton.

Benn, B., and S. Herrero. 2005. Grizzly bear mortality and human accessin Banff and Yoho National Parks: 1971–1998. Pages 63–72 in S.Herrero, editor. Biology, demography, ecology and managementof grizzly bears in and around Banff National Park and KananaskisCountry: the final report of the Eastern Slopes Grizzly Bear Project.Faculty of Environmental Design, University of Calgary, Alberta.

Bertch, B., and M. Gibeau. 2010. Grizzly bear monitoring in and aroundthe mountain national parks: mortalities and bear/human encoun-ters 1980–2009. Parks Canada, Banff, Alberta. Available by requestfrom http://www.pc.gc.ca/docs/v-g/oursgest-bearmanag/sec5.aspx(accessed January 2012).

Bow Valley Grizzly Bear Alliance. 2002. The bear necessities: a griz-zly bear conservation strategy for Banff National Park. Bow ValleyGrizzly Bear Alliance, Canmore, Alberta.

Brown, S. R. 1970. On the use of variance designs in Q methodology.Psychological Record 20:179–189.

Brown, S. R., 1980. Political subjectivity: applications of Q methodologyin political science. Yale University Press, New Haven, Connecticut.

Brown, S. R., D. W. Durning, and S. C. Selden. 1999. Q methodology.Pages 599–637 in G. J. Miller and M. L. Whicker, editors. Handbook



of research methods in public administration. Marcel Dekker, NewYork.

Byrd, K. L. 2002. Mirrors and metaphors: contemporary narratives ofthe wolf in Minnesota. Ethics, Place and Environment 5:50–65.

Clark, T. W. 2002. The policy process: a practical guide for natural re-source professionals. Yale University Press, New Haven, Connecti-cut.

Clark, T. W., A. P. Curlee, and R. P. Reading. 1996. Crafting effectivesolutions to the large carnivore conservation problem. ConservationBiology 10:940–948.

Clark, T. W., M. B. Rutherford, and D. Casey, editors. 2005. Coexist-ing with large carnivores: lessons from Greater Yellowstone. IslandPress, Washington, D.C.

Cooper, B., J. Hayes, and S. LeRoy. 2002. Science fiction or science fact?The grizzly biology behind Parks Canada management models. Crit-ical issues bulletin. Fraser Institute, Vancouver, British Columbia.

Cuppen, E., S. Breukers, M. Hisschemoller, and E. Bergsma. 2010. Qmethodology to select participants for a stakeholder dialogue onenergy options from biomass in the Netherlands. Ecological Eco-nomics 69:579–591

Dery, D. 1984. Problem definition in policy analysis. University Press ofKansas, Lawrence.

Durning, D. W. 2006. Using Q-methodology to resolve conflicts andfind solutions to contentious policy issues. Pages 601–620 in AsianDevelopment Bank (ADB), editor. The role of public administrationin building a harmonious society. ADB, Phillipines.

Durning, D. W., and S. R. Brown. 2007. Q methodology and decisionmaking. Pages 537–563 in G. Morcol, editor. Handbook of decisionmaking. CRC Press, New York.

Eastern Slopes Grizzly Bear Project. 1998. Grizzly bear population andhabitat status in Kananaskis Country, Alberta: a report to the De-partment of Environmental Protection, Natural Resources Service,Alberta. Eastern Slopes Grizzly Bear Project, University of Calgary,Alberta.

Ebbin, S. 2010. The problem with problem definition: mapping the dis-cursive terrain of conservation in two Pacific salmon managementregimes. Society & Natural Resources 24:148–164.

Fisher, R. A. 1960. The design of experiments. Hafner, New York.Garshelis, D. L., M. L. Gibeau, and S. Herrero. 2005a. Grizzly bear demo-

graphics in and around Banff National Park and Kananaskis Country,Alberta. Journal of Wildlife Management 69:277–297.

Garshelis, D., M. L. Gibeau, and S. Herrero. 2005b. Grizzly bear de-mographics in and around Banff National Park and KananaskisCountry—postscript for 2003–2004. Pages 50–51 in S. Herrero,editor. Biology, demography, ecology and management of grizzlybears in and around Banff National Park and Kananaskis Country:the final report of the Eastern Slopes Grizzly Bear Project. Facultyof Environmental Design, University of Calgary, Alberta.

Gibeau, M. L. 2000. A conservation biology approach to management

of grizzly bears in Banff National Park, Alberta. PhD dissertation.Resources and Environment Program, University of Calgary, Alberta.

Gibeau, M. L., S. Herrero, J. L. Kansas, and B. Benn. 1996. Grizzly bearpopulation and habitat status in Banff National Park: a report to theBanff Bow Valley Task Force. Eastern Slopes Grizzly Bear Project,University of Calgary, Alberta.

Herrero, S., editor. 2005. Biology, demography, ecology and man-agement of grizzly bears in and around Banff National Park andKananaskis Country: the final report of the Eastern Slopes GrizzlyBear Project. Faculty of Environmental Design, University of Calgary,Alberta.

Herrero, S., P. S. Miller, and U. S. Seal, editors. 2000. Population andhabitat viability assessment for the grizzly bear of the Central Rock-ies Ecosystem. Eastern Slopes Grizzly Bear Project, University ofCalgary, Alberta, and Conservation Breeding Specialist Group, Ap-ple Valley, Minnesota.

Kahan, D. M., H. Jenkins-Smith, and D. Braman. 2011. Cultural cognitionof scientific consensus. Journal of Risk Research 14:147–174.

Kellert, S. R., M. Black, C. R. Rush, and A. J. Bath. 1996. Human cultureand large carnivore conservation in North America. ConservationBiology 10:977–990.

Lasswell, H. D. 1971. A pre-view of policy sciences. Elsevier, New York.McKeown, B. F., and D. B. Thomas. 1988. Q methodology. Quantitative

applications in the social sciences series. Volume 66. Sage, ThousandOaks, California.

Mattson, D. J., K. L. Byrd, M. B. Rutherford, S. R. Brown, and T. W.Clark. 2006. Finding common ground in large carnivore conserva-tion: mapping contending perspectives. Environmental Science andPolicy 9:392–405.

Pal, L. A. 2005. Beyond policy analysis: public issue management in tur-bulent times. 3rd edition. Nelson Thomson Learning, Scarborough,Ontario.

Parks Canada. 2007. Banff National Park management plan, July2007 amendment. Parks Canada, Gatineau, Quebec. Avail-able from http://www.pc.gc.ca/pn-np/ab/banff/docs/plan1/chap1/plan1_E.asp (accessed August 2007).

Rutherford, M. B., M. L. Gibeau, S. G. Clark, and E. C. Chamberlain. 2009.Interdisciplinary problem solving workshops for grizzly bear con-servation in Banff National Park, Canada. Policy Sciences 42:163–187.

Sandbrook, C., I. R. Scales, B. Vira, and W. M. Adams. 2011. Valueplurality among conservation professionals. Conservation Biology25:285–294.

Sarewitz, D. 2004. How science makes environmental controversiesworse. Environmental Science & Policy 7:385–403.

Schmolck, P., and J. Atkinson. 2002. PQ method 2.11. Available fromhttp://www.qmethod.org (accessed June 2005).

Weiss, J. A. 1989. The powers of problem definition: the case of gov-ernment paperwork. Policy Sciences 22:97–121.


Documents

Human Perspectives and Conservation of Grizzly Bears in Banff National Park, Canada