Commentary: Electronically Available Surveys of Attitudes Toward Animals

Harold A. Herzog 1
Lorna B. Dorr
Western Carolina University

Much of the research on attitudes toward non-human species has been conducted with non-representative samples. Largely ignored in the literature on human/animal interactions are surveys conducted by commercial polling organizations using large probability samples of Americans. Many of these surveys contain information relevant to attitudes about animals and animal welfare issues. Much of this information is available to researchers electronically at little or no cost through organizations such as the Roper Center for Public Opinion Research and the National Opinion Research Center.

There is considerable interest among students of human/animal interactions in attitudes toward non-human species. Occasionally, reports on attitudes toward animal welfare issues are conducted using sophisticated sampling techniques coupled with large sample sizes (Kellert, 1996; Pifer, 1996). Unfortunately, such surveys of human/animal interactions are rare. More commonly, studies of attitudes toward animals are conducted using the most convenient group of subjects available to social scientists—college and university students, most often those taking introductory psychology courses. For example, recently we examined 15 issues of Society and Animals published between 1993 and 1998. The data in 11 of the articles were obtained using undergraduates. Of these, one article did not specify the source of the students, one used education students as subjects, and the other nine were based on students taking psychology classes. (One of these also included sociology students). Undergraduate psychology majors are a narrow source of information on human/animal relationships.

Each year thousands of public opinion polls are conducted in the United States, and some of these surveys contain questions pertinent to attitudes toward non-human species. Usually, private companies such as the Gallup Organization or Louis Harris and Associates conduct these polls. Typically, a news organization, a magazine, or a special interest group such as the National Shooting Sports Foundation commissions them. The results of many of these surveys are now available electronically. Further, these polls usually are based on large probability samples of the American public rather than on convenience samples of college students.

Here are a few examples. In 1994, Times Mirror commissioned Princeton Survey Research Associates to assess the views of Americans toward a variety of social causes. Of those sampled, 23% had a “very favorable” attitude toward the animal rights movement, 42% had a “mostly favorable” attitude, 21% had a “mostly unfavorable” view and 9% had a “very unfavorable” view of the movement. A 1989 survey of 1,004 Americans commissioned by Parents Magazine asked respondents about what adults would advise a woman who wanted a fur coat to do. Only 13% said that she should buy the fur coat, 41% said they would advise her to purchase a fake fur coat, and 8% said they would encourage the woman to buy a cloth coat.

A potential problem with large scale polling data relevant to the study of human/animal interactions is that the results of the surveys are widely scattered. Items related to attitudes toward animals may be imbedded in surveys that include a host of questions not related to the topic. In addition, the results are often published in isolated newspaper and magazine articles and are not readily accessed through most standard literature search engines. Indeed, the results may not be published at all. Sometimes the organization that commissioned the surveys may even deny the existence of the data. For example, in the 1980s and 1990s the American Medical Association (AMA) regularly commissioned the Gallup Organization to survey American attitudes toward animal research. Several years ago one of us (HH) telephoned the AMA to inquire about the results of recent polls that the organization may have conducted. We were told (incorrectly) by the AMA spokesperson that the organization had never commissioned these types of surveys.

Accessing Survey Data from POLL

Fortunately, data from numerous surveys conducted by national polling organization—including the AMA surveys—are now available from a central source, the Roper Center for Public Opinion Research. Founded in 1947, the Center is a non-profit organization that provides access to thousands of public opinion surveys. The database containing the survey results is referred to as POLL. POLL is a full-text question retrieval system that allows the user to search the database using key words. POLL presently contains over 10,000 individual survey files covering nearly 300,000 questions dating back to the 1930s. It is updated weekly.

There are several ways to access POLL electronically: (a) direct assess via the University of Connecticut computing facility; (b) the Lexis-Nexis data retrieval service; (c) Academic Universe, an electronic data retrieval service widely available to university students and faculty (also a part of Lexis-Nexis); and (d) Dialog, another on-line service. We used Dialog to search POLL for survey items related to human/animal relations. The number of survey questions for different sets of key words related to human/animal interactions together with sample items are shown in Table 1.

Table 1 Number of survey questions in the POLL database.

Key Words Number of Hits Sample Question
Animal treatment 18 Suppose your pet had an incurable disease or was in a serious accident, how much money would you be willing to pay for medical treatment?
Animal welfare 13 Please tell me whether you have a very favorable, somewhat favorable, somewhat unfavorable or very unfavorable impression of the following groups?...Animal welfare groups.
Animal rights 56 Do you give money to animal rights organizations?
Animal experiments 25 Did you ever have any courses in which either you or the instructor did experiments with live animals?
Animal testing 12 How much do you agree or disagree with each of the following statements?... It is right to use animals for medical tests if it might save human lives.
Vegetarian 15 In terms of your eating preferences, do you consider yourself a vegetarian?
Hunting 176 Which of the following sports and activities, if any, have you yourself participated in within the last twelve months?......Hunting.
Pets 92 Among the following groups of people, do you think that having a pet would provide a more satisfying life, a less satisfying life, or would pets make no difference?.....The average person....Persons living alone....People with mental health problems.
Cat 135 Who deserves greater priority in the Rose GardenSocks, the cat or Buddy, the dog?
Dog 188 Do you think that dogs have good and bad moods like people do?


Dialog searches may pick up survey questions that are unrelated to the topic of interest. For example, a search using the key words “cat” or “cats” revealed 135 questions in which the key words were in either the questions or the responses. Seven of these were related to CAT scans (an acronym for a body imaging technique), one to the Dr. Seuss classic, The Cat in the Hat, and another to “copycat crimes.”

However, many of the items are relevant to scholars of human/animal relationships. These included survey items asking about the proportion of people who “own” companion animals such as cats, superstitions regarding cats, whether or not owners displayed their cat's picture at their place of work, whether the respondents felt that cat owners had a better quality of life than non-owners, whether cats should be used in biomedical research, and whether the respondents identified themselves as “dog persons” or “cat persons.”

Dialog is not free. The researcher's institution must be a subscriber to the service and there is a fee for each search. For example, the on-line search illustrated in Table 1 included 11 key word phrases. The cost of this search was less than $5.00. However, retrieving the data of interest cost an additional $1.30 per survey question. Fortunately, faculty and students can access the POLL database at no cost through Academic Universe if their institution is a subscriber. Access to the database is simple. Once into the Academic Universe homepage (http://web.lexis-nexis.com/universe), click on Reference, then Polls and Surveys. The easiest way to search for information related to animals and society is through a key word search, (e.g. animal welfare). Dates, population, and sample size can narrow searches. Information for each question includes the exact wording of the question, the polling organization, the sponsoring organization, and details about the sample. For the most part, the data are presented in terms of summary percentages. However, some questions added to the database since 1992 include information that allows descriptions of the responses by demographic categories such as sex, age, political affiliation, and income level. In some surveys in which investigators would like to perform more extensive statistical analysis, sets that include the raw data can be purchased from the Roper Center. The Center's user services staff can also conduct data analysis on request. More information about Roper Center services and membership opportunities are available at the Roper Center Web site (http://www.ropercenter.uconn.edu).


National Opinion Research Center Surveys

Another source of readily available high quality data on attitudes of adults toward animals is the National Opinion Research Center (NORC) at the University of Michigan. The National Society for Medical Research commissioned the first NORC survey on animal issues in 1948 (Rowan & Loew, 1995). Of 2500 respondents, 84% thought that stray dogs should be sent to medical schools rather than euthanized, only 8% opposed the use of animals in medical research, and 63% indicated that they had never heard of individuals or groups working to end the use of animals in science. Since 1972, with support from the National Science Foundation, NORC has conducted a nearly annual survey of Americans on a broad range of topics. Called the General Social Survey (GSS), the data set contains questions related to over 3,500 variables. In 1993 and 1994, two animal related questions were included in the survey. In the first, (“anrights”) the subjects were asked the extent to which they agreed or disagreed with the statement, “Animals should have the same moral rights that humans do.” In the second, the participants were asked their opinion about the statement, “It is right to use animals for medical testing if it might save human lives.”

The GSS data set has several strengths. First, the data is easily available to researchers. It can be accessed at no cost over the Internet from the Survey Documentation and Analysis Archives at the University of California at Berkeley (http://socrates.berkeley.edu:7502/) or from Inter-university Consortium for Political and Social Research (ICPSR) at the University of Michigan (http://www.icpsr.umich.edu/GSS/extract.htm). Some analysis (cross tabs, correlations) of the GSS animal questions can be done directly at these sites.

For example, we used the GSS data to examine the relationship between a measure of attitudes toward abortion and animal rights: “Please tell me whether or not you think it should be possible for a pregnant woman to obtain a legal abortion if the woman wants it for any reason?” and “Animals should have the same moral rights that human beings do.” We found no significant relationship between responses to the abortion and animal rights items. Of 531 people, 46% who agreed or strongly agreed with the animal rights statement supported abortion as opposed to 44% of individuals who disagreed or strongly disagreed with the animal rights question. (X = 6.73, P = .15). Note that this pattern of results is different from that obtained by Nibert (1994) who found that animal rights supporters were more likely to support abortion than non-animal rights supporters. The difference may be due to sampling; Nibert's study was based on a sample of 500 adults in one Ohio county whereas the GSS data was based on a nationwide probability sample of 1,768 respondents.

More sophisticated analyses require the access to the full GSS data set. Some of this can be downloaded from the web sites at the University of Michigan or the University of California. The data are also available on CD-ROM from NORC or from the Roper Center at a cost of $250. Because of the number of variables in the GSS data set, a variety of research questions can be asked and answered based on the data. For example, Peek, Dunham, and Dietz (1997) used the 1993 GSS data to examine gender differences in attitudes toward the treatment of animals. Kruse (1999) used the 1994 GSS data set to explore the relationship between views of nature (Romantic versus Darwinian) and support for animal welfare issues.


Statistical Universe

Like Academic Universe, Statistical Universe is a product of Lexis-Nexis. The site (http://web.lexis-nexis.com/statuniv/) provides access to abstracts and some full texts of governmental statistical publications. Unlike the databases described above, Statistical Universe contains data related to activities and behavior rather than attitudes. For example, we searched the Statistical Abstract of the United States (one of the databases in Statistical Universe) for information about pet ownership. We found that in 1996 31.6% of American households included at least one dog and 27.3% included at least one cat. However, because the typical cat owner had more cats (2.2) than the average dog owner (1.7) there were actually more cats than dogs in American homes (59 million cats as compared to 53 million dogs). Statistical Universe contains fewer animal related items than does POLL. However, researchers interested in the economic impact of animals—such as the number of jobs related to animal agriculture, the amount of money spent annually on pets, and the number of Americans who hunt or fish—will find this a useful source.


Summary

An understanding of the role of animals in society requires accurate assessment of our collective attitudes and behaviors related to the many roles that animals play in human life. To date, many studies in this area have been based on non-representative samples. Many researchers are unaware that considerable information exists on a wide array of these attitudes—based on both commercial and non-commercial surveys of the American public. These surveys offer an alternative to studies of college students.


References

Kellert, S. (1996). The value of life: Biological diversity and human society. Washington, DC: Island Press.

Kruse, C. R. (1999). Gender, view of nature, and support for animal rights. Society and Animals. 7, 179-198.

Peek, C., Dunham, C. C., & Dietz, B. E. (1997). Gender, relational role orientation, and affinity for animal rights. Sex Roles, 37, 905920.

Pifer, L. (1996) Exploring the gender gap in young adults' attitudes about animal research. Society and Animals, 4, 37-52.

Rowan, A. & Loew, F. M. (1995). The animal research controversy: Protest, process & public policy. North Grafton, MA: Tufts University School of Veterinary Medicine.

1 Correspondence should be sent to Harold Herzog, Department of Psychology, Western Carolina University, Cullowhee, NC 28723. E-mail: í HYPERLINK mailto:herzog@weu.edu. We would like to thank Lisa Ferraro Parmelee of the Roper Center for Public Opinion Research for her comments on the manuscript.
 

For Abstracts of all issues, including the most current, click Article Abstracts

To order Society & Animals Journal, go to our secure online ordering page

You can Search the online issues of Society & Animals, as well as the entire Society & Animals Forum (formerly PSYETA) website,
for topics and keywords of your interest: