Trends in attitudes toward abortion, 1972-1975.

Two-thirds of a nationwide random sample of Americans polled in February 1976 said they agree with the statement that "the right of a woman to have an abortion should be left entirely up to the woman and her doctor." For the first time in the more than 25 years that Americans have been surveyed on their views regarding abortion, substantial majorities of women (69 percent), persons over 65 years of age (57 percent), Catholics (60 percent), blacks (67 percent) and people with less than a high school education (59 percent)-segments of the population that had been somewhat more reserved in their views on abortion-also agreed with the statement, documenting a growing consensus on this question among all Americans.(1)

Successive polls taken in the 1960s and early 1970s posed abortion questions in a variety of ways: Should it be legalized? In what circumstances, if any, should it be permissible? Who should make the decision? All of these surveys reported growing support from virtually all sectors of society for the right of women or couples to elect termination of pregnancy in consultation with their physicians.(2) During the same period, U.S. women had increasing first-hand experience with legal abortion. Between 1969 and 1975, some 3.5 million women obtained legal abortions in the United States-one in 14 of all women of reproductive age.(3)

Nonetheless, abortion remains a volatile political issue, in large part because of the activities of a relatively small but committed and articulate group of men and women who steadfastly oppose it on any grounds, maintaining that abortion should not be an option even in a pluralistic society. Leadership in the campaign to make abortion a criminal offense (as it was in most U.S. states until the mid-1960s) is provided by the Catholic hierarchy, especially the bishops, who, in a move unprecedented in American history, committed themselves last November to organize public and political opposition to abortion in every diocese and every election district in the land. The objective of the campaign is, in the words of the Jesuit publication America, "to secure antiabortion laws from the legislative, judicial and administrative departments of government."(4)

Considerable emphasis has been placed by antiabortion groups on obtaining a constitutional amendment that would reverse the Supreme Court decisions, and prohibit abortions in all circumstances (except, perhaps, to save the pregnant woman's life). Thirty-nine percent of Americans surveyed by the Louis Harris organization in early 1976 said that they opposed the Supreme Court decisions;(5) and 45 percent reported to the Gallup Poll that they favored a constitutional amendment that would prohibit abortions except in life-threatening circumstances.(6) These responses seem at variance with the overwhelming approval of legal abortion-at least where health is threatened, or where there is the likelihood of congenital defect, or where the pregnancy is the result of rape-that has been evident in a variety of polls taken since 1970.(0)

In light of the apparent ambivalence of some of these findings and in the face of the antiabortion campaign, because politicians (in office and running for office) are understandably unwilling to commit themselves to what they perceive are controversial or unpopular issues (in this instance, support of the Supreme Court's 1973 abortion decisions), it is important to understand and to make known, so far as possible, the nuances of the public's attitudes concerning abortion. There can be little quarrel with the often-expressed view that the way a question is posed influences the reply,(7) and that simply asking whether the decision to have an abortion should be up to a woman and her doctor, or whether an abortion should be easier or harder to obtain, provides only a rough estimate of the degree of public support for its decriminalization. Therefore, the responses to five virtually identical surveys (1965 and 1972-1975) asking whether or not an abortion is acceptable in six specific situations, provide a u nique opportunity to evaluate trends in abortion attitudes before and following the 1973 Supreme Court decisions. (The responses, however, may well underestimate the absolute support for or opposition to abortion in the enumerated situations, just because the respondents were provided with a range of choices.) The surveys-fielded to similar samples of Americans by the National Fertility Study (NFS) in 1965 and by the University of Chicago's National Opinion Research Center (NORC) in 1972-1975 asked the following question: Please tell me Whether or not you think it should be possible for a pregnant woman to obtain a legal abortion

* if the woman's (own) health is seriously endangered by the pregnancy ("own" was inserted in the 1973-1975 surveys but not in the 1972 question);

* if she became pregnant as a result of rape;

* if there is a strong chance of a serious defect in the baby;

* if the family has a very low income and cannot afford any more children;

* if she is not married and does not want to marry the man;

* if she is married and does not want any more children.

Responses were coded "yes, no, "don't know," or "no answer." The first three situations are usually described as the 'hard' reasons, those which are essentially beyond a woman's control; the rest are the 'soft' reasons, those over which a woman might be able to exercise some degree of control. (*)

This article analyzes the responses to the last four surveys (1972-1975, inclusive), the first one conducted before the Supreme Court's 1973 abortion decisions, the other three subsequent to the decisions. We find that following a sharp increase in public support for all six reasons after the Court's action, a plateauing of support appears to have occurred. The article is concerned with the possible reasons for this trend and with other changes in public opinion that may be attributable to the Supreme Court's decisions.

Findings

Table 1 shows that while all three hard reasons received widespread approval in every year, the soft reasons had substantially less support. In no year did fewer than three-fourths of respondents approve of abortion for any of the hard reasons. However, before the Supreme Court decisions, none of the soft reasons obtained approval from as many as half of all respondents; afterward, about one-half of those surveyed approved of abortion for reasons of poverty, and nearly half approved for the other two soft reasons. (+)

Important as these absolute rates are, the trend is most interesting. Following the substantial increase in approval for all reasons in 1973, two months after the decisions, approval did not continue to increase between 1973 and 1975.

A clearer picture of the structure of attitudes toward abortion can be gained by considering all six situational factors at once. By means of a Guttman scale, we compared nine patterns (++) of response to the question on abortion in the 1965 NFS with the patterns of response to the NORC surveys.

Table 2 documents two major changes in response patterns in the NORC surveys compared with those in the 1965 NFS. (The authors of a special analysis of the NFS response patterns reported that there were essentially three patterns: overwhelming predisposition to favor abortion if the mother's health is seriously endangered; even division on the questions of deformity and rape; and overwhelming opposition for the soft indications. (8)).

The first significant change is that there has been a sizable increase in the percentage of people who approve of abortion for all of the reasons, and a decrease in the percentage of those who do not approve of abortion under any circumstances. The second change is that approval of abortion if the mother's health is endangered is very likely to signify approval of abortion for the other two hard reasons as well. Similarly, approval of abortion for any of the soft reasons is likely to signify approval of abortion under all of the six circumstances. Thus, there are essentially two predominant response patterns: approval of abortion for the hard reasons only, and approval of abortion under all of the six circumstances. This defines a third category by exclusion-unequivocal disapproval of abortion. (This category, however, contains a very small number of respondents, which is why we refer to two predominant patterns.)

The results of a special statistical analysis we performed confirm that the six situational factors are scalable and that the scale is probably two-dimensional. (*) This permitted us to construct a variable which separates respondents into three meaningful categories: The first consists of people who said that they do not approve of abortion under any of the circumstances cited (scale score 0); the second consists of those who approve of it for the first two or for all three bard reasons (scale scores 2 and 3); the third consists of those who approve of abortion for all the hard reasons plus the first two or all three of the soft reasons (scale scores 5 and 6). People responding "don't know" or giving no response to any of the six items were excluded from the analysis. These summary measures are different from those used previously.

Previous researchers have used the Guttman scale score itself (or some minor variant) as the single measure of attitude, assigning nonscale or nonpattern respondents to their nearest or closest scale score. (9) Others have dichotomized on the basis of the number of approving responses. (10) We believe that dividing the sample on the basis of 0-3 approvals and 4-6 approvals leaves too many ambiguities for meaningful differentiation. Those who do not approve of abortion in any circumstances are expressing a substantially different viewpoint from those who approve of it at all. Furthermore, respondents on scales 1, 2a, 4 and 4a (which are excluded in our summary measures) are probably reasoning differently from those responding on the scale patterns we included (scale scores 0, 2-3 and 5-6). We believe that those responding according to patterns 4 or 4a may be regarded as essentially moderate in their views toward abortion, but there is one soft situation in which they would approve of abortion, and that situat ion is closely but not consistently linked with the more 'liberal' position. (+) Limiting attention to a special subset of 'consistent' respondents excludes a considerable number of people from the analysis. (++)

After such exclusion, the composition of the subsample was compared to the original NORC respondent pool for all four years on the basis of sex, race, religion, church attendance, income, education, region of residence, and size of place of residence. The only important deviations in the subsample were with respect to education and income, where the criteria for exclusion caused a small overrepresentation of the better educated and of those with higher incomes. (ss) This discrepancy is small enough, in our judgment, to justify use of the subsample in the following analyses.

Data from the 1972-1975 NORC surveys continue to show a strong relationship between the education and religion variables and attitudes toward abortion. But these are not the only two factors that affect such attitudes. Sex, race, region of residence and parity also have been shown to influence them. One problem encountered in the analysis of these attitudes and attitude differentials is that their correlates are not independent of each other, even though the mode of presentation suggests, at times, that this is the case. We know, too, that attitudes do not necessarily remain constant overtime. Remaining cognizant of these two methodological problems, we examine recent trends in attitudes toward abortion, discussing changes in the correlates of those attitudes.

Education and Religion

In 1972, the level of formal education remained the best predictor of attitudes toward abortion. After the Supreme Court decisions, differences in attitude between educational levels decreased rapidly, as Table 3 shows, largely because of the liberalization of attitudes among the least educated. That is, the percentage of those with less than a high school education disapproving of abortion for all reasons decreased substantially; while those expressing categorical approval increased by about the same amount. This shift results in the stability of the percentage of non--high school graduates approving for the hard reasons only. The distribution of respondents with a high school education or more remains remarkably constant during the entire 1972--1975 period.

The overall pattern reverses the trend of the 1960s. (11) During that decade, the differentials between educational levels first emerged, with the more highly educated becoming more liberal in their attitudes toward abortion, and the distribution of opinion among the less educated remaining fairly stable.

In 1972 education was still the strongest determinant of abortion attitude, but by 1974 religion had also become an important discriminating factor. We limit our discussion of this factor to a comparison of the attitudes of Protestants and Catholics by the degree of their commitment to the norms of their respective religions. (*) We measured religious commitment by frequency of church attendance and by self-report of commitment (the former available for all four years, the latter available only for the 1974 and 1975 surveys). One would expect similar results from each variable since the two are highly correlated. (+)

As may be seen in Table 4, among people who attend church once a month or less, there is little difference between the distributions of approval of Protestants and Catholics in any of the four years. Among those who attend church more than once a month, there is a consistent difference, in the expected direction, between Catholics and Protestants. Interesting as it is to find religious commitment to be a strong determinant of abortion attitude, the trends are perhaps of more importance. Distribution of opinion among those who seldom attend church remains fairly constant. Among more frequent church attenders, however, there is a strikingly different pattern. In 1973, after the Supreme Court decisions, there is a substantial increase in approval among both committed Protestants and Catholics, although the increase among Protestants is for soft reasons, while among Catholics this increase is for hard reasons only.

In 1974, the distribution of opinion among more religious Protestants remains fairly stable, while opinion among churchgoing Catholics changes again, this time in the direction of the extreme positions (with the shift to categorical approval more pronounced than the shift toward categorical disapproval).

We might speculate that the sharp division among Catholics is one result of the emergence of abortion as an issue. That it occurred among Catholics and not among Protestants might be the result of the strong reaction of the Catholic clergy to the Supreme Court decisions. This reaction might have placed the more committed Catholics in a position in which generally accepted standards, those of the Church and of the Court, were in conflict. Possibly, individual Catholics resolved the conflict by moving to one of the extreme positions (total approval or total disapproval) and avoiding a middle-of-the-road opinion. With intensified political action by the Church, Catholics who are regular attenders might be expected to hold a more conservative position on abortion. This appears to be the case, as indicated by the 1975 data for churchgoing Catholics showing a reduction in categorical approval of abortion to the 1972-1973 levels.

One might expect a disproportionate number of Catholics who do not approve of abortion under any circumstances to express greater confidence in their religion than in the Supreme Court. Table 5 presents data to test this expectation. (Because of the small numbers involved, data from the 1973, 1974 and 1975 surveys have been combined in the table.) As expected, a lower percentage of those who disapprove of abortion than of those who approve express high confidence in the Supreme Court, and a higher percentage of those who disapprove of abortion express high confidence in organized religion. This is true for both Protestants and Catholics for the three years following the Supreme Court decisions. It is notable that a significantly greater proportion of Catholics than Protestants who disapprove of abortion express high confidence in organized religion.

Age, Sex and Race

Next we consider the effects of age, sex and race on attitudes toward abortion. (Even if these factors do not have a direct effect, differences in attitudes may result through their interaction with social factors.)

Many surveys made during the middle and late 1960s found that a larger proportion of younger than older persons disapproved of abortion. It has been suggested that this might have been due to a combination of decreasing fertility pressure on the young and increasingly high standards of contraceptive effectiveness, the implication being that abortion was viewed as a "last resort" method of birth control by younger respondents. (12) Our analysis shows that by 1972 there was a virtually complete reversal of this situation. Younger age groups expressed more liberal opinions than older age groups. After the Supreme Court decisions, the youngest age group showed no significant change in attitudes toward abortion until 1975, when there was a noticeable decrease in approval, from 69 percent to 63 percent. The distribution of opinions among those over 30 remained fairly stable. (Differences by age not shown in tables.)

There was also, in the 1960s, an increasing gap reported between the attitudes on abortion held by men and women. (13) In 1972, however, while there are hints of this difference, it is not statistically significant. In 1973, immediately following the Supreme Court decisions, the differential reemerges slightly; but in 1974, it is in the opposite direction. By 1975, the differential again shifts in direction, with the proportions of men and women similar to those of 1973.

Educational differences in the sample partially explain the lag in liberalization of women's attitudes relative to those held by men (see Table 6). Among high school graduates and among those who have attended college, there is little difference from 1973 between men's and women's attitudes on abortion. For those without a high school diploma, the difference in 1973 is significant, women being less approving than men for both hard and soft reasons. By 1974, there remains no significant difference between men and women at any educational level.

When abortion attitudes are considered by race, whites are found to hold more liberal views than nonwhites. In 1972, almost one-third of the nonwhite population, compared with 11 percent of the white population, categorically disapproved of abortion (see Table 7). After the Supreme Court decisions, there was a significant decline in the percentage of nonwhites categorically disapproving of abortion (as low as 12 percent in 1974); however, the increase in approval was for hard reasons (from 23 percent in 1972 to 43 percent in 1974) rather than for hard and soft reasons together (which remained constant at 45 percent). In 1975, the percentage of nonwhites responding at each extreme increased, with categorical disapproval increasing to 16 percent and categorical approval increasing to 58 percent (about the same level as obtained for whites).

When racial differences are examined by educational attainment, different trends are observed. In 1972, more than half the nonwhite sample without a high school diploma disapproved of abortion. In subsequent years, this high percentage of disapproval fell to a low of 20 percent in 1974 (and remained at about that level in 1975). About four-fifths of the growth in approval between 1972 and 1975 among nonwhites of low education is attributable to an increase in the number of those approving for both the hard and soft reasons, and one-fifth to an increase in the proportion approving for the hard reasons only.

Among the more highly educated non-whites the situation is different. In 1972, almost two-thirds of this group approved of abortion for almost any reason, and the distribution of opinions of nonwhites who had a high school diploma or who had attended college was similar to that of whites of the same educational level. (Until 1972, the distribution of opinions of nonwhites had diverged from that of whites.) In 1973 and 1974, there appears to have been a significant decrease in the proportion of better educated nonwhites approving of abortion for all reasons; they seem to have shifted to an acceptance of the hard reasons only. By 1975, the situation had changed again, with the distribution of opinion among the nonwhite sample more nearly approximating the distribution of opinion for the white sample.

Residence

Place of residence has been suggested as a possible correlate of attitudes toward abortion. When region of residence is considered, the observed trend is similar to that reported by others.(14) In 1972, very little difference in attitudes appears except for the expected higher rate of disapproval in the South and Midwest (16 percent) than in the East and West (10 percent). In 1973 and 1974, there is no difference among regions in the rates of categorical disapproval, but the people of the South and Midwest still hold a more conservative position than do people living in other regions. (In 1973, the proportions approving of abortion were 57 percent in the South and Midwest compared to 67 percent in the East and West. In 1974, they were 55 percent compared with 72 percent.) In 1975, however, there is an increase in all regions in the percentage of those disapproving of abortion for any reason (seven percent in 1974 compared to nine percent in 1975), causing the distributions of opinion for the South and Midwest to resemble those of 1972. (Regional data are not shown in tables.)

Other investigators have explained this differential at least partly in terms of the influence of the fundamentalist religions which are strong in the South and Midwest.(15) In all four years, Protestants (in general) in the South were more conservative than Protestants elsewhere.(*) This difference was due mainly to the relatively conservative views of Baptists in the South (even compared to Baptists in other regions).

When level of education and region of residence are examined together, another pattern emerges (not shown in tables). In 1972, there was little difference among regions in distribution of opinion for all educational levels. The possible exception is for Southerners without a high school diploma, who tended to disapprove of abortion more than their counterparts in the other regions (32 percent total disapproval compared with 20-26 percent for other regions). By 1973, very noticable differences appeared. The distribution of attitudes remained stable for those with at least a high school diploma or some college, with an overall categorical approval rate of 67 percent and regional averages ranging from 62 to 71 percent. But among those without a high school diploma, people in the East and Far West moved closer to the more highly educated groups in their rate of unequivocal approval of abortion, increasing from 42 percent to 61 percent, while Southerners and Midwesterners without a high school diploma tended to a pprove of abortion only for the hard reasons. By 1974, regional differences existed at each educational level. These differences persisted into 1975.([dagger])

To examine differentials in attitude toward abortion based on widely used measures of traditional value structures, we examined the differentials by size of place of residence.([double dagger]

Not surprisingly, in 1972, people in rural areas were least favorable toward abortion. People in suburbs were the most favorable, and those in urban areas were in the middle. In 1973, attitudes were more liberal regardless of the size of place of residence, but the largest increase in approval was recorded in the rural areas-from 41 percent unequivocal approval in 1972 to 59 percent in 1973 (see Table 8). The percentage of rural people expressing approval in five and six situations had risen to almost the same level as in the urban regions. In 1974, except among rural residents, the views of people in all areas remained about the same as in 1973. Fewer rural residents in 1974 approved of abortion for both hard and soft reasons (50 percent), while more approved for the hard reasons only (an increase from 34 percent in 1973 to 43 percent in 1974). In 1975, there was a movement toward a more conservative position among rural people (an increase in unequivocal disapproval from seven percent to 16 percent), while those who approved for all reasons remained at 1974 levels (50 percent). This regression might have been a result of the church-based reaction to the Supreme Court decisions. If such is the case, we would expect different trends among rural people based on church attendance.

Table 8 shows two things: There is a difference in attitude between rural and nonrural people; and there are the expected differences among rural residents. For both groups of rural people- those who attend church once a month or less, and those who attend more often-there is a large jump in the rate of approval just after the Supreme Court decisions. However, during the next two years, unequivocal support among less frequent church attenders in rural areas drops off gradually; while among frequent churchgoers there is a sharp drop in 1974, followed by an increase to the 1973 level in 1975. These results suggest that if the Church influenced the regression of approval observed among rural people after 1973, that influence was not long-lasting.

Summary and Conclusions

Attitude differentials based on a number of characteristics have been presented. It appears that variables related to education and religion continue to be of primary importance in the study of abortion attitudes.

In the early 1970s, the level of educational attainment continued to have a strong influence on attitudes toward abortion, but the nature of this influence began to change. After 1972, there was no change in attitude among those who had been graduated from high school or who had had some college, but among those with less than a high school education, there was a noticeable switch to a more liberal position.

Among the factors that may be responsible for the shift in opinion among persons with less than a high school education is the relative decrease in the economic status of such persons. (0) A falling standard of living may have increased both the 'opportunity' costs and real costs associated with children to such an extent that the psychic and economic benefits which had been thought to result from having children were substantially less. It is also possible that more widespread acceptance of the positions and arguments arising from the women's movement may have reduced the value to the lower class of the "noneconomic goals and interests" embodied in children.

We noted also that religion is an important determinant, or at least correlate, of attitudes toward abortion. We found that religious commitment accounted for differences in attitudes to a much greater degree than did denominational identification. In the 1970s, there has been a negligible rate of disapproval of abortion among those with a low degree of religious commitment. We expect this differential to persist.

Nonetheless, there is still a salient differential with regard to religion based on denominational identification. We noted a decrease in resistance to abortion among both Catholics and Protestants during the early 1970s. However, in the years following the 1973 Supreme Court decisions, there was an apparent reversion in the opinions of churchgoing Catholics. We speculated that this might have been due to conflicting pressures created by the differing positions of the Church and the Supreme Court, and we presented evidence as to expressed confidence in religion and in the Court which supported this hypothesis. Corroborative evidence concerning the effect of the Catholic Church's position on contraception is provided by a recent study by NORC which attributes nearly onehalf of the decline in attendance at mass between 1963 and 1974 to the Church's stand on birth control. (16)

If this trend continues, we might expect a greater proportion of Catholics, those who are otherwise committed to church orthodoxy, to join in the move for even more liberal positions on abortion than presently exist. But, we would also expect certain sectors of the Catholic community to express even more staunch opposition than in the past to more widespread utilization of abortion.

It is notable that the 1973 NORC survey, fielded just two months after the 1973 Supreme Court abortion decisions, showed a remarkable liberalization of abortion attitudes on the part of all groups and subgroups of American society. That is, the very fact of the decisions apparently caused a rapid shift in abortion attitudes. Very little change occurred in the years following the decisions, although between 1973 and 1975, some 2.6 million U.S. women had legal abortions. This suggests that behavioral changes, particularly where such traditionally controversial subjects as abortion are concerned, may take some considerable time before they are reflected in changed attitudes. (Thus, U.S. women had been bearing children at the twochild level for some years before indicating in national surveys that they wanted or expected fewer than the three children averaged during the 'baby boom'.) The effect of changing the law, however, may-as in this case-have an immediately legitimating effect on public opinion.

(+.) Those polled in the NORC surveys included a national random sample of men and women ranging in age from 18 years to 65 years and older; the NFS was addressed to a sample of married women under 55 years of age.

William Ray Arney is an assistant professor in the Department of Sociology, Dartmouth College, and William H. Trescher is an alumnus of Dartmouth, now employed at First National Bank of Boston. This article is adapted and updated from a paper presented at the annual meeting of the American Statistical Association, held Aug. 26, 1975, In Atlanta. The authors thank the National Opinion Research Center for making the data used in this paper available through Project IMPRESS at Dartmouth.

(*.) See, for example, Table 2 (below) and reference 2.

(++.) In the NFS question, the six situations were the same but the introductory remark was, "Would it be all right for a woman to have a pregnancy interrupted...?"

(*.)The original order of the situations alternated hard and soft reasons to reduce response bias.

(+.)Disapproval for the hard reasons drew no more than 20 percent of responses in any year (rape and defect, in 1972); after the Supreme Court decisions, only poverty among married couples drew even a bare majority of disapproval (50-52 percent).

(++.)The nine patterns account for approximately 85 percent of the people who responded yes or no to every question, and for approximately three-fourths of the entire sample in each of the NORO surveys.

(*.) We conducted a principal components analysis on the six items. Resultant factors were rotated using a varimax solution. For each year, two orthogonal factors emerged which accounted for 75-80 percent of the trace. In all four studies, the situations denoted as soft reasons for approval loaded heavily on the first factor, while the hard reasons loaded heavily on the second. Loadings of the soft reasons on the second factor were negligible, as were the loadings of the hard reasons on the first factor. This lends empirical support to the notion that people are responding on two different dimensions rather than on a unidimensional scale of attitudes toward abortion.

(+.) It should be stressed that the liberal-conservative dimension does not necessarily have any meaning for the respondents.

(++.) The percentages excluded are the following: 41 percent in 1972; 33 percent in 1973; 36 percent in 1974; and 30 percent in 1975.

(*.) Those who report themselves to be Jewish or of a religion other than Catholic, Protestant or Jewish, or who claim no religious affilation, hold very liberal attitudes toward abortion, with about 85-90 percent approving abortion for five or all six situational reasons.

(+.) In fact, for 1974 and 1975, differences between results using the two indicators are so small that tables based on self-reported religious intensity are not presented here.

(*.) During the 1960s, the relative economic situation of those in the lower income percentiles improved slightly. People in the lowest 20 percent experienced the fastest rate of growth in their income until about 1968. At that point, income growth leveled off, with the plateau persisting until 1971. only those in the top five percent of the income distribution experienced growth which was at all comparable to that experienced by most people during the 1960s. The net effect was a decrease in the relative economic status of those in the lower income groups during the 1970s. (See: Office of Management and Budget, Executive Office of the President, Social Indicators, 1973, U.S. Government Printing Office, Washington, D.C., 1973, Chap. 5.)

References

(1.) H. Reinhold, "Poll Finds Voters Judging '76 Rivals on Personality," New York Times, Feb. 13, 1976, P. 1; and H. Liberman, New York Times, personal communication, Feb. 10, 1976.

(2.) J. Blake. "Abortion and Public Opinion: The 1960-1970 Decade," Science, 171:540, 1971; J. Blake, "Elective Abortion and Our Reluctant Citizenry: Research on Public Opinion in the United States," in H. J. Osofsky and J. D. Osofsky, eds., The Abortion Experience: Psychological and Medical Impact, Harper & Row, Hagerstown, Md., 1973, p. 447; E. F. Jones and C. F. Westoff, "Attitudes Toward Abortion in the United States in 1970 and the Trend Since 1965," in Commission on Population Growth and the American Future, Demographic and Social Aspects of Population Growth, C. F. Westoff and R. Parke, Jr., eds., Vol. 1 of Commission Research Reports, U. S. Government Printing Office, Washington, D.C., 1972, p. 569; R. Pomeroy and L. C. Landman, "Public Opinion Trends: Elective Abortion and Birth Control Services to Teenagers," Family Planning Perspectives, Vol. 4, No. 4, 1972, p. 44; and N. B. Ryder and C. F. Westoff, Reproduction in the United States, 1965, Princeton University Press, Princeton, N.J., 1971.

(3.) E. Weinstock, C. Tietze, F. S. Jaffe and J. G. Dryfoos, "Abortion Need and Services in the United States, 1974-1975," Family Planning Perspectives, 8:58,1976.

(4.) "The Bishops' Plan for Pro-Life Activities," editorial, America, Dec. 27, 1975, p.454.

(5.) L. Harris, "Poll: Abortion Is a Safe Issue," New York Pa at, Apr. 12, 1976, p. 26.

(6.) American Institute of Public Opinion, "Women in America," The Gallup Opinion Index, Report No. 128, Princeton, N.J., Mar. 1976.

(7) J. Blake, 1973, op. cit.

(8) C. F. Westoff, E. C. Moore and N. B. Ryder, "The Structure of Attitudes Toward Abortion," Milbank Memorial Fund Quarterly, 47:11, 1969.

(9.) Ibid.

(10.) G. E. Hendershot and J. W. Grimm, "Abortion Attitudes Among Nurses and Social Workers," American Journal of Public Health, 64:438, 1974.

(11.) J. Blake, 197.1, op. cit.

(12.) Ibid.

(13.) Ibid.

(14.) Ibid.; and D. S. Mileti and L. D. Barnett, "Nine Demographic Factors and Their Relationship to Attitudes Toward Abortion Legislation," Social Biology, 19:43, 1972.

(15.) C. F. Westoff, E. C. Moore and N. B. Ryder, 1969, op. cit., p. 19.

(16.) A. M. Greeley, W. C. McCready and K. McCourt, Catholic Schools in a Declining Church, Sheed & Ward, Kansas City, Kans., 1976, Table 5.10, p. 135.

Table 1

Percent of respondents approving of abortion for each of six reasons, by
reason and year of survey, 1972-1975

Reason (*)  1972    1973 (+)  1974    1975
            (N=     (N=       (N=     (N=
            1,613)  1,504)    1,484)  1,490)

Mother's    83      91        90          88
 health
Rape        74      81        83          80
Defect in
 child      74      82        83          80
Family
 poor       46      52        52          50
Mother un-
 married    40      47        48          46
No more
 children   38      46        45          44

(*)The first three are the hard reasons.

(+)Survey made in Mar. 1973, two months after the Supreme Court
decisions.
Table 2

Percent of respondents to 1965 NFS and 1972-1975 NORC surveys
responding "yes" or "no" to each of the six reasons for abortion, by
pattern of response on a scale of 0-6

Scale                               Reason for abortion

                        Moth-                  Rape      Defect
                        er's                             in
                        health                           child


0                       No                     No        No
1                       Yes                    No        No
2                       Yes                    Yes       No
2a                      Yes                    No        Yes
3                       Yes                    Yes       Yes
4                       Yes                    Yes       Yes
4a                      Yes                    Yes       Yes
5                       Yes                    Yes       Yes
6                       Yes                    Yes       Yes
All other combinations
No. of respondents
 with at least one
 "don't know" or
 "no answer"
Coefficient of
 reproducibility
Coefficient of
 scalability

Scale                        Reason for abortion

                        Family    Mother    No
                        poor      unmar-    more
                                  ried      chil-
                                            dren

0                       No        No        No
1                       No        No        No
2                       No        No        No
2a                      No        No        No
3                       No        No        No
4                       Yes       No        No
4a                      No        Yes       No
5                       Yes       Yes       No
6                       Yes       Yes       Yes
All other combinations
No. of respondents
 with at least one
 "don't know" or
 "no answer"
Coefficient of
 reproducibility
Coefficient of
 scalability

Scale                            % of respondents,by pattern
                                         of response
                        1965                  1972        1973
                        NFS (*)               NORC        NORC



0                         9                    10           5
1                        23                     5           3
2                        11                     3           5
2a                       10                     4           5
3                        24                    18          19
4                         5                     5           2
4a                        2                     3           5
5                         2                     5           3
6                         5                    36          43
All other combinations    9                    12          10
No. of respondents
 with at least one
 "don't know" or
 "no answer"            101                   294         155
Coefficient of
 reproducibility          u                   .93         .94
Coefficient of
 scalability              u                   .80         .82

Scale                    % of respondents,by
                              pattern
                             of response
                        1974      1975
                        NORC      NORC



0                         5         7
1                         3         3
2                         3         3
2a                        4         5
3                        19        19
4                         5         4
4a                        3         6
5                         4         4
6                        42        41
All other combinations   12         8
No. of respondents
 with at least one
 "don't know" or
 "no answer"            196       204
Coefficient of
 reproducibility        .94       .94
Coefficient of
 scalability            .78       .81

(*)N. B. Ryder and C. F. Westoff, Reproduction in the United States,
1965, Princeton University Press, Princeton, N. J., 1971. Table X-2, p.
271 (N = 5.617).

Notes: u = unavailable; percents may not add to 100 because of rounding.
Table 3

Percent distribution of respondents approving various reasons for
abortion, by education, 1972-1975

Education                        1972 % approving
               N         Total       Never     Hard      All

0-11 years     321       100         26        37        37
H.S. graduate  306       100          9        30        60
Atleast
some college   314       100          5        19        75

Education                        1973 % approving
               N         Total       Never     Hard      All

0-11 years     298       100         12        39        49
H.S. graduate  333       100          7        34        59
Atleast
some college   373       100          4        22        74

Education                        1974 % approving
               N         Total       Never     Hard      All

0-11 years     269       100         10        38        52
H.S. graduate  330       100          8        32        60
Atleast
some college   347       100          5        20        75

Education                        1975 % approving
               N         Total       Never     Hard      All

0-11 years     275       100         14        37         49
H.S. graduate  333       100          9        32         59
Atleast
some college   336       100          5        20         75

Note: Percents may not add to 100 because of rounding.
(ss)The following tables compare the percent distributions of the
original sample and subsample according to income and education,
1972-1975. (Percents may not add to 100 because of rounding.)

Income

                       1972                  1973             1974
               Orig.     Sub.      Orig.     Sub.      Orig.

Low            30        26        31        27        31
Moderate       39        40        42        43        38
High           31        34        27        31        32

                  1974             1975
               Sub.      Orig.     Sub.

Low            25        38          36
Moderate       39        29          27
High           35        33          34

Education

                       1972                  1973             1974
               Orig.     Sub.      Orig.     Sub.      Orig.

0-11 years     40        34        37        30        35
H.S. graduate  32        32        32        33        33
At least some
College        27        33        31        37        32

                  1974             1975
               Sub.      Orig.     Sub.

0-11 years     28        36          29
H.S. graduate  35        34          35
At least some
College        37        30          36

Note: Low income in 1972 is defined as less than $6,000; in 1973, 1974
and 1975, it is less than $7,000. Moderate income in 1972 is
$6,000-$12,500; in 1973, it is between $7,000 and $14,000; in 1974 and
1975, the upper boundary is increased to $15,000.
Table 4

Percent distribution of Protestants and Catholics approving various
reasons for abortion, by frequency of church attendance, 1972-1975

Attends [less than or equal to] once per month

Religion and                    1972 % approving
frequency of  N         Total       Never     Hard      All
attendance

Total         400       100          7        27        66
Protestant    313       100          9        26        66
Catholic       87       100          2        32        65

Religion and                    1973 % approving
frequency of  N         Total       Never     Hard      All
attendance

Total         452       100          3        28        69
Protestant    343       100          2        27        71
Catholic      109       100          4        30        66

Religion and                    1974 % approving
frequency of  N         Total       Never     Hard      All
attendance

Total         449       100          3        27        70
Protestant    333       100          4        26        70
Catholic      116       100          2        28        71

Religion and                    1975 % approving
frequency of  N         Total       Never     Hard      All
attendance

Total         456       100          6        27         67
Protestant    333       100          6        28         66
Catholic      123       100          4        26         70

Attends > once per month

Religion and                    1972 % approving
frequency of  N         Total       Never     Hard      All
attendance

Total         417       100         22        37        40
Protestant    275       100         20        36        44
Catholic      142       100         27        39        34

Religion and                    1973 % approving
frequency of  N         Total       Never     Hard      All
attendance

Total         409       100         14        41        45
Protestant    280       100         13        36        51
Catholic      129       100         18        52        30

Religion and                    1974 % approving
frequency of  N         Total       Never     Hard      All
attendance

Total         386       100         14        39        47
Protestant    265       100         10        40        50
Catholic      121       100         23        37        40

Religion and                    1975 % approving
frequency of  N         Total       Never     Hard      All
attendance

Total         367       100         16        37         47
Protestant    264       100         11        35         54
Catholic      103       100         27        42         31

Note: Percents may not add to 100 because of rounding.
Table 5

Percent of Catholics and Protestants expressing "high confidence" in
organized religion and in the Supreme Court, by abortion attitude,
1973-1975

Religion            Religion                Court
and          N           %         N          %
abortion
attitude

Catholics
Never           88       50           87     26
Hard           249       41          241     30
All            355       33          354     35

Protestants
Never          123       40          124     24
Hard           540       38          549     29
All          1,082       35        1,094     37
Table 6

Percent of respondents approving of abortion for all reasons, by sex and
education, 1972-1975

Sex and                        1972                  1973           1974
education
                       N         %         N         %         N

Male total             471       60        456       66        434
0-11 years             157       37        131       57        136
H.S. graduate          128       63        118       60        113
At least some college  186       78        207       75        185
Female total           470       55        548       58        512
0-11 years             164       37        167       43        133
H.S. graduate          178       58        215       59        217
At least some college  128       71        166       72        162

Sex and                   1974             1975
education
                       %         N          %

Male total             62        409       65
0-11 years             52        110       50
H.S. graduate          55        128       61
At least some college  74        171       77
Female total           64        535       59
0-11 years             53        165       48
H.S. graduate          63        205       58
At least some college  76        165       73
Table 7

Percent distribution of respondents approving various reasons for
abortion, by race and education, 1972-1975

Race and                         1972 % approving
education
                N                   Total     Never     Hard

White total     805                 100       11        30
0-11 years      254                 100       19        41
H.S.graduate
 or more        551                 100        7        25
Nonwhite total  136                 100       32        23
0-11 years       67                 100       52        22
H.S. graduate
 or more         69                 100       12        25

Race and          1972 %                1973 % approving
                approving
education
                All       N                   Total     Never

White total     59        897                 100        6
0-11 years      40        240                 100        8
H.S.graduate
 or more        68        657                 100        5
Nonwhite total  45        107                 100       20
0-11 years      25         58                 100       28
H.S. graduate
 or more        64         49                 100       10

Race and          1973 % approving           1974 % approving
education
                Hard      All       N                   Total

White total     30        64        855                 100
0-11 years      40        52        229                 100
H.S.graduate
 or more        27        68        626                 100
Nonwhite total  39        41         91                 100
0-11 years      38        34         40                 100
H.S. graduate
 or more        41        49         51                 100

Race and               1974 % approving           1975 % approving
education
                Never     Hard      All       N

White total      7        28        65        867
0-11 years       8        38        54        241
H.S.graduate
 or more         6        24        69        626
Nonwhite total  12        43        45         77
0-11 years      20        38        42         34
H.S. graduate
 or more         6        47        47         43

Race and                    1975 % approving
education
                Total     Never     Hard      All

White total     100        8        30         62
0-11 years      100       13        39         48
H.S.graduate
 or more        100        7        26         67
Nonwhite total  100       16        26         58
0-11 years      100       21        29         50
H.S. graduate
 or more        100       11        23         65

Note: Percents may not add to 100 because of rounding.
Table 8

Percent distribution of respondents approving various reasons for
abortion, in rural and nonrural areas, according to frequency of church
attendance, 1972-1975

Residence                                     1972 % approving
and frequency                     N                   Total     Never
of attendance

Rural total                       118                 100       19
Attends [less than or equal to]
once per month                     55                 100       13
Attends > once
per month                          63                 100       24
Nonrural total                    817                 100       13
Atttends [less than or equal to]
once per month                    451                 100        6
Attends > once
per month                         366                 100       22

Residence                           1972 % approving    1973 % approving
and frequency                     Hard      All       N
of attendance

Rural total                       40        41        121
Attends [less than or equal to]
once per month                    42        45         68
Attends > once
per month                         38        38         53
Nonrural total                    28        59        877
Atttends [less than or equal to]
once per month                    21        74        514
Attends > once
per month                         36        42        363

Residence                                     1973 % approving
and frequency                     Total     Never     Hard      All
of attendance

Rural total                       100        7        34        59
Attends [less than or equal to]
once per month                    100        3        31        66
Attends > once
per month                         100       13        38        49
Nonrural total                    100        7        31        62
Atttends [less than or equal to]
once per month                    100        2        23        75
Attends > once
per month                         100       12        43        44

Residence                                     1974 % approving
and frequency                     N                   Total     Never
of attendance

Rural total                       127                 100        7
Attends [less than or equal to]
once per month                     70                 100        6
Attends > once
per month                          57                 100        9
Nonrural total                    821                 100        7
Atttends [less than or equal to]
once per month                    484                 100        2
Attends > once
per month                         337                 100       14

Residence                           1974 % approving    1975 % approving
and frequency                     Hard      All       N
of attendance

Rural total                       43        50        114
Attends [less than or equal to]
once per month                    36        58         67
Attends > once
per month                         53        39         47
Nonrural total                    28        65        829
Atttends [less than or equal to]
once per month                    21        76        499
Attends > once
per month                         36        49        330

Residence                                     1975 % approving
and frequency                     Total     Never     Hard      All
of attendance

Rural total                       100       16        34        50
Attends [less than or equal to]
once per month                    100       12        37        51
Attends > once
per month                         100       21        30        49
Nonrural total                    100        8        29        63
Atttends [less than or equal to]
once per month                    100        4        23        73
Attends > once
per month                         100       15        37        48

Note: Percents may not add to 100 because of rounding.
[degrees] The following table shows the proportion of Protestants
approving of abortion for all reasons, by region of residence,
1972-1975.

Year         Region
             South                 Other
          N         %         N          %

1972      115       50        212       59
1973      143       58        245       65
1974      109       54        257       65
1975      111       51        249       66

(+)The following table shows the proportion of persons approving of
abortion for all reasons, by education and region of residence,
1974-1975.
Education                                       Year
(in yrs.)                             1974                  1975
and region                    N         %           N       %

East and West
[less than or equal to]11      64       62          66      62
[greater than or equal to]12  235       76         207      73

Midwest and South
[less than or equal to]11      76       45          68      40
[greater than or equal to]12  222       60         241      62

(++)The measure we employed to do this is the NORC Belt Code, which is
divided into four categories:

* Large urban: one of the 100 largest Standard Metropolitan Statistical
Areas (SMSAs).

* Urban: urban counties which have towns of 10,000 people or more and
which are not "large urban" or "suburban".

* Suburban: suburbs of the 100 largest SMSAs; and

* Rural: counties having no towns of 10,000 people or more.