ATTITUDES OF RURAL AND URBAN PUBLIC TOWARD WOLVES IN CROATIA FINAL REPORT

ATTITUDES OF RURAL AND URBAN PUBLIC TOWARD WOLVES IN CROATIA FINAL REPORT 1

AUTHORS: ALEKSANDRA MAJIĆ SKRBINŠEK ALISTAIR J. BATH ZAGREB,

Attitudes of Rural and Urban Public toward Wolves in Croatia TOWARDS A BETTER UNDERSTANDING OF HUMAN DIMENSIONS IN WOLF MANAGEMENT IN CROATIA Authors: Aleksandra Majić Skrbinšek Kranjčevićeva 28 43500 Daruvar, Croatia almajic@gmail.com Alistair J. Bath Department of Geography Memorial University of Newfoundland St.John s, NL Canada A1B 3X9 abath@mun.ca This report presents results of the studies which were carried out in, 2004 and. The studies were carried out as a part of the LIFE Third Countries project: Protection and Management of Wolves in Croatia of The State Institute for Nature Protection. 2

TABLE OF CONTENTS Preface...11 Introductory part...12 Overview of the history of wolf management in Croatia...12 Human dimensions in wildlife management: research and application...14 Methods...17 Study area...17 Questionnaire...18 Sample...19 Data collection...20 Results...22 About the sample...22 Attitudes toward wolves...34 Management...44 Hunting...53 Tourism...55 Livestock issues...56 Fear of wolves and beliefs about wolves...66 Information and public participation...69 Knowledge about wolves...75 Conclusions...80 Sources...81 3

TABLE OF FIGURES Figure 1: The cognitive hierarchy model of human behaviour. Adapted from (Fulton, Manfredo, and Lipscomb, 1996)...15 Figure 2: Schematic conception of attitudes adopted from Rosenberg and Hovland, 1960 (as cited in Halloran, 1967) and modified by adding the fourth behavioural intention component.16 Figure 3: Croatian wolf range was divided into three study zones for the purpose of this research.17 Figure 4: Share of female respondents in the studies....22 Figure 5: Age structure of the respondents from Gorski Kotar in the 1999, and studies. Thicker lines represent a moving average calculated from 5 neighbouring numbers...23 Figure 6: Age structure of the respondents from Lika in the 1999, and studies...23 Figure 7: Age structure of the respondents from Dalmatia in the 1999, and studies...24 Figure 8: Respondents by education....25 Figure 9: Share of respondents that are members of a hunting club....25 Figure : Share of respondents that have seen a live wolf in the wild...26 Figure 11: Share of respondents that have seen a wolf in captivity (for example in a zoological garden)...27 Figure 12: Share of respondents that have shot / killed a wolf....27 Figure 13: Share of respondents sheep owners...28 Figure 14: Share of respondents goat owners....28 Figure 15: Share of respondents cattle owners...29 Figure 16: Share of respondents horse owners....29 Figure 17: Respondents were asked to mark on a scale from 1 to how important wolf management in Croatia is to them. This chart shows the results for Gorski Kotar...30 Figure 18: Respondents were asked to mark on a scale from 1 to how important wolf management in Croatia is to them. This chart shows the results for Lika...31 Figure 19: Respondents were asked to mark on a scale from 1 to how important wolf management in Croatia is to them. This chart shows the results for Dalmatia....31 Figure 20: Mean values of the responses to the question on importance of wolf management in Croatia (scale 1 to )....32 Figure 21: The respondents were asked to mark on a scale from 1 to how important is to them to be informed about wolf management in Croatia. This chart shows the results for Gorski Kotar...32 4

Figure 22: The respondents were asked to mark on a scale from 1 to how important is to them to be informed about wolf management in Croatia. This chart shows the results for Lika...33 Figure 23: The respondents were asked to mark on a scale from 1 to how important is to them to be informed about wolf management in Croatia. This chart shows the results for Dalmatia...33 Figure 24: Mean values of the responses to the question: How important is to you personally to be informed about the wolf management in Croatia? (scale from 1 to )....34 Figure 25: Which of the following best describes your feelings toward wolves?...35 Figure 26: Mean values of the responses to the question: Which of the following best describes your feelings about wolves? On a scale from 1 to 5, three represents a neutral answer...35 Figure 27: To have wolves in Croatia is: bad, not important or good...36 Figure 28: Mean values of the responses to the question: To have wolves in Croatia is: Bad, not important or good. On a scale from 1 to 5, three represents a neutral answer...36 Figure 29: To have wolves in Gorski Kotar / Lika / Dalmatia (respectively to the respondent s region) is bad, not important or good...37 Figure 30: Mean values of the responses to the question: To have wolves in Gorski Kotar / Lika / Dalmatia is: Bad, not important or good. On a scale from 1 to 5, three represents a neutral answer...37 Figure 31: Has your attitude toward wolves changed over the last two years?...38 Figure 32: If yes, has it become: more negative, neutral or more positive?...39 Figure 33: It is important to maintain wolves in Croatia for future generations....40 Figure 34: Mean values of the answers to the statement It is important to maintain wolves in Croatia for future generations (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...40 Figure 35: It is important to maintain wolves in the area of Gorski Kotar / Lika / Dalmatia (respectively to the respondent's region)....41 Figure 36: Mean values of the answers to the statement It is important to maintain wolves in Gorski Kotar / Lika / Dalmatia for future generations (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)....41 Figure 37: We should assure abundant wolf population for the future generations....42 Figure 38: Mean values of the answers to the statement We should assure abundant wolf population for the future generations (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...42 Figure 39: It is unnecessary to have wolves in Gorski Kotar / Lika / Dalmatia (respectively to the respondent s region) because there already are wolves in other parts of Croatia....43 Figure 40: Mean values of the responses to the question: It is unnecessary to have wolves in Gorski Kotar / Lika / Dalmatia (respectively to the respondent s region) because there already are wolves in 5

other parts of Croatia (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...43 Figure 41: It is not necessary to have wolves in Croatia since abundant populations of wolves already exist in other European countries....44 Figure 42: Mean values of the responses to the question: I is not necessary to have wolves in Croatia since abundant populations of wolves already exist in other European countries (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)....44 Figure 43: Wolves should be completely protected in Croatia....45 Figure 44: Mean values of the responses to the question: Wolves should be completely protected in Croatia (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...45 Figure 45: Wolves should be completely protected in the area of Gorski Kotar / Lika / Dalmatia (respectively to the respondent's region)....46 Figure 46: Mean values of the responses to the question: Wolves should be completely protected in the area of Gorski Kotar / Lika / Dalmatia (respectively to the respondent's region) (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)....46 Figure 47: Wolves should be allowed to be hunted in specific hunting season in the area of Gorski Kotar / Lika / Dalmatia (respectively to the respondent s region)...47 Figure 48: Mean values of the responses to the question: Wolves should be allowed to be hunted in specific hunting season in the area of Gorski Kotar / Lika / Dalmatia (respectively to the respondent s region) (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)....47 Figure 49: In Gorski Kotar / Lika / Dalmatia wolves should be killed using all possible means, including killing of the pups in dens and use of poisons...48 Figure 50: Mean values of the responses to the question: Figure 49: In Gorski Kotar / Lika / Dalmatia wolves should be killed using all possible means, including killing of the pups in dens and use of poisons (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...48 Figure 51: I would agree with increasing wolf numbers in Croatia...49 Figure 52: Mean values of the responses to the question: I would agree with increasing wolf numbers in Croatia (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...49 Figure 53: I would agree with increasing wolf numbers in the area of Gorski Kotar / Lika / Dalmatia...50 Figure 54: Mean values of the responses to the question: I would agree with increasing wolf numbers in the area of Gorski Kotar / Lika / Dalmatia (respectively to the respondent's region) (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)..50 Figure 55: We already have enough wolves in Croatia...51 6

Figure 56: Mean values of the responses to the question: We already have enough wolves in Croatia (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...51 Figure 57: We already have enough wolves in the area of Gorski Kotar / Lika / Dalmatia (respectively to the respondent's region)....52 Figure 58: Mean values of the responses to the question: We already have enough wolves in the area of Gorski Kotar / Lika / Dalmatia (respectively to the respondent's region) (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)....52 Figure 59: Wolves keep roe deer populations in balance...53 Figure 60: Mean values of the answers to the statement: Wolves keep roe deer populations in balance (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...53 Figure 61: Today, there are fewer roe deer and red deer in Croatia than two years ago....54 Figure 62: Wolves reduce the numbers of red deer and roe deer to unacceptable level...54 Figure 63: Mean values of the responses to the question: Wolves reduce the numbers of red deer and roe deer to unacceptable level (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...55 Figure 64: Having wolves in Gorski Kotar / Lika / Dalmatia increases tourism in that area...55 Figure 65: Mean values of the responses to the question: Having wolves in Gorski Kotar / Lika / Dalmatia increases tourism in that area (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...56 Figure 66: Wolves cause a lot of damage to livestock....56 Figure 67: Mean values of the responses to the question: Wolves cause a lot of damage to livestock (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...57 Figure 68: If a wolf killed livestock, I would agree with killing this problem animal...57 Figure 69: Mean values of the responses to the question: If a wolf killed livestock, I would agree with killing this problem animal (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...58 Figure 70: Farmers should receive compensations for the damages that wolves cause...58 Figure 71: Mean values of the responses to the question: Farmers should receive compensations for the damages that wolves cause (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...59 Figure 72: Government should pay the compensations for the damages caused by wolves...59 Figure 73: Mean values of the responses to the question: Government should pay the compensations for the damages caused by wolves (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...60 7

Figure 74: Farmers should pay insurance against wolf attacks to livestock...60 Figure 75: Mean values of the responses to the question: Farmers should pay insurance against wolf attacks to livestock (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...61 Figure 76: Government should help paying for this insurance...61 Figure 77: How much of the total insurance should be covered by the government? Answers are in percentages from 0 to 0 for Gorski Kotar...62 Figure 78: How much of the total insurance should be covered by the government? Answers are in percentages from 0 to 0 for Lika....62 Figure 79: How much of the total insurance should be covered by the government? Answers are in percentages from 0 to 0 for Dalmatia...63 Figure 80: Mean values of the responses to the question: How much of the total insurance should be covered by the government? (1= 0, 6= 50, 11 = 0)...63 Figure 81: Farmers should not receive compensations for the damages caused by wolves if they did not use prevention measures, such as guarding dogs....64 Figure 82: Mean values of the responses to the question: Farmers should not receive compensations for the damages caused by wolves if they did not use prevention measures, such as guarding dogs (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)...64 Figure 83: Farmers in wolf areas should receive subsidies instead of compensations for the damages...65 Figure 84: Mean values of the responses to the question: Farmers in wolf areas should receive subsidies instead of compensations for the damages (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)....65 Figure 85: In areas where wolf attacks to livestock are frequent, wolves should not be protected even if farmers would not receive the compensations....66 Figure 86: In areas where wolves live in close proximity to people, wolf attacks to people are common....66 Figure 87: Mean values of the responses to the question: In areas where wolves live in close proximity to people, wolf attacks to people are common (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)....67 Figure 88: In areas where wolves live near domestic animals, their main food are domestic animals...67 Figure 89: Mean values of the responses to the question: In areas where wolves live near domestic animals, their main food are domestic animals (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)....68 Figure 90: I would be afraid to hike in the woods if wolves were present...68 8

Figure 91: Mean values of the responses to the question: I would be afraid to hike in the woods if wolves were present (1 = disagreement with the statement, 5 = agreement with the statement, 3 = neutral opinion)....69 Figure 92: Which of the following animals is the most dangerous to humans?...69 Figure 93: Opinions of hunters were considered when making wolf management decisions in Croatia....70 Figure 94: Opinions of farmers were considered when making wolf management decisions in Croatia....70 Figure 95: Opinions of environmental NGOs were considered when making wolf management decisions in Croatia....71 Figure 96: The government takes into consideration my opinions about wolf management:...71 Figure 97: Share of affirmative answers to the question: Have you ever heard of LIFE project on wolf management in Croatia?"....72 Figure 98: Share of affirmative answers to the question: Have you ever seen SOS Wolf" poster?"....72 Figure 99: Share of affirmative answers to the question: Have you ever seen a brochure about wolves in Croatia from the LIFE project?"...73 Figure 0: Share of affirmative answers to the question: Have you ever attended a presentation about wolves?...73 Figure 1: Share of affirmative answers to the question: Have you ever seen a documentary of the HRT (Croatian national TV network) about wolves in Croatia?....74 Figure 2: Share of affirmative answers to the question: Have you ever seen a brochure on damage prevention measures?...74 Figure 3: Share of affirmative answers to the question: Have you ever heard about the local offices for the region of Dalmatia, Lika and Gorski Kotar with the main purpose of aiding farmers in prevention of the damages caused by wolves?....75 Figure 4: Knowledge score (0-7)...76 Figure 5: Share of affirmative answers to the question: What is an average weight of an adult male wolf in Croatia? (Correct answer: 21-40kg)....77 Figure 6: Share of affirmative answers to the question: Whether the wolves have always existed in the area of Gorski Kotar / Lika / Dalmatia or not (Correct answer: Yes)...77 Figure 7: Share of affirmative answers to the question: Wolves are completely protected in Croatia" (Correct answer: Yes)...78 Figure 8: Share of affirmative answers to the question: Wolves kill sheep and goats only if there is not enough of deer and other game animals. (Correct answer: False)...78 Figure 9: Share of affirmative answers to the question: How often are wolves successful in catching a wild prey? (Correct answer: Once in attempts)...79 9

Figure 1: Share of affirmative answers to the question: What is the average size of a wolf pack in Croatia? (Correct answer: 1- wolves)...79

PREFACE This report presents descriptive analysis of the data collected during survey of public attitudes toward wolves and wolf management. The data were collected during the year, 2004 and. Some of the results are compared with results of a similar study conducted in 1999 by Bath and Majić (2000). The report is organized into three larger parts. First, Introductory Part includes a section on history of wolf management in Croatia, which offers a background story to the reader for better understanding of the presented results. Following section concisely explains the theory of human dimensions research with a brief glance towards different kinds of issues human dimension research can address. Second part explains the methodology of research used in our study and finally the third part presents the results of the survey. 11

INTRODUCTORY PART OVERVIEW OF THE HISTORY OF WOLF MANAGEMENT IN CROATIA The history of wolf (Canis lupus) management in Croatia is similar to that seen in many other European countries. After many years of significantly reducing wolf numbers in the country, wolf numbers today are increasing according to some sources. The species has been under complete protection in Croatia since 1995. After the legal protection in 1995, it seems that public attitudes toward wolves have become more negative (Bath and Majić, 2000). The evaluation was based on the content analysis of the newspaper articles about wolves and on the documented illegal killings of the wolves. But an accurate representation of existing attitudes toward wolves and wolf management has not been done. It is important at this time, as the wolf population continues to increase, that public attitudes toward and knowledge about wolves and their management be better understood. Gathering representative data from the general public and key interest groups using a quantitative approach is essential for successful wolf management in Croatia. Wolf management seems to be more socio-political in nature than biological, thus requiring an understanding of this human dimension. Up until 1894 the wolf was present in all parts of Croatia. During that year at least one wolf was killed in each of the former municipalities of Croatia. The wolf distribution has been reduced dramatically since 1894. The species has disappeared from the continental lowlands, but has survived in the areas of the Dinara Mountains and in Dalmatia. While no scientific studies estimating the Croatian wolf population have been completed based upon the size of available habitat and the number of wolves killed based on hunting statistics, the wolf population between 1954 and 1972 may have been as high as 600 to 00 individuals. Today the number of wolves is estimated between 130 and 170 individuals (Štrbenac at al., ). Before the wolf was completely protected in 1995, wolves were heavily persecuted using all available means, particularly shooting, poisoning, and killing pups in dens. Hunting data provide the best indication of changes in the wolf population size over this period. In Croatia and Slavonia (lowland part of Croatia today) 1,324 wolves were killed between 1891 and 1921 (approximately 42 wolves per year with a peak of 120 wolves in 1892). After WWII, a major effort was made to exterminate wolves in Croatia. The wolf was classified as an unprotected game species allowing it to be hunted by all ways and means of hunting. An Order for the extinction of wolves was issued in 1948 by the government and a bounty was paid for each killed wolf. Between 1946 and 1986, approximately 540 wolves were killed in Gorski Kotar alone. During the same period, a single hunter in the Dalmatian mountains (Svilaja, Dinara and Kamešnica) killed approximately 68 wolves. Between 1954 and 1972, approximately 5,206 wolves were killed in Croatia resulting in an average of 274 dead wolves per year. Between 1960 and 1961, wolf mortality numbers decreased to 50, and further decreased in 1980-1981 to 32 animals. The change in the number of wolves killed could be attributed partly to a change in attitudes toward the species, but predominately is a reflection of fewer wolves in the country to kill. In Gorski Kotar, the mean number of wolves killed per year dropped from 15 to 9, and then to one, in the periods 1945-1976, 1977-1986, and 1987-1993 respectively. Similar patterns were observed in other parts of Croatia. For example in Dalmatia by the end of the 19th and beginning of the 20th centuries, ten wolves had been killed on average annually. By the end of eighties of the XX century, the number of wolves in Croatia was drastically reduced. The wolf population of Croatia was estimated at fewer than 50 individuals. These remaining individuals survived in Gorski Kotar and in Lika; the wolf was believed to have been exterminated from Dalmatia. While the legal status of the wolf did not change until 1995, over time various extermination methods have become less popular. Poisoning was abandoned for the most part in 1972 and traps and bounties were removed shortly after in 1976. In 1984, the municipality of Vrbovsko in 12

Gorski Kotar took steps to ensure wolf numbers would not be completely eliminated; a decision was made to not kill wolves in the municipality unless there was more than one breeding pair. The road to getting the wolf completely protected in Croatia after many years of persecution was legally relatively quick and easy. An initiative by concerned scientists to legally protect the wolf started on 12 Jan 1994 with the document, The Justified Proposal for the Legal Protection of the Wolf in Croatia. The wolf became legally protected on 09 May 1995 ( Pravilnik o zaštiti pojedinih vrsta sisavaca, N.N. 31/95). With official protection of the wolf, the government became responsible for payment of compensation to livestock owners for wolf damages to livestock. During the first three years of legal protection, wolf mortality actually increased at least 5 times and perhaps as much as 11 times suggesting that the public was not consulted or supportive of the change in legislation. At that time no human dimensions data had been collected to assess the attitudes, nor any work towards building partnerships with interest groups completed. Without such public involvement in the decision-making process, successful implementation of the new policy remained challenging. Most of the illegal killing was by shooting and while full protection remains as the national law, no one has been prosecuted for poaching. Despite the illegal killings, the wolf population appears to have steadily increased rising. Prior to full protection for the wolf, the general public in 1986 ranked the wolf fourth in a list of pests, behind the bear, wild boar and fox. In 1993, the wolf was ranked seventh on the pest list behind wild boar, insects, bear, lynx, rodents and fox. In 1994, the wolf appeared on a postage stamp for Earth Day. Since 1995 several seminars on assessing predator damages have occurred and preventative measures for livestock promoted, especially the use of guard dogs. Biological research on wolves and radio-collaring the first Croatian wolf began in October of 1998. The first Human Dimension in wolf management project in Croatia was initiated in spring of 1999. The study has included residents of the entire wolf area of Croatia. As expected the differences were found among the three regions (Gorski Kotar, Lika and Dalmatia) with Gorski Kotar residents being the most positive toward wolves and Dalmatia residents being the most negative toward wolves. Interest groups included in the study were hunters, foresters and high school students from the same regions. Overall the most positive were high school students, followed by foresters and hunters. The most negative group were general public residents from Dalmatia. As the wolf population continues to grow and policies and management options be reconsidered, it will be important to understand the viewpoints of the general public and various interest groups toward wolves and possible management options. For some individuals and interest groups, the increasing wolf population provides increased opportunities for eco-tourism, the chance to hear a howl, see a paw print in the mud or a wolf scat. For others, more wolves bring more fear of human safety and greater threats to livestock. For many, myths and misunderstandings about the species remain high, thus creating challenges for governments who are forced to manage the species and to gain an informed public consent about wolf management. Wildlife managers and government officials need to have scientific data to allow them to balance views heard by the loud minorities and truly understand how the majority of residents feel about the issue. This study establishes that baseline assessment of attitudes toward and beliefs about wolves and their management from a representative public in three zones (Gorski Kotar, Lika, and Dalmatia). The study documents how attitudes and beliefs differ between groups, and also within groups across space and time, thus providing managers with information that should allow flexibility in management options by region. 13

HUMAN DIMENSIONS IN WILDLIFE MANAGEMENT: RESEARCH AND APPLICATION Often, there is a debate on how a wildlife population should be managed. The debate occurs because different interest groups hold different values and therefore have different management goals. What are values and what is management? MANAGEMENT has been defined as a dynamic, objective, and goal-driven process that functions with an environment having cultural, political and ecological components (Decker, Brown, and Siemer, 2001b). Rokeach (1973) defines a VALUE as an enduring belief that a specific mode of conduct is personally or socially preferable to an opposite or converse mode of conduct or end state of existence. So, it is important for managers to understand values and how values influence attitudes and consequently support or opposition for the management options. HUMAN DIMENSIONS RESEARCH is about how people value wildlife, how they want wildlife to be managed, and how they affect or are affected by wildlife and wildlife management decisions (Decker, Brown, and Siemer, 2001a). Human dimensions research can address various objectives (Bath and Majić, 2000): Baseline assessment to begin attitudinal and belief monitoring in order to assess, for example if an educational effort has made a difference. Educational role targeting specific weaknesses in knowledge to affect attitudes. Working toward designing more effective educational materials. Building partnerships working toward understanding the issues of a variety of interest groups, building trust, and initiating the first steps toward working together. Identification of areas of support and disagreement over management options, thus assessing the feasibility of approaches being successfully implemented. Identification of types of conflicts as a first step toward conflict resolution. In theory there are several distinct types of conflict, but in reality several types of conflict are usually combined The following are the different types of conflict (Mitchell, 1989):. o o o o Cognitive disagreement over knowledge can be solved through environmental education, especially if the education programme is well targeted (there are many examples of more or less successful environmental education programmes). Values although the two sides agree on the information which makes up the knowledge (for example, number of wolves in Croatia), their values can be different. This type of conflict can be resolved through negotiations about principles of management. Costs /benefits conflict an interest group can argue that benefits created by a specific management action go to another privileged group. Behavioural the disagreement is because of the messenger. It does not matter what the message is, it will not be accepted because of the messenger. 14

By understanding attitudes and beliefs, more effective targeting of educational messages is possible. Also, such information provides managers an indication of support or opposition for certain management practices in the area. By integrating views of all parties interested in agency decisions into the actual decision making conflicts can be conquered and avoided in the future. The following chapter offers a brief description of the theoretical background for the human dimensions research. Cognitive approach Human dimensions research such as this one is based on a cognitive approach. Cognitive approach is concerned with social psychology theory called ATTITUDE AND VALUE THEORY (Fishbein and Ajzen, 1975) and typically suggests that people s values determine their attitudes and that their attitudes, in turn, affect their behaviours (Pierce, Manfredo, and Vaske, 2001). This approach has been used to explore people s values, attitudes and norms regarding wildlife and its management and to understand how those concepts influence their behaviour (Pierce, Manfredo, and Vaske, 2001;Vaske and Dötterer, 1999;Lohr, Ballard, and Bath, 1996). Behaviours Behavioural Intentions Numerous Faster to Change Peripheral Specific to Situations Attitudes and Norms Value Orientations Values Few in Number Slow to Change Central to Beliefs Transcend Situations Figure 1: The cognitive hierarchy model of human behaviour. Adapted from (Fulton, Manfredo, and Lipscomb, 1996). VALUES are commonly defined as desirable end states, modes of conduct, or qualities of life we individually or collectively hold dear, such as freedom or equality (Rokeach, 1973). Values are often formed early in life and are extremely resistant to change once they have been established (Rokeach, 1973). We cannot observe values directly, so we look for evidence of people s values. In human dimensions research, values are often measured using questionnaires that ask people what they consider important. A person s small number of values can direct a large number of attitudes that express those values (Pierce, Manfredo, and Vaske, 2001). The influence of values on attitudes and behaviour occurs indirectly via other components of cognitive hierarchy. For example, basic beliefs serve to strengthen and give meaning to fundamental values (Vaske and Dötterer, 1999). Patterns of these basic beliefs create value 15

orientations (Fulton, Manfredo, and Lipscomb, 1996). Some researchers suggest that value orientations regarding natural resources can be arrayed along a continuum ranging from anthropocentric to biocentric (Shindler et al., 1993; Steel et al., 1994; Thompson and Barton, 1994 as cited in Vaske et al., 1999). In social psychology literature there are many statements that are offered as definitions of ATTITUDES. For example, Katz (1960, as cited in Halloran, 1967) defines an attitude as the predisposition of an individual to evaluate some symbol or object or aspect of his world in a favourable or unfavourable manner. An attitude is a part of the bridge between the outside stimulus situation and the behavioural outcome and is made of three components: affect (liking or disliking), cognition (decision on support or disapproval) and actual behaviour (Halloran, 1967). Fishbein and Ajzen (1975) later introduced the fourth component behavioural intention (the intention of a behaviour is not always the same as the actual behaviour) (Figure 2). Measurable independent variables Intervening variables Measurable dependent variables Sympathetic nervous responses Verbal statements of affect STIMULI (individuals, situations, social issues, social groups and other attitude objects) ATTITUDES AFFECT COGNITION BEHAVIOURAL INTENTION Perceptual responses Verbal statements of beliefs Verbal statements of intended actions (behaviour) BEHAVIOUR Overt actions Verbal statements concerning behaviour Figure 2: Schematic conception of attitudes adopted from Rosenberg and Hovland, 1960 (as cited in Halloran, 1967) and modified by adding the fourth behavioural intention component. NORMS have been defined as standards of behaviour they specify what people should do or what most people are doing (Pierce et al., 2001). Norms directly influence behaviour. In human dimensions, the importance of understanding norms have three main reasons, as proposed by Pierce et al. (2001). Development of normative standards allows for better understanding of issues. Explaining and predicting behaviour. Influencing behaviour (making people aware of norms can affect their actions). 16

METHODS STUDY AREA Human dimensions research is interdisciplinary in nature and is most effective when information can be collected and blended directly with biophysical data over the same geographic space. In an effort to be most relevant for management decision-making concerning the wolf in Croatia, the HD study area included the entire wolf range in Croatia. Presently, wolves in Croatia are distributed over the entire Dinara Mountain Range, from the Slovenian border to the borders with Bosnia and Herzegovina and Montenegro. On the margins of this area wolves are occasionally present. Wolves, however, are not found on the Istria peninsula (excluding Učka and Čičarija mountains), on any of the Croatian islands, and also not on the lowlands. In order to allow comparisons, we have used the same zones (Gorski Kotar, Lika and Dalmatia Figure 3) defined by Bath and Majić (2000) in the previous study. Additional data were collected from the urban population the residents of the capital city, Zagreb. Figure 3: Croatian wolf range was divided into three study zones for the purpose of this research. Zone one, labelled Gorski Kotar, is the most northern and mountainous zone and includes the entire region of Gorski Kotar and the north-western parts of Lika. The size of the zone is approximately 5,245 km2. Forestry provides the main source of income for the region. The northwest border of the zone is defined by the state border to Slovenia, and towards the Istrian peninsula by the presence of wolves. The west border of the zone stretches along Velibit Mountain near the Adriatic Sea coast. The north border of the zone is defined by the constant presence of wolves. The eastern border represents the state border with Bosnia and Herzegovina 17

while the southern border to zone two is defined by the biophysical and demographic characteristics (rivers, different vegetation, change in the number of livestock, and higher rates of attacks to livestock by wolves). This zone one region is the most densely forested of the three zones (beech, silver fir, spruce and pine mixed forest dominate) and of Croatia. Gorski Kotar is about 60-70 forested and therefore represents the best wolf habitat in the country. Wild ungulates are present. Number of sheep in the area is relatively low as well as the occurrence of wolf attacks to livestock.. According to scat analysis, wild prey is the predominant wolf food in this zone. Zone 1, Gorski Kotar, includes three national parks that make up an important part of large carnivore habitat in Croatia: Risnjak National Park, North Velebit National Park and Plitvice Lakes National Park. Zone 2 of the HD study zone includes the remaining parts of Lika and is approximately 4,396 km2 in size. The eastern border of the zone is the state border with Bosnia Herzegovina while the western border spreads along the Adriatic Sea coast (defined by constant wolf presence). The southern border of zone 2 is with Dalmatia defined by different biophysical and demographic characteristics. Zone 2 is less forested than zone one. Beech forest dominates, with spatial walleyes, turned to grazing land. The number of sheep is considerably larger than in zone 1. Despite the relatively high sheep density, wolf attacks to livestock are relatively rare. Wild prey forms the majority of the diet of wolves in this zone. Zone 3 includes the inland parts of Dalmatia and is 6,170 km2 in size. The human population density is the highest of the three zones. The coastal, very densely populated areas of Dalmatia are excluded from the zone as they are not in wolf range. The eastern border of the zone is the state border with Bosnia and Herzegovina. The Neretva River forms the southern border. Poor Mediterranean vegetation and a rocky countryside make livestock raising challenging in this zone. Only hare and wild boar are present. Scat analysis reveals that 86 of the wolf s food is livestock. The density of sheep is relatively high in Dalmatia, and very high wolf damage occurs in this region. Although the numbers of sheep are relatively high, sheep owners usually own small flocks of sheep (average 30 sheep) that they use as a form of secondary income QUESTIONNAIRE In surveys, answers are of interest not intrinsically but because they are in relationship to something they are supposed to measure. In that sense designing a question for a questionnaire is designing a measure, not a conversational inquiry (Fowler, 2002). For the purpose of this study, three different questionnaires were designed. Two questionnaires were used in the wolf range zones (Gorski Kotar, Lika and Dalmatia). Those were very complex questionnaires which consisted of 82 and 85 questions, respectively. The other questionnaire was developed and used in Zagreb with the goal to obtain insight into urban population attitudes toward and beliefs about wolves. This questionnaire consisted of 70 questions covering the same issues as the previous one. Consistency in the both questionnaires was preserved in order to allow comparisons between the groups. In general, following topics were included in the questionnaire: Attitudes toward wolves. Beliefs about wolves and a knowledge section made up of factual questions. Attitudes toward various management issues, such as livestock issues, hunting, public involvement in decision-making and others. Personal experience with wolves. 18

Socio-demographic information about each respondent. The development of the questionnaires was based on the questionnaire used in 1999 by Bath and Majić (2000). There are two main reasons for that: The questions in the 1999 questionnaire were tested and therefore we could omit additional testing of our questionnaire. Similar questionnaires would allow direct comparisons of the results from 1999 with our data, thus we have also directed our research towards monitoring of the attitudes and beliefs. Some of the questions used in the 1999 study were omitted because they were of less use than expected. For example, question It is important to have healthy population of wolves in Croatia was not understood by all respondents in the same way. Some interpreted "healthy" as "viable", while others thought of it as "free of disease". Because ensuring consistent meaning of the questions to all respondents was one of our requirements, this question had to be excluded from the questionnaire. Additional sections related mainly to public involvement in decision making and familiarity with the LIFE project were added to the questionnaires. SAMPLE The goal of sampling was to obtain the sample that represents the population as well as possible. According to Fowler (2002), there are three key aspects of sample selection: 1. The SAMPLE FRAME is the set of people that has a chance to be selected. Statistically speaking, a sample can be representative only of the population included in the sample frame. One design issue is how well the sample frame corresponds to the population a researcher wants to describe. For example, if we wish to document public attitudes in a geographical area and we use telephone directory for sampling, than the sample frame is made of users of the telephone service and not all the inhabitants of the area. In our study the sample frame included adult (15 years and older) residents of the three zones (Gorski Kotar, Lika and Dalmatia) and Zagreb. 2. Probability SAMPLING PROCEDURES must be used to designate individuals to be included in the sample. Each person must have a known chance of selection. This chance is set by a sampling procedure. However if respondent's characteristics such as his or hers availability or initiative affect the chances of selection, there is no statistical basis for evaluating how well or how poorly the sample represents the population. In our study, there is a probability that chance of selection was influenced by individuals availability (where interviews were carried our in person - rural areas) and by individuals initiative (where interviews were done by mail Zagreb). 3. The details of sampling design, such as its SIZE and SPECIFIC PROCEDURES used for selecting individuals. A sample size of 400 respondents per zone was chosen in order to allow for results to be accurate 19 times out of 20, plus or minus 5. Such a sample size results in a 95 confidence level and a 5 confidence interval. Specifically, stratified random sampling was used where for each community in the study zones a sample was calculated that is proportional to the number of residents that live in this community. In Zagreb survey in 2004, and in the survey of public attitudes in Gorski Kotar, Lika and Dalmatia in, when the surveys were carried out by mail, it was expected that the response rate will be relatively low. Because of 19

that, the samples were increased to 800 for Gorski Kotar, Lika and Dalmatia and to 00 for Zagreb. Respondents were selected randomly by a computer-generated random key from an electronic version of telephone book. DATA COLLECTION We have used two methods of data collection. In rural areas in respondents were interviewed in person in their homes, while the survey of urban population of Zagreb in 2004 and follow-up survey of rural population in utilized interviewing by mail. Each of the used methods has its advantages and disadvantages. Fowler (2002) has summarized those as follows: 1) Potential advantages of personal interviewing: a) Some sample designs can be implemented best by personal interview (e.g. area probability samples). b) Personal interview procedures are probably the most effective way of enlisting cooperation for most populations. c) Interviewer can accurately follow complex instructions or sequences. d) Multimethod data collection is possible (e.g. observations). e) Confidence building is possible. f) Longer survey instruments are possible than by any other method. 2) Potential disadvantages of personal interviewing: a) It is likely to be more costly than the alternatives. b) Trained interviewers are needed. c) The total data collection period is likely to be longer than other procedures. d) Some samples may be more accessible by some other mode (urban areas). 3) Potential advantages of self-administered data collections: a) Ease of presenting questions requiring visual aids. b) Asking questions with long or complex response categories. c) Asking batteries of similar questions. d) The fact that the respondent does not have to share the answers with the interviewer. 4) Potential disadvantages of self-administration: a) Especially careful questionnaire design is needed. b) Open questions usually are not useful. c) Good reading and writing skills by respondents are needed. d) The interviewer is not present to exercise quality control with respect to answering all questions, meeting question objectives, or the quality of answers provided. e) Cannot control who answers the questions. 5) Advantages of mail procedures: 20

a) Relatively low cost. b) Can be accomplished with minimal staff and facilities. c) Provide access to widely dispersed samples and samples that for other reasons are difficult to reach by telephone or in person. d) Respondents have time to give thoughtful answers, to look up records, or to consult with others. 6) Disadvantages of mail procedures: a) Ineffectiveness of mail as way of enlisting cooperation (depending on group to be studied and the topic). b) Various disadvantages of not having interviewer involved in data collections. c) Need for good mailing addresses for sample. 21

RESULTS ABOUT THE SAMPLE During the project we have interviewed a total of 1 606 respondents. It is important to point out that the success of interviewing considerably differs with respect to the chosen survey method. Personal contact with potential respondents in their homes resulted with response rates higher than 80. Interviewing by mail reduced the success to around 20 (in Zagreb) and (in rural areas). General public respondents are coming from the three zones of the wolf range (Gorski Kotar (n=406 in, n=88 in ), Lika (n=384 in, n=84 in ) and Dalmatia (n=382 in, n=43 in )) and from the urban population of Croatia (Zagreb, n=219). The sample is made up mostly of male respondents, except in Zagreb where both genders were equally represented (49,3 were females and 50,7 were males). Figure 4 shows share of female respondents in the surveys. 50 45 40 35 1999 1999 1999 30 25 20 15 5 0 Gorski Kotar Lika Dalmatia Figure 4: Share of female respondents in the studies. Average age of the respondents was approximately the same in all the surveyed groups of public. Figures 5, 6 and 7 present the age structure of the respondents. In average, the age structure has not changed during the three surveys. However, it is possible to note the increased presence of middle aged respondents (40-50 years old) in the sample from which was collected through mail. We can assume that interviewing in person in the respondent's home during the usual working hours probably eliminates potential respondents from this age group, as they are away at their working place. 22

7.0 Gorski Kotar 6.0 5.0 4.0 3.0 2.0 1.0 0.0 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81 84 87 90 93 Dob 1999 Moving average@5 Moving average@5 Moving average@5 Figure 5: Age structure of the respondents from Gorski Kotar in the 1999, and studies. Thicker lines represent a moving average calculated from 5 neighbouring numbers. 8.0 7.0 Lika 6.0 5.0 4.0 3.0 2.0 1.0 0.0 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81 84 87 90 93 Dob 1999 Moving average@5 Moving average@5 Moving average@5 Figure 6: Age structure of the respondents from Lika in the 1999, and studies. 23

9.0 8.0 Dalmatia 7.0 6.0 5.0 4.0 3.0 2.0 1.0 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81 84 87 90 93 0.0 Dob 1999 Moving average@5 Moving average@5 Moving average@5 Figure 7: Age structure of the respondents from Dalmatia in the 1999, and studies. For the purpose of data analysis, the education level of the respondents was categorized into three groups: (1) attendance to elementary school, (2) attendance to high school and (3) university degree. In all three rural zones most of the respondents belong to the high school group, while respondents from Zagreb included equal representation of high school (48.8) and university-level education (48.4) (Figure 8). Explanation for such a high share of respondents with university degree in Zagreb probably can be found in the fact that Zagreb survey was done by mail. This method requires that the potential respondents independently fill out the questionnaire. In order to avoid any possible bias, results from Zagreb have to be carefully analyzed and interpreted. Similar effect can be noted also with the results from, where the percentage of elementary school education has decreased in comparison to, and percentage of university education has increased. 24

80 70 60 50 40 30 20 0 Elementary School High School University Elementary School High School University Gorski Kotar Lika Dalmatia Figure 8: Respondents by education. Results from show that relatively small number of the respondents, especially those from Zagreb, are members of a hunting club. The highest number of hunters (Figure 9) was documented in Gorski Kotar,5 and in Dalmatia,1, followed by Lika 7,2 and Zagreb with only 1,4. In sample, the percentage of hunters has increased, probably as a result of better motivation of the hunters to respond to the questionnaire through mail. 25 20 15 1999 1999 1999 5 0 Gorski Kotar Lika Dalmatia Figure 9: Share of respondents that are members of a hunting club. Given the cautious nature of wolves, which are known to avoid people, it is interesting that most of the respondents claimed they have seen a wolf in the wild (Figure ). As expected the 25

percentage of respondents that have seen a wolf in the wild was considerably lower among the respondents from Zagreb (20.6). Large majority of all respondents have seen a wolf in captivity, for example in a zoo (Figure 11). And while there were respondents from rural areas that have killed a wolf (Figure 12), it was not a case among the respondents from Zagreb. On average 45 of all respondents from Gorski Kotar, Lika and Dalmatia knew a person that has killed a wolf. Many respondents from Dalmatia and Lika owned sheep and/or goats. As expected, among the respondents from Gorski Kotar there are fewer livestock owners (Figures 13, 14, 15 and 16). Important to note is that many of the respondents have had a damage caused to their livestock by wolves (8 in Gorski Kotar, 18 in Lika and 14 in Dalmatia). And while 25 of those coming from Gorski Kotar and Lika have reported the damage, in Dalmatia this percentage was 50. 80 70 60 50 1999 1999 1999 40 30 20 0 Gorski Kotar Lika Dalmatia Figure : Share of respondents that have seen a live wolf in the wild. 26