Public perceptions of animal welfare - an international perspective

Grahame Coleman

^{Animal Welfare Science Centre
School of Psychology, Psychiatry and Psychological Medicine
Monash University 3800
Australia}

• Abstract
• Attitudes to animal welfare as a generic issue
• Species preferences
• Animals in research
• Companion animals
• Wildlife
• Livestock
• Zoo animals
  
  
• Conclusions
• References
  
  

Abstract

The aim in this paper is to provide a review of current public attitudes to animal welfare mainly in Australasia, Europe and North America. Data available across the various sectors is very uneven with the majority focusing on farm animals, followed by animals used in research, companion animals and very little on the other sectors. Equally, depending on the sector, most data relevant to livestock come from Europe and Australia, while much of the data on animals in research come from North America and Europe. Nevertheless, across all of the animal sectors for which there are data, there clearly is a widespread belief that animal welfare is important. To the extent that community values alone provide a barometer by which decision makers can adopt strategies that reflect those values, it is important to know what the community thinks. However a consistent approach for obtaining community attitudes over time needs to be adopted, in a way similar to the Eurobarometer approach with regard to farm animal welfare. Perhaps one objective for the Australian Animal Welfare Strategy could be to auspice such an approach.

Attitudes to animal welfare as a generic issue

Attitudes are learned dispositions which are often invoked both in everyday use as well as in research to explain behaviour. The antecedents of attitudes include a range of demographic, personality and other variables that underlie a range of beliefs. Before reviewing current community attitudes relevant to the various animal sectors, it is relevant to identify some of the key demographic variables that are associated with attitudes to animal welfare as a generic issue. These variables tend to underpin all attitudes to animal welfare.

Gender is the most prevalent variable that is related generally to animal welfare attitudes. Matthews and Herzog (1997) studied the personalities of 99 undergraduate psychology students from West Carolina University. In general, personality showed little or no relationship with attitudes towards animal welfare but women generally had more positive attitudes to animals than did men. Herzog (2007) in a later study reviewing gender differences in human animal interactions, found that in 31 studies, women were found always to be more sympathetic to the treatment of animals than were men.

A variable that is also correlated with gender is empathy with women generally showing higher scores on empathy than men. Taylor and Signal (2005) found a significant correlation between empathy and attitudes towards animals in a sample of undergraduate sociology and psychology students. Once again, females scored more positively on attitudes towards animals than did males. In a later study, Signal and Taylor (2007) surveyed 543 respondents from the general community and 389 from animal protection groups in Australia. As expected, respondents from the animal protection groups showed a more positive attitude towards animals and higher levels of empathy than did respondents from the general population. Females in both groups reported more positive attitudes and higher levels of empathy than did males from the corresponding groups.

People’s beliefs about the mental life of animals often shape their attitudes to animal welfare. Knight, Vrij, Cherryman and Nunkoosing (2004) found that, more than a range of demographic variables, belief in an animal mind was consistently negatively associated with approval of the uses of animals across a variety of situations including experimentation, teaching and entertainment. Again, women tended to be less supportive of the uses of animals in these situations than were men, and, not surprisingly, meat eaters were more in favour of the uses of animals in these contexts then were non meat eaters.

A more recent study by Knight and Barnett (2008) involving a qualitative analysis of interview material from eight UK participants concluded that people tend to differentiate between animals on the basis of the perceived mental capacity of each animal and the participant’s familiarity with that animal.
Similarly, while people tended to believe that it was appropriate to use animals for certain purposes, their views with regard to particular animals was also based on their experience with that species. In general, respondents had limited knowledge of animal use procedures and practice, and there was a suggestion that this involved an active avoidance on the part of participants so that they could avoid the cognitive dissonance associated with their uses of the animals on one hand (eg eating meat) and of the practices associated with the management (eg on-farm housing and husbandry) of those animals on the other. This is consistent with Knight, Nunkoosing, Vrij and Cherryman’s (2004) earlier finding that people adapt their attitudes to specific contexts and may actively avoid learning about specific animal procedures and uses to minimise cognitive conflict.

Pagani, Robustelli and Ascione (2007), in a study of children aged from nine to 18 years, found that most were against hunting, the use of furs and leather clothes and the use of animals in zoos. While a small majority of girls were against zoos (64%) boys were evenly divided. Generally girls were more opposed to the uses of animals in these ways than were boys. Clearly the gender differences in attitudes to the uses of animals begin at an early age.

It is unclear what the reasons for gender differences in attitudes to animal welfare are. To the extent that gender differences in empathy may be linked to both gender and gender role - feminity (Karniol, Gabay, Ochion and Harari,1998) however, it is clear that women represent a significant grouping in the community in regard to attitudes to animal welfare. Community surveys need to ensure adequate representation of both males and females to ensure that the results are representative. Also, any analysis of the behavioural consequences of attitudes to animal welfare should probably consider males and females separately.

Species preferences

Attitudes to animal welfare differ across the different animal sectors and across species. Coleman, Hay and Toukhsati (2005) surveyed 1061 Victorians and obtained the perceived importance of welfare for animals in general, farm animals, domestic pets and native animals. While a minority of respondents regarded animal welfare in general to be important (41%) almost half (49.3%) considered farm animal welfare to be important and a majority thought welfare of domestic pets (65%) and native animals (56.6%) to be important.

Driscoll (1995) argued that in order to evaluate attitudes to the uses of animals, it is important to understand people’s generic attitudes to different animal species. She obtained ratings of 33 species from 133 respondents. She also obtained respondents’ attitudes to the various uses of animals. On the basis of the ratings, she found that she could identify five groups of animals. In terms of lovability, primates and larger mammals were rated highest whereas spiders, insects and some mammals including rats, skunks and lions were rated poorly. Interestingly, many food species including turkeys, chickens, lobsters and trout were rated as relatively unlovable.

More recently, Bjerke and Ostdahl (2004) reported that people most liked small wild animals including small birds, squirrels, butterflies and hedgehogs but also showed a moderate liking for the common domestic species of ducks, geese, dogs and cats. Interestingly, birds of prey and fox’s were also liked. Those animals most disliked included rats, mosquitoes, mice, snails, wasps, bees and bats. Many of the common birds were slightly disliked including crows, seagulls, pigeons and magpies. While the authors identified some effects of gender and age on preferred species they did not identify the specific attitudes that underpin this hierarchy of species preferences.

As Driscoll (1995) pointed out, despite the argument from animal rights activists that animals should be valued in themselves, the general community clearly discriminates amongst the different species.

Animals in research

There appears to be a trend for the general community to become progressively more opposed to the use of animals for research purposes. Plous (1996) in a random sample of 5,000 psychologists found a generational difference in attitudes to animal experimentation amongst psychology graduates. Whereas almost 50% of PhD graduates from prior to 1970 supported the use of animals in research, only 20% of 1990s graduates expressed such support. However, this may be confounded with gender because only 21% of the pre-1970 graduates were female while 64% of the 1990s graduates were female.

Broida, Tingley, Kimball and Miele (1993) found that students from seven US universities who were likely to encounter animal experiments in their studies (Psychology, biology, pre-medicine and pre-veterinary medicine) were more likely to oppose animal experimentation than those studying in different areas. As is generally found, women were more opposed than were men. In addition, animal rights supporters tended to be more sceptical of science.

Pifer, Shimizu and Pifer (1994) also found that there were gender differences in attitudes towards the use of animals in research with more women opposing than men. This was true for all animals studied. A more in depth investigation of this gender effect by Eldridge and Gluck (1996) showed that womens’ responses reflected a caring nurturant perspective while men took a more utilitarian perspective which emphasized the potential benefits of animal research. This is consistent with the link between empathy and gender discussed earlier.

There is some evidence for a trend in attitudes to the uses of animals in research. Crettaz von Roten (2008) reviewed trends in animal experimentation in Switzerland over the past decade. Although it was not possible to track trends using the same specific attitude question, there was a clear reduction in the percentage of people who agreed that it was acceptable to use animals in research if it was needed to improve health outcomes. In 1994, the percentage who agreed was 62.4%, and this had reduced to 34.7% in 2005. In seeking to explain these attitudes, they determined that men were more in favour of animal experimentation than were women and, similar to Broida, Tingley, Kimball and Miele (1993), people with more positive attitudes towards science were also likely to support animal experimentation. There was some evidence that there was a concomitant reduction in positive attitudes toward science with the reduction in positive attitudes to animal experimentation. For example, the support for research decreased from 81.3% in 2000 to 68.3% in 2005.

In a survey of Chinese university students from Guangdong province, Davey (2006) found that 80% disagreed with the statement that “continued research with animals will be necessary if we are ever to conquer diseases such as cancer, heart disease and AIDS”. Similarly, 69% agreed with the statement that “the use of animals such as rabbits for testing the safety of cosmetics and household products is unnecessary and should be stopped”. Respondents were less clear about the use of animals in medical research more generally, with 31% undecided and 62% disagreeing (but not strongly disagreeing). This variation in attitude according to context is consistent with Herzog’s (2007) findings. It is unclear the extent to which these student attitudes reflect social desirability responses to the researcher, and to what extent they are attitudes that would motivate specific behaviours.

Coleman (2004) reviewed attitudes to the use of animals in research. He found that opposition from the general public ranged from over 60% in France and West Germany to around 40% in the USA, Greece and Portugal. There has been some evidence from the USA to suggest that, at least for primates and dogs, support for their use in experimentation has steadily declined over that past 20 years from 63% to around 50%. No consistent relationship was found between general scientific knowledge and opposition to animal experimentation but there was an association with general environmental concerns. Approval for the use of animals in biomedical research appeared to be higher than that for behavioural research.

In all of this research, it is difficult to determine the extent to which community attitudes reflect social desirability or actual strongly held views. Unless the attitudes are seen to relate to some behavioural outcome, this will always be of some concern. Paul (1995) in her review of attitudes to the uses of animals in experiments, commented that much of the research to that time said “little about the beliefs and feelings of people directly involved in the animal experimentation debate: anti-experimentation campaigners and the scientists who use animals as part of their work. Politicians, for example, often have very different beliefs than those of their general public supporters … Both scientists and animal rights campaigners have strong feelings about animal experimentation; they have probably thought long and hard about the issues involved, yet they have arrived at opposing conclusions” (Paul, 1995, p3).

Nevertheless, there does seem to be a trend in all countries where surveys have been done - Europe, and North America in particular, for community attitudes towards the uses of animals in research to have become more negative.

Companion animals

The data on attitudes towards companion animals is quite uneven. Much has been done on human companion animal interactions, and reasons for owning a pet and attributes of both owners and pets that characterise the human pet relationship. However, community attitudes to companion animals are much less researched.

In a recent USA study, 126 participants identified themselves as either a dog person, cat person, both or neither and rated their own masculinity, femininity, independence, dominance and athleticism (Perrine and Osbourne, 1998). Cat owners and females were found to hold more favourable attitudes towards cats than were non-owners and males. Females (65%) were more likely to label themselves as ‘cat persons’ than were males (23%). There was also evidence that people tend to display more extreme feelings towards cats than dogs; 18% of the sample said they did not like cats whereas only two participants reported a dislike for dogs.

On a love-hate scale, the majority of 400 randomly selected respondents from Victoria, Australia expressed love for dogs (57%) while only 29% indicated that they loved cats (Toukhsati, Bennett and Coleman, 2007). The average rating for cats on a 7-point scale was near the mean (4.0) for dog owners and non-owners indicating neutral feelings but was 6.1 for cat owners indicating a high degree of love. Interestingly dog owners rated their love for their dogs at 6.64, substantially above cat owners’ ratings for their cats.

Poss and Bader (2007) assessed attitudes towards companion animals amongst Hispanic residents in Texas. They reported that there was anecdotal evidence of companion animal overpopulation, and a substantial number of free-roaming dogs. A substantial majority (between 89.7% and 98.1%) of 206 randomly surveyed residents supported sterilization of dogs and cats and between 80.9% and 99.5% indicated a variety of concerns about free roaming dogs. Fewer than 10% supported free-roaming of dogs and cats. Residents born in Mexico were much less likely to have a fenced yard than were those born in the USA. Residents with an elementary level education were much more likely to chain their dogs than were those who had completed some high school and women were much more concerned about free roaming dogs than were men.

In Dominica, over 90% of respondents reported that Roseau, the capital of Dominica, had a roaming dog problem (Alie, Davis, Fielding, and Maldonado, 2007). Just over 50% thought the best way to deal with this was to take dogs to a shelter, 11.6% thought euthanizing was the best option while less than 3% thought desexing to be an appropriate option. Approximately 70% of respondents were unaware of any animal welfare laws. 91.8% of respondents thought that their pets had feelings but 15.2% condoned violence against animals. Unlike most other research, no clear differences between male and female respondents were found in this study.

Al-Fayez, Awadalla, Templer and Ariwaka (2003) compared the attitudes towards pets of Kuwaiti families with published data from American families and found the attitude scores to be about one standard deviation less positive for the Kuwaitis. This was attributed to the fact that dogs are regarded as “dirty” in the Islamic religion, notwithstanding the view that this does not come from the Holy Qur’an. The authors took the view that animal welfare research in Islamic countries is needed, particularly in view of wide discussions in regard to livestock handling and slaughter.

It is difficult to draw meaningful conclusions from these data. There is little consistency in the focus of the available research. What little evidence there is seems to indicate that, at least in Western countries, the majority of people like dogs but fewer like cats. In Islamic countries people may see only a utilitarian function for dogs and may not actually like them but the available data are too limited to make any confident conclusions on this. The available data do suggest that people from all communities see roaming dogs as a problem and favour desexing as a control technique. The more fundamental question that these data raise is whether liking of an animal is necessarily linked to welfare concerns and whether the belief in animals having a utilitarian function is necessarily linked with more negative welfare attitudes.

Wildlife

Data on attitudes to the welfare of wildlife are also limited. However, there does seem to be a tendency for the community to favour the preservation of wildlife for a variety of reasons.

In an Australian Rural Industries Research and Development Corporation (RIRDC) report (1998), it was found that many people concurrently held the view that kangaroos were a pest to graziers and should be controlled while at the same time believing that they should be protected because they are unique to Australia. This theme seems to occur quite frequently in attitude research on wildlife. For example, Wilkinson and Fitzgerald (1997) found in a large survey of new Zealand respondents, that there was wide acceptance that feral rabbits damage the environment and are a threat to farm production, but that at the same time rabbits were seen to be warm and furry and able to provide a useful economic resource if managed appropriately. There was no clear consensus on appropriate methods for control of rabbits. When Johnston and Marks (1997) conducted a survey on public attitudes to vertebrate pest management, they found that while 58% of respondents supported the use of some biological controls, only 39% were prepared to accept collateral death of native wildlife and only 38% were prepared to accept death of domestic pets. Most people (63%) believed that pest animals were being adequately controlled.

In relation to wildlife species that predate on stock and are seen as a risk to humans, Wolch, Gullo and Lassiter (1997) evaluated media coverage in relation to cougars in California between 1985 and 1995. Historically, Cougars had been hunted for sport and had sometimes been culled. Estimates of the cougar population were variable. Over the period of the survey, there had been seven non-fatal cougar attacks on humans in California and two fatalities. Community views were divided, with ranchers wanting a reduction in the cougar population because of livestock predation and conservationists wanting protection because of risk of extinction. Interestingly, California residents in 1996 voted against a proposition (Prop. 197) to permit hunting of cougars in California. This was in the context of a vigorous media debate with 35% of stories positive, 43% neutral and 22% negative. Of letters to the editor, 88% (21 of 24 letters) were supportive of the cougars, while only 18% of reported stories were.

In a more recent US study of attitudes towards crocodiles, Smithem and Mazzotti (2008) found a positive correlation between perception of the risks posed by crocodiles and attitudes towards them. Neither knowledge of crocodiles nor demographic variables were related to risk perceptions so respondents presumably formed their attitudes through other sources.

In his survey of Chinese university students, Davey (2006) found that 85% of respondents agreed that “it is morally wrong to hunt wild animals just for sport”. Respondents were equally divided on the statement that “there is nothing morally wrong with hunting wild animals for food” with few people showing strong agreement or disagreement. However most agreed that the slaughter of whales and dolphins should be stopped immediately.

Tisdell, Wilson, and Nantha (2005) surveyed 204 Australians from the general community. Based on the authors’ likeability index, mammals were rated highest followed by birds and then reptiles. While likeability was correlated with the extent to which the survival of the species was favoured, it was not a major determinant. All respondents strongly favoured the survival of the species even for the most disliked class, reptiles.

In a more recent study of community attitudes and management strategies for wildlife recovery, Morzillo, Mertig, Garner and Liu (2007) surveyed 1006 residents from southeastern Texas in the United States. Respondents were asked to indicate whether or not they supported the black bear population to increase naturally, whether natural resource agencies should assist in increasing the population, whether these agencies should restock bears or whether they supported the view that black bears should not exist in southeast Texas. Only 6% of respondents endorsed a view that bears like this should not exist. The most popular choice was 50% of respondents supporting restocking with only 23% not supporting this option while the remainder was unsure. Those respondents closest to the national park where the bears were to be located, were most in support of the “no bear” strategy.

Broadly speaking, while members of the community appear to be aware of some of the negative aspects of wildlife encroaching on their safety or their environment, they generally favour appropriate management and are opposed to culling. Wildlife are considered to be valuable and this tends to overshadow community concerns in relation to personal risk or encroachment on livestock production.

---

Livestock

Much of the debate in relation to livestock welfare centers on how livestock are housed and handled and the various husbandry practices that are deemed necessary to manage them.

Heleski and Zanella (2006) surveyed 87 students from Michigan State University to assess their attitudes and knowledge about farm animals. Approximately 80% of the students were female and 58 of the 87 students were in an introductory animal science course while the remaining 29 were in an applied animal behavior course. When asked the question “are you comfortable with how agricultural animals in modern, intensive production system
are housed and managed”, around 40% of introductory students showed at least some concern, while about 70% of the animal behaviour students expressed similar concern. Their knowledge of animal husbandry practices was assessed. The introductory students scored below chance in identifying cages as the main housing for egg laying chickens while the animal behaviour students scored above 80%. Following this, Heleski, Mertig and Zanella (2006) surveyed veterinary college faculty members and animal science faculty members on their attitudes towards farm animal welfare. While veterinary staff members generally showed lower scores than did animal science staff, both groups showed a clear differentiation with regard to the different livestock sectors in the extent to which they agreed that “the predominant methods that are currently used to produce animal products provide an appropriate level of animal welfare in the (relevant) industry”. Beef, sheep and dairy were considered by most (>70%) to receive an appropriate level of welfare, while the percentage agreeing for pigs and poultry was lower, with less than 50% agreeing with this statement in regard to laying hens.

Veterinarians tended to regard most housing and husbandry practices to be of some concern whereas animal science faculty tended to express lower levels of concern. Nevertheless a substantial majority of both groups agreed that cage space for laying hens, stockmanship, lameness in dairy cattle and flooring effects on lameness warranted concern. Early weaning, beak trimming and toe trimming were of less concern.

The Eurobarometer is a tool to monitor public opinion in Europe and was developed by Jacques-Réne Rabier. The standard Eurobarometer was established in 1973. Each survey consists of approximately 1000 face-to-face interviews per Member State (except Germany: 2000, Luxembourg: 600, United Kingdom 1300 including 300 in Northern Ireland). Reports are published twice yearly. Special Eurobarometers are commissioned from time to time. The special Eurobarometer published in 2007 provides a wealth of detail in relation to community attitudes towards the welfare of farm animals across the European Union. While approximately two thirds of respondents had visited a farm which rears animals, the distribution varied markedly across the different countries. While at one extreme, only approximately 7% of Scandinavian respondents had not visited a farm that rears animals, over 60% of respondents from Greece and Portugal had not visited such a farm. There was wide variability in attitudes towards the welfare of laying hens. For example around 70% of respondents from the Netherlands, Denmark, Germany, France and the Czech Republic regarded laying hen welfare as fairly bad or very bad, but less
than 30% of respondents from Cyprus and Estonia held this view. On the other hand even though there was still considerable variability, in no country did more than 50% of respondents view the welfare of dairy cows to be fairly bad or very bad. This was also true for pig welfare, although many more respondents reported that they did not know in regard to pig welfare. In general, laying hens and broiler chickens were regarded as the farm species most in need of welfare improvement.

Kjaernes, Lavik and Kjoerstad (2005) conducted a survey in seven EU countries: Hungary, Italy, France, Great Britain, the Netherlands, Norway and Sweden. Although there was some variation amongst the countries, a minimum of 65% (France) rated animal welfare as important or very important. The issue was regarded as most important in Italy where 87% of the population rated animal welfare as important or very important. Also, there was variability across the countries in the level of agreement that some animal welfare had improved over the past ten years. Fewer than 40% of respondents in Hungary agreed with this statement, while between 60 and 70% of respondents in the other countries agreed. Levels of concern about pig, chicken and dairy cow welfare also varied. Greatest concern was expressed for chicken welfare (typically around 50% of respondents) followed by pigs (somewhat variable from 12% in Norway to 44% in the Netherlands) and fewer than 15% expressing concern about the welfare of dairy cows Lusk, Norwood and Prickett (2007) reported that 97% of 1019 respondents surveyed nationwide in the US agreed or strongly agreed that “it is important to me that animals on farms are well cared for”. When they attempted to assess the influence of social desirability on that response by asking whether the respondent agreed with the statement “the average American thinks that farm animal welfare is important”, only 52% agreed or strongly agreed. This may indicate that people overstate their personal beliefs.

Because long distance transport of animals to slaughter has become an issue in the European Union, Carlsson, Frykblom and Lagerkvist (2007) investigated the price premium that Swedish consumers were willing to pay for the use of mobile abattoirs compared to transportation to slaughter.
Consumers were found to be willing to pay more for the use of mobile abattoirs for cattle, but not for broiler chickens.

Mayfield, Bennett, Tranter and Wooldridge (2007) reported data from approximately 1500 consumers in Italy, Sweden and Great Britain. In Italy and Sweden approximately 85% of respondents felt farm animal welfare was important or very important compared to 73% of respondents from Great Britain. When asked about preference for free range eggs, 71% in Great Britain, 65% and Sweden and 47% in Italy stated that free range was their first choice. However when asked the question “how important is the treatment of hens…..”, 70% of respondents from Italy said very important, 64% from Great Britain and 59% from Sweden. When asked “how good do you think welfare conditions are for chickens….”, 49% from Italy said poor, 56% from Great Britain and 40% from Sweden. Clearly there is not a simple relationship between concerns about hen welfare and purchasing patterns at least in these countries. This raises the broader question of the relevance of attitudes to animal welfare policy and to the consumption of animal products. This will be briefly discussed at the end of the paper.

Nevertheless, it is clear that there is a widespread view in the community that farm animal welfare is important and that laying hens are seen to be at the greatest welfare risk, followed by pigs. Dairy cows were seen to be at lower risk and there appears to be no data available on sheep, beef cattle or some other farmed species such as turkeys, goats, etc. Although there is variability amongst the EU countries, most countries throughout the Western world show similar patterns of attitudes to farm animal welfare.

Zoo animals

There are limited data available on community attitudes to zoo animal welfare. In 1994, when plans were developed to upgrade the Stanley Park Zoo in Vancouver, voters decided in a referendum to phase out the zoo. The Stanley Park Zoo closed completely in December 1997, after the last remaining animal, a polar bear named Tuk, died at age 36. The polar bear pit, which was often criticized by Animal Rights Activists, was converted into a demonstration salmon hatchery. Lindberg (1999) observed that zoos, in the face of criticism that keeping of captive wild animals is not morally defensible, have redefined their mission by aligning themselves with conservationist objectives. Lindberg commented that “it is fair to presume that zoo professionals are strongly committed to animal welfare but less so to animal rights”. He goes on to say that where a choice between individual animal welfare and species preservation are in conflict, higher priority is given to the preservation of the species. Notwithstanding these comments, there is virtually no data on the attitudes and knowledge of either zoo professionals or the general community in regard to zoo animals. Nevertheless, this view accords with the available data discussed earlier on the community’s views on animals in the wild where preservation of the species appears to override any community concerns.

In his survey of Chinese university students discussed earlier, Davey (2006) found that 84% of respondents agreed that “I sometimes get upset when I see wild animals in cages at zoos”. It is difficult to know how prevalent such a view might be.