|
Tracing the Profile of Animal Rights
Supporters: A Preliminary Investigation
Colin Jerolmack
ABSTRACT
A question about the “moral rights” of nonhuman animals in the
1993 and 1994 General Social Survey (GSS) effected an
understanding of some of the demographics of those supporting
animal rights. This study checked results against related
questions concerning attitudes toward animal testing and meat
consumption. The stereotypical profile of an animal rights
supporter is female, well educated, upper-middle class,
middle-aged, and white. The data in this study do not support
the stereotype. Instead, the young, non-black minorities, and
the less educated were more likely to support animal rights;
income was not a significant predictor. Other predictors
examined included religious denomination, frequency of church
attendance, and attitudes toward environmental protection. This
study’s findings suggest the need to rethink “post-materialist”
and “post-citizenship” theories about who supports animal
rights. This paper also points the way toward future studies to
examine the causal linkages between the predictors and animal
rights attitudes.
Much attention has been paid recently to animal rights attitudes
and activism, as social scientists continue to heed Flynn’s
(2001) call to integrate nonhuman animal studies and
human-animal relations into our fields of study. Many people
today appear willing to support some sort of rights for animals.
In a 1995 poll conducted by the Associated Press, two-thirds of
respondents agreed with the statement, “’an animal’s right to
live free of suffering should be just as important as a person’s
right to live free from suffering’ ” (Flynn quoting Agnew, p.
78). In this study, close to one-third of the respondents (827
people of 2, 771) agreed that “animals should have the same
moral rights that human beings do.”
This study uses the statement about “moral rights” and two other
related questions asked in the General Social Survey (GSS)
published by the National Opinion Research Center (NORC, 2000)
to explore the demographic characteristics of animal rights
supporters. The GSS is a cross-national annual personal
interview survey of American households. Conducted by NORC, it
asks hundreds of broad and varied questions to thousands of
randomly selected respondents. Data for this study were drawn
from the GSS to create a database of respondents from 1993 and
1994, the years in which the “moral rights” questions appeared.
Animal Rights Supporters Evidence of a Profile?
A profile seems to be consolidating about the supporters of
animal rights. Again and again, it has been shown that women are
more supportive of animal rights than are men (Gallup &
Beckstead, 1988; Galvin & Herzog, 1992; Kruse, 1999; Nibert,
1994; Peek,. Bell, & Dunham, 1996; Plous, 1998; Uyeki, 2000).
Other features of the stereotypical profile indicated by
previous research include being white, middle-aged, well
educated (at least a bachelor’s degree), and upper-middle class
(Galvin & Herzog; Lowe & Ginsberg, 2002; Plous).
Inglehart (1997) argues in his thesis of “post-materialist
values” that new values have recently emerged in certain groups
because they have achieved material security. Now in a position
to care about others and have concerns beyond their personal
needs, these people attempt to extend rights and promote
acceptance of different lifestyle choices beyond their current
acceptable boundaries. Franklin, Tranter, and White (2001)
(2001) argues that, if such is the case, we would expect support
for animal rights to be higher among members of the
stereotypical profile described above, whom Inglehart identifies
as those in a position to adopt post-materialist values. Such a
demographic profile also fits with the typical
“post-citizenship” movement participant described by Jasper
(1997).
Yet, there have been challenges to the stereotypical profile:
Franklin et al. (2001) argue that pro-animal rights attitudes
are becoming more diffuse across categories such as age, race,
class, and education. Although virtually every study still finds
women more supportive than men of animal rights, some
large-sample surveys find income to be unrelated to animal
rights’ attitudes. Minorities, the less educated, and the young
are more likely to be supportive of animal rights (Franklin et
al., 2001; Nibert, 1994; Peek et al., 1996; and Uyeki, 2000).
This study will weigh in on a debate about the demographics of
animal rights supporters. Using a large national survey, this
study has a robust sample from which it can make an important
contribution to the “profile debate.”
Flynn (2001) argues that “the Judeo-Christian tradition
contributes to the norms that enable humans to mistreat animals”
(p. 79). This study seeks to test empirically this statement by
seeing if people who are members of Judeo-Christian
denominations or frequently attend church are less likely to
support animal rights than those who are not religious, belong
to “nontraditional” religions, or do not attend church. Previous
evidence for this hypothesis can be found in Galvin and Herzog
(1992), Kruse (1999), Nibert (1994), and Peek et al. (1996).
There is growing evidence of a connection between
“pro-environment” attitudes and “pro-animal” attitudes (Kruse,
1999; Peek et al., 1996; and Uyeki, 2000). The GSS contains a
battery of questions about environmental attitudes and activism.
This study will look for evidence of such a relationship.
Data and Method
Dependent Variables
After excluding cases containing missing values on one or more
of the variables (732 respondents), the study examined the
characteristics of 2,081 respondents who answered “yes” or “no”
to the statement, “animals should have the same moral rights
that humans beings do.” Originally, the statement was a
five-point Likert-type question; for comparison in a logistical
regression, however, it was turned into a dichotomous variable.
“Agree” and “strongly agree” were grouped together as supporting
animal rights while “disagree” and “strongly disagree” were
grouped together as not supporting animal rights. Those who were
neutral (542 respondents) have been excluded from the analysis.
The same technique was used to recode, “it is OK to test on
animals to save human lives” into a yes/no question (418
excluded). The third dependent variable, “refuse to eat meat for
moral reasons,” was also recoded: Those who answered “always”
and “often” were grouped as being morally against meat
consumption; those who answered “sometimes” and “never” were
grouped as not morally opposed to meat consumption.
Independent Variables
This study used sex, race, education, age, and total family
income to explore the post-materialist and post-citizenship
theses. To examine the effect of religion on support for animal
rights, the study used items about religious denomination and
frequency of church attendance. Eight environmental attitude and
activism questions were employed to examine a “pro-environment”
stance as a predictor of support for animal rights (Table 1).
Note that GSS 1994 was a split-ballot survey, so 1,655
respondents in that year did not receive the questionnaire
containing the dependent variables used in this analysis.
Table 1. Frequencies of the Demographics and Opinion Items
Variable Frequency Percent (valid) Mean St. Dev.
Age 46.00 17.15
Young 1,383 30.2
Middle-aged 2,198 47.9
Old 1,006 21.9
Race
White 3,830 83.3
Black 567 12.3
Other 201 4.4
Sex
Male 1,975 43.0
Female 2,626 57.0
Education (years completed) 13.12 3.00
Less than HS 801 17.5
High school + 2,699 58.9
Bachelor + 1,084 24.7
Religious preference
Protestant 2,799 61.1
Catholic 1,110 24.2
Jewish 92 2.0
Other 157 3.4
None 420 9.2
Frequency of church attendance
Never 740 16.4
Rarely 1,862 41.3
Regularly 656 14.5
Often 1,251 27.7
Animals should have the same moral rights as people
Strongly agree 168 6.1
Agree 659 23.8
Neutral 522 19.6
Disagree 963 34.8
Strongly disagree 439 15.8
Refuse to eat meat for moral reasons
Always 76 2.6
Often 204 7.1
Sometimes 594 20.6
Never 2,008 69.7
OK to test on animals to save human lives
Strongly agree 400 14.4
Agree 1,435 51.7
Neutral 418 15.0
Disagree 357 12.9
Strongly disagree 168 6.0
**four yes/no questions were asked related to environmental activism:
1. Respondent is a member of an environmental group
2. Have given money to an environmental group
3. Have signed an environmental petition
4. Have participated in an environmental demonstration
**four other five-point scale environmental questions were asked:
1. I do what I can to help the environment
2. I try to buy chemical-free produce
3. I try to drive less to help the environment
Nominal variables were converted into “dummy” variables for
logistical regression analysis. In regression, one value from
each nominal variable, usually the most frequently occurring
value, is taken as the reference category against which the
other “dummy” categories are compared. Race was recoded into
“dummies” for (a) black, (b) white, and (c) other.
Unfortunately, these are the only easily managed categories in
the 1993, 1994 GSS. Those who select “other” are asked to write
in their ethnicity. The reference group was “whites.”
Religious denominations were categorized as (a) Catholic, (b)
Protestant, (c) Jewish, (d) Other, and (e) None (no religion),
with Protestants as the reference group. Educational categories
were (a) less than high school, (b) high school degree, (c) some
college, and (e) bachelor’s degree or higher. Those with a high
school degree served as the reference group.
Three age groups were identified: young (<18-34), middle-aged
(35-55), and old (56 +). The middle-aged were the reference
group. The age categories were employed for the sake of
parsimony, as significant effects were not found for more
discrete categories. The stereotypical animal rights supporter
is middle-aged; once this category was created, making general
comparative categories became a simple matter.
After recoding responses into the above categories, the new
variables were analyzed using logistical regression, the tool of
choice when working with dichotomous dependent variables. The
recoded “moral rights” has only two values: for supporters (1)
and for non-supporters (0); the same is true for the animal
testing and meat questions. This technique illuminates which
predictors have a significant relationship (p<.05) to the
dependent variable and shows the direction of that relationship,
while controlling for all other independent variables used in
the model. Logistical regression also calculates an odds-ratio
score, the factor by which the odds of a particular value of the
dependent variable occurring change when the independent
variable increases by one unit (in the case of nominal variables
like race, a “change by one unit” means a change from the value
of the reference category to the independent variable in
question). The greater the odds-ratio number (eb), the greater
the particular independent variable influences the odds of the
outcome of a certain value of the dependent variable. Negative
“B” values indicate a decreased likelihood to support animal
rights.
Results
Table 2 shows the results of the analyses for the “moral rights”
variable. Only significant relationships are listed, and the
reference category for each nominal variable is in parentheses.
Taking the post-materialist and post-citizenship profile as the
assumption, one would expect to find that those more likely to
support “moral rights” for animals (herein referred to as
“animal rights” are female, middle-aged, white, well educated,
and affluent.
As expected, males were significantly less likely than were
females to support animal rights; we can see, however, that the
other characteristics of the stereotypical profile are not
supported. Against the expected profile, “other race” was
strongly tied to an increase in support of animal rights
compared to whites (blacks, too, were more likely than were
whites to support animal rights, although the relationship is
not quite significant at the .05 level). Against the expected
profile, young people were more likely than were the middle-aged
to support animal rights; compared to high school graduates,
those with a bachelor’s degree or higher were less likely to
support animal rights; those with less than a high school degree
were more likely to support animal rights. In addition, the
expected association between total family income and support for
animal rights was not observed.
Once frequency of church attendance was controlled for, there
was no significant relationship between religious denomination
and support for animal rights, although the Jewish and “other
religion” categories are close to being significantly more
likely than Protestants to support animal rights. Evidence,
however, was found for the hypothesis that church attendance
correlates negatively with support for animal rights.
Several of the “environmental” variables were significant:
Members of an environmental group or who give money to an
environmental group, claim to do what they can to help the
environment, and buy chemical-free produce are all more likely
to support animal rights than are those who do not support the
above activities. The overall model explained about 14% of the
total variance of the dependent variable, and knowing the values
of the significant predictors allows for prediction of about 18%
of those who support animal rights.
Table 2: Animals should have the same moral rights as humans
(Logistical regression analysis)
\
Model Summary
|
Step |
-2 Log likelihood |
Cox & Snell R Square |
Nagelkerke R Square |
|
1 |
2481.124 |
.095 |
.139 |
Classification Table(a)
|
|
Observed |
Predicted |
YESRIGHT |
Percentage Correct |
.00 |
1.00 |
|
Step 1 |
|
YESRIGHT |
.00 |
1504 |
92 |
94.2 |
1.00 |
535 |
116 |
17.8 |
Overall Percentage |
|
|
72.1 |
a
The cut value is .500
Variable B S.E. Wald df Sig EB
Sex
Male -.517 .103 25.351 1 .000 .597
(female)
Race
Other .665 .225 8.764 1 .003 1.945
(white)
Age
Young .510 .112 20.830 1 .000 1.665
(middle-aged)
Church attendance
(never)
Rarely -.309 .148 4.354 1 .037 .734
Regularly -.459 .192 5.745 1 .017 .632
Often -.866 .173 25.096 1 .000 .421
Education
<high school .635 .147 18.710 1 .000 1.886
(HS diploma)
Bachelor’s + -.615 .145 18.087 1 .000 .541
Give money to environmental group
Yes .245 .105 5.447 1 .020 1.277
(No)
Member of environmental group
Yes .433 .176 6.052 1 .014 1.543
(No)
I do what I can to help the environment
.277 .062 13.500 1 .000 .797
I try to buy chemical-free produce
.154 .050 9.403 1 .002 .858
Insignificant predictors: Black (compared to White); Catholic,
Jewish, “Other,” None (compared to Protestant); Old (compared to
Middle-aged); Some college (compared to High school diploma);
Signed an environmental petition (compared to Non-signer); Took
part in an environmental demonstration (compared to
Non-participants); Try to drive a car less to help the
environment; Try to recycle; and Income.
Again, when looking at the above table, the greater the
odds-ratio number (eb), the greater the particular independent
variable influences the odds of the outcome of a certain value
of the dependent variable. Being a member of the “other race”
category was the most significant predictor of support for
animal rights, followed by having less than a high school
degree, being young, being a member of an environmental group,
giving money to an environmental group, buying chemical-free
produce, and doing what one can to help the environment. The
most significant predictors of being unsupportive of animal
rights are rarely and regularly attending church, followed by
being male, having a bachelor’s degree or higher, and often
attending church.
Two other questions pertaining to nonhuman animal welfare were
examined to aid in verifying the results obtained from the
question on “moral rights” and to attempt to construct a more
encompassing measure of support for animal rights. One question
asked whether one abstains from eating meat for “moral reasons,”
while the other asked if one agrees that it is acceptable to
experiment on animals if human lives will be saved. Both these
questions are phrased in ways that could influence the results,
and it is questionable how directly related they are to
supporting animal rights. Such limitations are elaborated below,
and it is for these reasons that both questions were relegated
to a secondary part of the analysis of support for animal
rights. These two questions and the “moral rights” question were
incompatible as a scale.
Examining the regression model on animal testing reveals a
relatively high amount of consistency between the results and
those of the “moral rights” model (Table 3). Being male,
attending church, and having a bachelor’s degree or higher are
associated with being more likely to condone animal testing than
their reference groups (variables are listed in terms of the
magnitude of effect on the dependent variable). These same
groups were less likely to support “moral rights” for animals.
The young (compared to the middle-aged), those who have given
money to environmental groups, and those who try to buy
chemical-free produce are all more likely to be against animal
testing (again, listed in order of magnitude of effect). These
groups are the same ones who were more likely to support “moral
rights” for animals. Being a member of the “other race” category
and having less than a high school degree, however, are not
significantly associated with the animal testing question (as
they are with the “moral rights” question). The model explained
10% of the total variance.
Table 3: It is OK to test on animal to save human lives
(Logistical regression analysis)
Model Summary
|
Step |
-2 Log likelihood |
Cox & Snell R Square |
Nagelkerke R Square |
|
1 |
1993.582 |
.063 |
.103 |
Classification Table(a)
|
|
Observed |
Predicted |
NOTESTS |
Percentage Correct |
.00 |
1.00 |
|
Step 1 |
|
NOTESTS |
.00 |
1822 |
11 |
99.4 |
1.00 |
402 |
10 |
2.4 |
Overall Percentage |
|
|
81.6 |
a
The cut value is .500
Variable B S.E. Wald df Sig EB
Sex
Male -.454 .120 14.369 1 .000 .635
(Female)
Age
Young .353 .128 7.565 1 .006 1.423
(Middle-aged)
Church attendance
(Never)
Rarely -.464 .167 7.749 1 .005 .629
Regularly -.856 .230 13.795 1 .000 .425
Often -.518 .190 7.448 1 .006 .596
Education
(HS diploma)
Bachelor’s + -.858 .172 24.971 1 .000 .424
Give money to environmental group
Yes .260 .120 4.675 1 .031 1.297
(No)
I try to buy chemical-free produce
.307 .058 28.253 1 .000 .736
In looking at the regression model for those who refuse to
eat meat for moral reasons (Table 4), again there is a fair
| 