Summarize 4 research articles due anytime before tomorrow!! Must be in own words

Wiley and Society for the Scientific Study of Religion are collaborating with JSTOR to digitize, preserve and extend access to Journal for the Scientific Study of Religion.

http://www.jstor.org

The Abortion Controversy: Conflicting Beliefs and Values in American Society Author(s): Joseph B. Tamney, Stephen D. Johnson and Ronald Burton Source: Journal for the Scientific Study of Religion, Vol. 31, No. 1 (Mar., 1992), pp. 32-46 Published by: on behalf of Wiley Society for the Scientific Study of Religion Stable URL: http://www.jstor.org/stable/1386830 Accessed: 03-11-2015 02:08 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

The Abortion Controversy: Conflicting Beliefs and Values in American Society

JOSEPH B. TAMNEY* STEPHEN D. JOHNSON* RONALD BURTON*

This study examines the ideological and structural bases for attitudes and political action relating to abortion. Data on abortion attitudes from a random sample of "Middletown" in the fall of 1989 (N = 400) indicated that the strongest effects on abortion attitude stemmed from a belief that life begins at conception, a belief in privacy rights, and religion. Social traditionalism was also found to play an important role in a pro-life stand, but it was influential only for conservative Protestants and not for Catholics. A general commitment to avoid taking anyone's life was significantly related to abortion attitudes but only for Catholics. For political action, the data showed that the pro-life movement was still more important in 1989 than was the pro-choice movement. Party identification was not related to an expressed willingness to use abortion as a political litmus test, but attending pro-life churches did increase the likelihood that pro-life adherents would use abortion in making voting decisions.

THEORETICAL BACKGROUND

Contemporary societies are being shaped by class and cultural conflicts (Brandmeyer and Denisoff 1969; Inglehart 1990; Wald et al. 1989). Class conflicts are those rooted in a person's place in the economy and stratification system. Of course, class conflicts can also involve cultural differences, but under such circumstances these cultural differences are derivative of more fundamental class or economic antagonisms. A cultural conflict as such is one in which values are of primary importance, and their origin is not explained in class terms. A good example of such culturally based conflicts is the present controversy over the issue of abortion (Clarke 1987; Himmelstein 1986; Himmelstein and McRae 1988; Woodrum 1988).

In this study we built a model of the reasons people do or do not support legalized abortion. In doing so, we paid particular attention to the cultural or ideological bases of abortion attitudes. Then we examined the implications of the abortion issue for voting decisions. As national studies have found, the American population has remained divided on the abortion question for the last 15 years (Dionne 1989; Gallup and Castelli 1987; Gallup Report 1989; Hugick 1989). We focused in this study on the conflicting ideological perspectives which serve as the bases for people's stands on this complex issue (Tamney 1986:163).

*Joseph B. Tamney and Stephen D. Johnson are professors of sociology, and Ronald Burton is an associate professor of sociology at Ball State University, Muncie, IN 47304.

? Journal for the Scientific Study of Religion, 1992, 31 (1): 3246 32

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

THE ABORTION CONTROVERSY 33

Ideology and Legalized Abortion

The defense of legalized abortion is rooted in individualism. Most important is the libertarian desire to limit the role of government, which has led to efforts to expand the zone of privacy. As Justice Harry Blackmun of the Supreme Court wrote in a case upholding the legalization of abortion: "Our cases long have recognized that the Constitution embodies a promise that a certain private sphere of individual liberty will be kept largely beyond the reach of government" (USA Today 1986). Not surprisingly, Scott (1989) has found that the right of an individual to choose is the reason most frequently given by people who support legalized abortion.

Individualism has also taken on a gender-based meaning relevant to the abortion issue. Feminists have argued that if the individual is esteemed, a person has the right to control his or her own body. As Harrison (1983:197) has written: "Bodily integrity, or the power of self-direction as an embodied human being, is even more substantively conditional of human worth and dignity than most of the political rights reputed to be basic in a liberal society." For a feminist, then, the abortion issue is "the cutting edge of feminism," because to allow legal abortions is to deny that "anatomy is destiny" (Willis 1985:16). Legalized abortion, in this perspective, is one means for achieving the conditions which will allow women to have dignity and reach fulfilLment as persons.

During the course of Western history, Judeo-Christian culture has been a signifi- cant source of limits on individual freedom. One Christian tradition has viewed nature as "God's rules" and has therefore tended to oppose human tampering with natural processes. From this tradition has emerged the idea of the sacredness of human life, a belief which has affected modern attitudes toward the abortion issue. Some Catholic leaders, and most notably Cardinal Bernardin, have placed the protest against abortion within a larger ideological framework of the need to preserve all human life (Cleghorn 1986). From this perspective, the rejection of abortion accompanies the condemnation of nuclear warfare and capital punishment, along with efforts to eliminate such deadly problems as poverty and malnutrition. All such issues form a "seamless garment," the purpose of which is to preserve human life. Conceivably, the nobility of this cause has led some people, and especially Catholics, to refuse any withdrawal from the line of con- ception as the beginning of human life. However, research suggests that the "seamless garment" argument might apply only to a small, perhaps elite, segment of Americans such as Cardinal Bernardin (Neitz 1981; Goldman 1984; Jelen 1990).

Another Christian tradition relevant to abortion emphasizes the sinfulness of human nature, views desires for physical pleasures or comforts as corrupting, and emphasizes the need to control the human's baser nature. This puritanical perspective finds expres- sion in a commitment to various traditional moral rules. Thus, research has consistently found that opponents of legalized abortion hold traditional ideological views about gender and family such as the belief that woman's place is in the home (Cable 1984; Luker 1984:160); a condemnation of sex education in the public schools (Schwartz 1985); and an opposition to various forms of sexual freedom, divorce, and pornography (Neitz 1981:270; Barnartt and Harris 1982; Johnson and Tamney 1988). Such attitudes have been collectively identified as "social traditionalism" (Himmelstein 1983). The head of the Pro-Life Action has said, "There's more to life than the visceral twitch, and I think

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

34 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION

using sex purely for pleasure is sort of animalistic" (quoted in Ridgeway 1985:29). Thus, a second ideological basis for opposition to legalized abortion is a commitment to social traditionalism.

Another secular, ideological basis for limiting arguments based on individualism is political conservatism. Prominent conservatives tend to distrust human nature and spontaneity (e.g., Kirkpatrick 1982). "Historically, conservatives have been distrustful of man's basic nature and have emphasized the need for restraining institutions" (Lenski 1971:221). Not surprisingly, a positive relationship has been found between political conservatism and social traditionalism (Tamney and Johnson 1988). Also, an anti- abortion stand has generally been associated with politically conservative people and groups (Wilcox 1989).

In sum, the acceptance of legalized abortion among Americans is a consequence of individualism which finds expression in libertarian and feminist ideologies. On the other hand, opposition to legalized abortion finds its ideological justification in the religious tradition, specifically in natural law or puritanical doctrines. It was also expected in our study that political conservatives, perhaps for reasons similar to those of puritans, would tend to oppose legalized abortion. People accepting such doctrines generally believe that human life begins at conception, despite the absence of any scientific basis for this judgment (New York Times 1989), and they often conclude that abortion is murder, which is by far the reason most frequently given for rejecting legalized abortion (Scott 1989).

The Theoretical Model

Diagram 1 depicts the model which we tested for explaining abortion attitudes. It was based on the expectation that a person's attitude toward legalized abortion is most directly related to cultural/ideological factors such as a belief in privacy, feminism, a belief that life begins at conception, social traditionalism, political conservatism, and a general belief that all life is worth preserving. Furthermore, the model tries to explain the abortion stance of Catholics and conservative Protestants. We suggest that the Catholic church has had an impact on anti-abortion attitudes through its theology that all life begins at conception and should be preserved. However, it is only among funda- mentalist Protestants, and not among Catholics or mainline Protestants, that religiosity has been found to be related in a consistently conservative manner to social traditionalist issues (Jelen 1990). Thus, we argue that conservative Protestants are anti-abortion because they are social traditionalists, which relates directly to an anti-abortion attitude in our model and indirectly to this attitude via the belief that life begins at conception.

Of course, religious people might be influenced by religious doctrines without personally knowing about theological matters. CHURCH INFLUENCE was a variable we included in our analysis to measure directly whether or not a person perceived that church leaders had an effect on the position he/she took on abortion. Our reasoning led us to predict that Catholics and conservative Protestants would be the most influenced by church leaders on the abortion issue, and what they hear at church would lead them to take a negative view of legalized abortion. We propose that these church influences are mainly anti-abortion, and that a major justification for this stand is the view that

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

THE ABORTION CONTROVERSY 35

DIAGRAM 1

PARTIAL THEORETICAL MODEL

w

_ WM Z _~~+ _ __ .

+ Church_ ,Influence .A

_~~~~~~ _ Age Catholic ,

+ > ~~Life at Conception .l

Education Conservative ......... t .\o

Protestant\ Support

. ; For Legalized Preservation Abortion

of All Income

+ life

I I _ ~~~~Social tf . . . _ _ Sex Traditionalism -

? / ~~~>N} Privacyl +1 / Orientation

Race Political l l Conservatism

Feminism

life begins at conception; we further predicted a relationship between church influence and this view. In addition, we presumed that social traditionalists would be more likely to be influenced by the teachings of the church (Cavanaugh 1986).

Since valuing control over one's life to establish one's self-respect is an important component of feminism, we predicted that feminism would be related to a privacy orien- tation. We also assumed that among social traditionalists an emphasis on submission to traditional norms would lead them to reject arguments based on privacy because such arguments imply the importance of individual autonomy.

Five structural variables related to social inequality were added to the model as exogenous variables: age, education, income, sex, and race. Paths between these five exogenous variables (respectively) and other variables are omitted from Di)agram 1 in order to reduce the "clutter" in the model and make it more readily understandable. However, the relationships among these five exogenous variables were assessed in the LISREL analysis which was run to test the model. These variables were chosen because they have been shown in previous studies to be directly related either to abortion attitudes or to the other nine variables which we propose are directly or indirectly related to abortion attitudes (Colasanto and DeStefano 1989; Gallup Report 1989; Hugick 1989; Johnson 1986; Plutzer 1988; Tamney 1986; Tamney and Johnson 1988).

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

36 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION

Abortion and Politics

Studies have found that abortion opinions had little, if any, effect on elections during the 1970s and 80s (Clymer 1984; Granberg and Burlison 1983; Johnson et al. 1989; Lewin 1988; Traugott and Vinovskis 1980). However, in the recent Webster vs. State of Missouri case, the Supreme Court gave states new leeway to impose restrictions on abortion (Newsweek 1989). Americans reported that as a result of the Supreme Court's ruling they would be more likely to consider the abortion issue when deciding how to vote (Hugick 1989). In our research we explored the role of religion in making abortion a political litmus test.

METHODS

Our sample consisted of 400 residents from the SMSA of Lynd's (1929) "Middle- town" (Muncie, Indiana). The respondents were interviewed over the phone and were selected by a standard random-digit-dialing technique in which the proportion of telephone numbers for each telephone exchange was the same as the latter's proportion of the Muncie SMSA.

Although this was not a nationwide sample, Middletown has been shown to be representative in religious and political respects. In four previous studies of presiden- tial elections, the percentages voting for the major presidential candidates were virtually the same in Muncie as they were nationwide (Johnson 1978; Johnson and Tamney 1982, 1985; Johnson et al., n.d.). Also, nationwide studies have replicated Middletown findings about the lack of influence of the Christian Right on national elections (Lipset 1985; Lipset and Raab 1981).

The interview schedule contained a series of questions which measured basic demographic characteristics such as the respondent's sex, age, race, total income, and education.

The impact of religion on abortion attitude was assessed by creating two dummy variables. One was high church-attending Catholics vs. low church-attending Catholics, along with four other religious groupings (Protestants, Jews, other religion, and no religious preference). This variable was called the CATHOLIC variable (N = 24). The other dummy variable was high church-attending, fundamentalist, conservative Protestants vs. all other religious groupings (low church-attending or non-fundamentalist or non-conservative Protestants, Catholics, Jews, other religion, and no religious preference). Fundamentalism is a personal belief which was measured by asking the respondents the extent to which they believed every word of the Bible was true; conser- vative Protestantism meant membership in a church classified as such.' This variable was called CONSERVATIVE PROTESTANT (N = 43). These particular variables were created on the assumption that Catholics and fundamentalist Protestants who frequently attend church would be the most opposed to abortion. Another religious factor was

1. CONSERVATIVE PROTESTANT included affiliation with the Lutheran Church (Missouri Synod), Baptist churches, the Christian Church (not the Disciples of Christ), the Church of the Nazarene, and other, small non- mainline churches. The item assessing fundamentalism has been shown in a series of previous studies to have the most reliability and validity of seven items used to measure religious fundamentalism (Johnson 1986).

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

THE ABORTION CONTROVERSY 37

obtained by asking respondents the extent to which they believed that church leaders had influenced their positions on abortion.

Social traditionalism was measured by a two-item scale. One question asked respondents whether sexual abstinence or improved sex education was the better way to handle teenage pregnancy. Another question determined the extent to which the respondent agreed with the following item: "It is much better for everyone involved if the man is the achiever outside the home and the woman takes care of the home and family" (ALPHA = .39). Possible reasons for the relatively low alpha are that there were only two items in the scale and they assessed two distinct aspects of social tradi- tionalism: The former measured traditional attitudes about sex, while the latter measured traditional family structure.

Another set of questions concerned life issues. Belief in life at conception was measured by asking the respondents when human life begins. The available responses were: the moment a woman becomes pregnant (coded 3), when a fetus can survive (coded 1), or not sure (coded 2). This variable was called NEWLIFE. Two questions concerned a respondent's beliefs about preserving all life. One question determined the extent to which respondents thought that, in order to preserve life, both euthanasia and capital punishment should be eliminated. This variable was called ALLLIFE. Its purpose was to determine if respondents believed it was wrong as a general principle for one person to take the life of another. The second question determined the extent to which a respondent believed that it is morally wrong to conduct medical experiments in which animals are killed. This was called PROTECT. The Pearson r correlation bet- ween these two life measures was rather small (r = .09, n.s.). Thus, we decided that these two variables, for the most part, measured different pro-life attitudes.

Two ideological positions which generally support a pro-choice stand are feminism and belief in the privacy of the abortion decision. As a measure of privacy belief, respondents were asked the extent to which they agreed or disagreed with this state- ment: "Decisions about divorce and abortion are private matters, and we should not pass laws that regulate these matters." This question (PRIVACY) was meant to deter- mine if the respondent believed there is a private part of our lives from which govern- ment should generally be excluded. A measure of feminism was obtained by having the respondent indicate agreement with the following item: "Only if women organize to change laws can anything really be done about sex discrimination." Another question assessed general political ideological position by asking respondents how they would classify themselves along a scale ranging from very conservative to very liberal. The latter type of item has been shown in a number of studies to be related to actual positions on a number of political issues (Flanigan and Zingale 1987:116).

The abortion attitude itself was measured by the following question:

Which of the following statements best describes your opinion about abortion?

1. Abortion should never be permitted.

2. Abortion should be permitted only if the life or the health of the woman is in danger.

3. Abortion should be permitted if, due to her particular situation, the woman would have difficulty in caring for the child.

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

38 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION

4. Abortion should never be forbidden.

Alternative #1 was considered a strong pro-life position, alternative #2 a moderate pro- life position, alternative #3 a moderate pro-choice position, and alternative #4 a strong pro-choice position.

One question concerned the political importance of the abortion issue. Respondents were asked how concerned they were about abortion on a range from "think very little about it" to "a political candidate's stand on abortion would greatly determine" whether or not the respondent would vote for that candidate.

RESULTS

The percentages of people who took the four basic positions on the abortion measure were as follows: should never be permitted, 10.0%; permitted only if the woman's life is in danger, 44.9%; permitted if the woman would have difficulty caring for child, 18.9%; should never be forbidden, 26.2%. Such divided sentiment on abortion has also been found in Gallup polls which have indicated that from 1975 to 1988, about 25% of the American public supported abortion under any circumstances, about 20% would have denied abortion under all circumstances, and about 55% believed abortion should be allowed under certain circumstances (e.g., cases of rape or if a family cannot afford to care for another child) (Gallup Report 1989).

A Causal Model for Abortion Attitude

LISREL analysis was used to test the proposed model of factors related to a person's abortion attitude (see Diagram 1). In this LISREL analysis, measurement modeling was not used; the variables were assumed to be directly assessed by the measures (e.g., ALLLIFE and PROTECT were considered to be two different variables measuring two different aspects of preserving all life; they were not modeled as two measures of a life- preserving latent variable). LISREL analysis was used in preference to ordinary least squares analysis, because LISREL allows a test of a model as a whole. At the same time, a LISREL analysis can indicate which originally proposed individual paths are significant and which paths not originally proposed should be present, given the observed correlations among all variables in the model.

Since CATHOLIC and CONSERVATIVE PROTESTANT were dummy variables created out of the same religious preference variable, a "0" relationship between these two was forced into the model. The correlation matrix for all variables in the model, which the LISREL used as basic data, can be seen in Table 1.

The first LISREL analysis, which tested the original model, indicated that the RACE and ALLLIFE measures had neither direct nor indirect relationships with the abortion attitude measure. These two variables were thus eliminated from the model. In addi- tion, INCOME related only to the PROTECT measure (i.e., those of lower income wanted to protect animals from medical experiments), and FEMINISM related only to the PRIVACY measure at just barely the .05 level (i.e., feminists were most likely to support

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

THE ABORTION CONTROVERSY 39

a00

0 S 0.4~~~~~~~~~C

CM 00

S X X o X ~~ o o o Lo o

> - N N 3> ~~~~o o co o o N ,

?~~~~~~~~i ci 1 4 OOC 3

X O.O Co.b Co . . . ?. O. . O. O. 4 CS

E0 >

X X Z X z Z X Z Z ? 8 X X N X X 6 c

0-

m cq C?3 00 t- V CII Lo I II II I

-o o coc

o E~~~ kt~~ 0 O~~ O~~O c E-4 E 1)`CI-

- 0~~~~~~~~~~~Z~

r n 4E--4 o~~ - ~~~ kt~~ C~~ - ~~O c~ja -Z4 g

kt0 O., E- kt z0 C Q~C~

;s 4 0 0 04 0 00 O. ;so 0 0

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

40 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION

the privacy of the abortion decision). When both income and feminism were removed from the model, the probability that the model would fit the data increased by more than two-fold. Thus, FEMINISM and INCOME were also eliminated from the analysis.

The model which was eventually derived is seen in Diagrams 2a and 2b. It was obtained by repeating LISREL analyses until only those paths remained which were significant and no additional paths were indicated for inclusion according to residual analysis (Pedhazur 1982). The CHI-SQUARE test for model fit had a probability of .47. If the path between CATHOLIC and CONSERVATIVE PROTESTANT had not been fixed at 0, the model would have fit the data with a probability of .872 Of all the models considered in these analyses, this one best accounted for the observed correlations among all the variables in Diagrams 2a and 2b. The proportion of variance explained in abortion attitude by the other variables in the final model was 35%.

DIAGRAM 2A

FINAL MODEL PATHS AMONG ENDOGENOUS VARIABLES

.11 | CATIIOLIC C1URCI

C_________ _ ,1, ,,, INFLUENCE

CONSERVATIVE

\NEWLIFE .30 PROTESTANT

.23 ABORTION

SOCIAL PROTECT

. 3 TRADITIONALISM ANIMALS

2 \ -1 .13

i POLITICAL PRIVACY

.08* CONSERVA'I'ISM |_-__ ORGANIZATION

*p < .05, all other paths significant at p < .01.

Previous work by the authors on support for leaders and ideologies related to abortion (such as support for Pat Robertson and religious fundamentalism) has indicated that different factors played different roles for people of varying education levels (Burton et al. 1989; Johnson et al. 1989). The final model (see Diagrams 2a and 2b) was thus tested separately for those of lower education (high school graduates or less) and for

2. LISREL analysis further indicated that the two separate items of the only scale in our model, i.e., social traditionalism, also fit the data. That is, when the item concerning sexual abstinence vs. sex education, or the item about women staying home, was substituted into the model instead of the social traditionalism scale, the model still fit the data equally well (the probability was .33 for the former and .35 for the latter).

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

THE ABORTION CONTROVERSY 41

DIAGRAM 2B

FINAL MODEL: PATHS AMONG EXOGENOUS AND ENDOGENOUS VARIABLES

CATHiOLIC - .09 CliURCII l _____ l____,___ INFLUENCE

.1

AGE

- .09* _ _ _ _ _ _ _

CONSERVATIVE NEWLIFE 09* PROTESTANT |_-_____X _14

10 1~~~~~~~- 14

EDUCATION ABORTION

.12 1

\ SOCIAL _ PROTECT

\ TRADITIONALISM / ANIMALS

SEX \J

POLITICAI, PRIVACY CONSERVA'I'ISM ORGANIZATION

*p < .05, all other paths significant at p < .01.

those of higher education (some college or more). After adjustment of the paths in the two models to fit the data better for the two respective sub-groups, two major differences were obtained. First, for the high-education group, political ideology had a direct effect on abortion attitude (beta = -.22, p < .01) while social traditionalism did not (beta = -.05, n. s.). However, for the low-education group, social traditionalism had a direct effect on abortion attitude (beta = -.15, p < .01) while political ideology did not (beta = -.06, n. s.). Second, the model accounted for almost twice as much of the variation in abortion attitudes for the high-education group as it did for the low-education group (52% vs. 28%).

Political Activism

Thirty-one percent of the respondents claimed that the abortion issue was important enough to them that a candidate's stand on abortion would determine whether or not they would vote for that candidate. The pro-life respondents were more likely to make this claim than were the pro-choice respondents (36.4% vs. 23.6%, Chi-Square = 6.60, p < .05; but 51.6% vs. 29.5% for strong pro-life respondents and strong pro-choice respondents respectively, with Chi-Square = 17.5, p < .01).3

3. Our justification for classifying people into just two categories, pro-life and pro-choice, came from an

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

42 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION

We found that there were no differences among Democrats, Independents, and Republicans in the extent to which they would use candidates' stands on abortion as a basis for choosing among them. We also found that pro-choice Republicans were just as likely to use abortion in candidate choice as were pro-life Republicans, and that there were just as many pro-life Democrats as pro-choice Democrats who would use abortion in candidate choice. While some have observed that abortion has become a divisive issue for Republicans (Ellis 1989), our data indicate that Democrats are just as divided over abortion as are Republicans.

Although party affiliation did not influence respondents' use of the abortion issue as a criterion for voting, we wondered if there might have been a religious affiliation influence. To determine this, the theoretically and empirically most pro-life religious environments (i.e., conservative Protestants and Catholics) were combined into one category which we called "pro-life religious environments." The most pro-choice religious social contexts (i.e., mainline Protestants and those with no religious affiliation) were combined into a second category which we called "pro-choice religious environments." The mean score on the abortion attitude measure for conservative Protestants and Catholics was 2.36. The mean abortion score for the mainline Protestants and for those with no religious affiliation was 2.81 (F for difference was 7.40, p < .001).

This new dichotomized variable was then brought into a two-factor ANOVA design with an individual's position on abortion as a second factor (see Table 2). The main effect difference in means between the pro-life and pro-choice religious environments was significant: The former was associated with a heightened political relevance of abortion (F = 8.02, p < .01). The ANOVA table further indicates that it is the pro-life religious

TABLE 2

MEANS FOR POLITICAL ACTIVISM ON ABORTION FOR THE FACTORS ABORTION ATTITUDE AND ABORTION RELIGIOUS ENVIRONMENT

Abortion Religious Environment

Pro-Life Pro-Choice

Pro-Life 2.30 1.91

Abortion __________________ (92) (75) 2.13

Attitude Pro-Choice 1.98 1.94

(43) (100) 1.95

2.20 1.93

Note: Numbers in parentheses are the frequencies in each cell.

examination of the changes in the means for all of those variables which had a direct relationship with abortion attitude in Diagrams 2a and 2b. We found that, in general, changes in the means took place between the moderate pro-life position and the moderate pro-choice position (see abortion attitude item in Methods section). For example, the greatest increase in a pro-privacy orientation took place between the two moderate positions, and there was not as great an increase from the strong pro-life position to the moderate pro-life position, or from the moderate pro-choice position to the strong pro-choice position.

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

THE ABORTION CONTROVERSY 43

environment which stimulates an individual to base his/her political decisions on abortion, and that the mean scores of pro-choicers in pro-life religious environments is no different from those for both pro-choicers and pro-lifers in pro-choice environments. Almost twice as many pro-life respondents attending pro-life churches said they would use the abortion issue in making political decisions as did the three other types in Table 2. For pro-choicers in a pro-choice religious or non-religious environment (vs. a conservative Protestant or Catholic environment), there was no greater tendency for single-issue voting on the basis of abortion than was the case for pro-choicers in pro-life environments. The F for inter- action was 4.22 (p < .05). The correlation between a pro-life vs. pro-choice attitudinal position and political action was .21 (p < .01) for those in pro-life religious environments and -.02 (n.s.) for those in pro-choice religious environments.

None of the structural variables was related to political activism on the abortion issue. That is, neither age, education, income, sex, nor race had significant zero-order correlations with the use of abortion in political decision-making.

DISCUSSION AND CONCLUSION

As anticipated, support for legalized abortion was related to cultural variables: a commitment to privacy, a belief that human life begins at conception, and social tradi- tionalism. The importance of religion in determining abortion attitudes was also substan- tiated by the significance of the CHURCH INFLUENCE variable.

Feminist writers have linked their acceptance of legalized abortion to their feminist beliefs (e.g., Harrison 1983; Willis 1985). The absence of a relationship between our feminism measure and the abortion variable, even at the zero-order level, must be the result of methodological or conceptual problems.4

Although the ALLLIFE measure did not relate to abortion attitude, another measure presumed to indicate a belief in maintaining life, the PROTECT ANIMALS measure, did relate. The negative relationship between PROTECT ANIMALS and support for legalized abortion might have occurred because some respondents were generally pro- life regardless of species (we found a slight, positive relationship between PROTECT and ALLLIFE, as can be seen in Table 1). In fact, the Seamless Garment Network includes animal rights groups (Johnson 1990). As for the ALLLIFE measure, it assessed whether respondents believed it was categorically wrong for one person to take the life of another (the specific examples were euthanasia and capital punishment). The fact that PROTECT, but not ALLLIFE, was significantly related to abortion attitude suggests that while some pro-life people do not seem interested in protecting all living

4. There are different types of feminism. Our question measured assimilationist feminism (Rossi 1972). However, studies using a measure of radical feminism, which concerns changing gender identities, might find significant relationships with abortion attitude. One problem is maintaining a distinction between radical feminism and social traditionalism. In our study, we asked about the desirability of women staying at home and leaving the public world to men. While this was part of our measure of social traditionalism, it could also be used as an indicator of radical feminism. This question about gender roles was indeed significantly related to abortion attitudes (r = -.24, p < .01). In exploring the feminism issue, we examined relationships among three ques- tions used in the 1990 GSS. As mentioned, respondents were asked whether women should stay at home. Their responses were related to a second question about legalized abortion (r = .31). A third question asked whether the respondent would vote for a qualified female candidate for the presidency (a measure which probably assesses assimilationist feminism). Responses to this question were only weakly related to abortion attitudes (r = .11).

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

44 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION

creatures, they might be sympathetic to protecting "innocent" living beings. A second qualification of our conclusion about ALLLIFE is that although ALLLIFE

did not have a main effect relationship with abortion attitude, additional analysis indicated that it correlated significantly with a person's abortion stand for Catholics (r = -.34, p < .02; i.e., those Catholics most likely to take a pro-life stand on abortion were also the most likely to believe in preserving all life), whereas for Protestants there was a non-significant positive correlation (r = .07, n.s.). This additional analysis indicates that the "seamless garment" ideology could possibly play some role in supporting a Catholic's pro-life stand.

Overall, it seems that a pure concern with all forms of life is not, except possibly for Catholics, a normative idea which serves as a basis for the "pro-life" movement. A minority of those opposing legalized abortion might be motivated by a desire to preserve "innocent" life. By and large, however, the data do not support the conclusion that the movement against legalized abortion is based primarily on a general pro-life perspective.

As for political conservatism, it had a significant independent effect on abortion attitudes only among the more educated. An earlier study that used samples from nine European countries found that abortion attitudes related to a respondent's self-placement on a Left-Right scale more strongly for an elite subsample than for the general popula- tion subsample (Inglehart 1990:295). Other studies have found that political conservatism was related to being a fundamentalist and to supporting Pat Robertson's bid for the presidency, but these relationships were significant only for the more educated respondents (Johnson et al. 1989:394-395; Burton et al. 1989:353). It seems that among the more educated, conservative religious beliefs and values are more likely to be related to political philosophy (Burton et al. 1989:355).

When we examined social traditionalism, we found it to be important for conservative Protestants, but not for Catholics. Also, as a moral foundation for an attitude toward abortion, traditionalism was significant for the less educated, but not for the more educated.

Our data indicate that in the fall of 1989 pro-life people were still significantly more likely to base their vote on a candidate's stand on the abortion issue than were pro- choice people. This finding is consistent with the results of nationwide samples from 1972 to 1987 (Scott and Schuman 1988) and with a Gallup survey which was, like our study, conducted in the fall of 1989 (Colasanto and DeStefano 1989). Moreover, a tendency for single-issue voting among pro-life people was not related to political party affiliation but rather to religious affiliation (Wald et al. 1988). For pro-lifers, being a part of a pro-life religious environment seems to mobilize them to take a political stand on abortion. Furthermore, by focusing members' attention on the abortion question, conservative churches in effect have helped remove class issues, such as discrimination and poverty, from the electoral process.

The debate about legalized abortion is a battleground for people possessing different values. To a significant extent, it is a fight between individualists and traditionalists. Among the latter are those who are committed to the puritanical moral tradition; political conservatives who emphasize the need for tradition as the basis for social order; and those, perhaps influenced by the natural law tradition, who seek to prevent human

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

THE ABORTION CONTROVERSY 45

manipulation of innocent life. Obviously, religion is an important influence in this cultural context. Moreover, religion seems to play a role in mobilizing political action on the abortion issue. However, it should not be overlooked that the variables in the model worked less well in explaining the attitudes of the less educated. Possibly, such people are more influenced by events of the moment than by long-term ideological commitments (Kiecolt and Nelsen 1988; Johnson et al. 1989:395). This issue warrants further study.

REFERENCES

Barnartt, Sharon N., and Richard J. Harris 1982 Recent changes in predictors of abortion at-

titudes. Sociology and Social Research 66 (April):320-334.

Brandmeyer, Gerard, and R. Serge Denisoff 1969 Status politics: An appraisal of the applica-

tion of a concept. Pacific Sociological Review 12(1):5-11.

Burton, Ron, Stephen Johnson, and Joseph Tamney 1989 Education and Fundamentalism. Review of

Religious Research 30(1):344-359. Cable, Sherry

1984 Professionalization in social movement organization: A case study of Pennsylva- nians for biblical morality. Sociological Focus 17 (October):287-304.

Cavanaugh, Michael 1986 Secularization and the politics of tradi-

tionalism The case of the right to life move- ment. Sociological Forum 1(2):251-283.

Clarke, Alan 1987 Moral protest, status defense, and the anti-

abortion campaign. British Journal of Sociology 38(1):2.

Cleghorn, J. Stephen 1986 Research note on Cardinal Bernardin's

'seamless garment'. Review of Religious Research 28(2):129-142.

Clymer, Adam 1984 Long-range hope for Republicans is found

in polls. New York Times, 11 November, Y1, Y16.

Colasanto, Diana, and Linda DeStefano 1989 Abortion: 'Pro-choice' position stirs increas-

ed activism in abortion battle. The Gallup Report, No. 289 (October):16-20.

Dionne Jr., E. J. 1989 Poll of abortion finds the nation is sharply

divided. New York Times, 26 April, Y1, Y13.

Ellis, John 1989 "Wedge issue" that pierces GOP's Heart.

Los Angeles Times, 19 November, M3, M8. Flanigan, William, and Nancy Zingale

1987 Political behavior of the American elec- torate. Boston: Allyn and Bacon.

Gallup, Jr., George, and Jim Castelli

1987 Opinions stable on abortion. Indianapolis News, 17 January, p. 3.

Gallup Report 1989 Attitudes on abortion little changed since

Supreme Court's 1973 ruling. No. 281 (February):16-19.

Goldman, Ari L. 1984 New York's controversial archbishop. The

New York 7imes Magazine, 14 October, pp. 38-42, 46-51, and 70.

Granberg, Donald, and James Burlison 1983 The abortion issue in the 1980 elections.

Family Planning Perspectives 15 (September/October):231-238.

Harrison, Beverly Wildung 1983 Our right to choose. Boston: Beacon.

Himmelstein, Jerome L. 1983 The new right. In The New Christian Right,

edited by Robert C. Liebman and Robert Wuthnow, 13-30.

1986 The social basis of antifeminism: Religious networks and culture. Joural for the Scien- tific Study of Religion 25(1):1.

Himmelstein, Jerome L., and James A. McRae, Jr. 1988 Social issues and socio-economic status.

Public Opinion Quarterly 52(4):492-512. Hugick, Larry

1989 Abortion: Majority critical of abortion deci- sion, but most Americans favor some new restrictions. The Gallup Report, No. 286 (July):5-12.

Inglehart, Ron 1990 Cultural shifts. Princeton, N.J.: Princeton

University Press. Jelen, Ted

1990 Religious belief and attitude constraint. Journal for the Scientific Study of Religion 29(1):118-125.

Johnson, Mary 1990 An empirical test of American attitudes

toward the consistent ethic of life. Un- published paper.

Johnson, Stephen D. 1978 Judgments of equity and vote in a presiden-

tial election. Sociological Focus 11(2): 161-172.

1986 The Christian right in Middletown. In The

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions

46 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION

Political Role of Religion in the United States, edited by Stephen D. Johnson and Joseph B. Tamney, 181-189.

Johnson, Stephen D., and Joseph B. Tamney 1982 The Christian right and the 1980 presiden-

tial election. Journal for the Scientific Study of Religion 21(2):123-131.

1985 The impact of the Christian right on the 1984 presidential election. Review of Religious Research 27(2):124-133.

1988 Factors related to inconsistent life-views. Review of Religious Research 30(1):40-46.

Johnson, Stephen D., Joseph B. Tamney, and Ronald Burton 1989 Pat Robertson: Who supported his can-

didacy for president? Journal for the Scien- tific Study of Religion 28(4):387-399.

n.d. Economic satisfaction vs. moral conser- vatism in the 1988 presidential election. Sociological Focus, forthcoming.

Kiecolt, K. Jill, and Hart M. Nelsen 1988 The structuring of political attitudes

among liberal and conservative Pro- testants. Journal for the Scientific Study of Religion 27(1):48-59.

Kirkpatrick, Jeane J. 1982 Dictatorships and double standards. New

York: Simon and Schuster. Lenski, Gerhard

1971 Conservatives and radicals. In Contem- porary Sociological Theory, edited by Fred E. Katz, 220-222.

Lewin, Tamar 1988 Abortion's foes switch tactics - and

prepare for a new day. New York Times, 27 November, E9.

Lipset, Seymour 1985 Beyond 1984: The future of American

politics. Tocqueville Review 7(2):159-178. Lipset, Seymour, and Earl Raab

1981 The election and the evangelicals. Commen- tary 1(1):25-31.

Luker, Kristin 1984 Abortion and the politics of motherhood.

Berkeley: University of California. Lynd, Robert S., and Helen M. Lynd

1929 Middletown. New York: Harcourt, Brace, and World.

Neitz, Mary Jo 1981 Family, state, and God: Ideologies of the

right-to-life movement. Sociological Analysis 42(3):265-276.

Newsweek 1989 The future of abortion: The court drills a

crack in the foundation of Roe. 17 July, pp. 14-16.

New York Times 1989 What is right and wrong with Roe vs.

Wade? The view from friends of the court. 23 April, E6.

Pedhazur, Elazar 1982 Multiple regression in behavioral research.

New York: Holt, Rinehart and Winston. Plutzer, Eric

1988 Work life, family life, and women's support of feminism. American Sociological Review 53(4):640-649.

Ridgeway, James 1985 The profile juggernaught. The Village

Voice, 16 July, pp. 28-29. Rossi, Alice

1972 Sex equality: The beginnings of ideology. In Toward a sociology of women, edited by Constantina Safilios-Rothchild, 344-353. Lexington, Mass.: Xerox.

Schwartz, Amy E. 1985 Bitter Pill. The New Republic, 18 February,

pp. 10-12. Scott, Jacqueline

1989 Conflicting beliefs about abortion: Legal ap- proval and moral doubts. Social Psychology Quarterly 52(4):319-326.

Scott, Jacqueline, and Howard Schuman 1988 Attitude strength and social action in the

abortion dispute. American Sociological Review 53(5):785-793.

Tamney, Joseph B. 1986 Religion and the abortion issue. In The

Political Role of Religion in the United States. See Johnson 1986.

Tamney, Joseph B., and Stephen D. Johnson 1988 Explaining support for the moral majority.

Sociological Forum 3(2):234-256. Traugott, Michael W., and Maris A. Vinovskis

1980 Abortion and the 1978 congressional elec- tions. Family Planning Perspectives 12 (September/October):238-246.

USA Today 1986 Inquiry topic: Abortion. 12 June, A9.

Wald, Kenneth, Dennis Owen, and Samuel Hill, Jr. 1988 Churches as political communities.

American Political Science Review 82(2): 531-548.

1989 Evangelical politics and status issues. Jour- nal for the Scientific Study of Religion 28(1): 1-16.

Wilcox, Clyde 1989 Political action committees and abortion: A

longitudinal analysis. Women and Politics 9(1):1-19.

Willis, Ellen 1985 Putting women back into the abortion

debate. The Village Voice 30 (16 July): 15-16, 24.

Woodrum, Eric 1988 Moral conservatism and the 1984 presiden-

tial election. Journal for the Scientific Study of Religion 27(2):192-210.

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:08:12 UTC All use subject to JSTOR Terms and Conditions