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

Springer is collaborating with JSTOR to digitize, preserve and extend access to Population and Environment.

http://www.jstor.org

Comparison of Pro-Choice and Pro-Life Activists: Their Values, Attitudes, and Beliefs Author(s): Donald Granberg Source: Population and Environment, Vol. 5, No. 2 (Summer, 1982), pp. 75-94 Published by: Springer Stable URL: http://www.jstor.org/stable/27502957 Accessed: 03-11-2015 02:29 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 02:29:28 UTC All use subject to JSTOR Terms and Conditions

Comparison of Pro-Choice and Pro-Life Activists: Their Values, Attitudes; and Beliefs

Donald Granberg

University of Missouri, Columbia

This article reports a survey of two statewide abortion related

organizations: Missouri Citizens for Life and the Abortion Rights Alliance

(N =

524). The two groups differed signficantly on 12 of 18 Rokeach terminal values and on eight of 13 NORC instrumental values. Salvation

was the value which most successfully differentiated the two groups. The two groups were used to establish the construct validity of 18 abortion attitude items. Members of the two groups also differed significantly on each of 18 beliefs, indicating that they tend not only to accept assertions made by their own group but also to reject assertions made by the other

group. Anti-abortion were more likely than pro-abortion respondents to

expect an increase in restrictions on abortion five years from now. Pro abortion respondents attributed significantly more pro-abortion sentiment to U.S. adults than did the anti-abortion respondents.

Several authors have sought to explain variation in abortion

attitudes as a function of demographic and social status variables

(Rossi, 1966; Blake, 1971; Arney & Trescher, 1976; Granberg, 1978). In a recent, comprehensive attempt, eight surveys of U.S. adults

done by the National Opinion Research Center between 1965 and 1978 were used, in which the same six items asked whether a legal abortion should be available.

The cooperation of Maureen McCarthy, Ann O'Donnell and Kathy Edwards is

gratefully acknowledged. This research project benefited greatly from the assistance of

Mary Harbourt, Kathy Dunsford, Sherry Kilgore, Barbara Breen, Janice Warneke, Lynn Taylor, Janice Sato, Beth Wellman Granberg, Karen Quiring, and Patricia Shanks. Data collection was funded by a grant from the Graduate Research Council of the University of

Missouri ?Columbia. The analysis was funded by National Institute of Child Health and Human Development Grant No. 1-HD-82834. Tom Smith of the NORC at the University of

Chicago provided the marginals on the abortion items from the 1980 General Social Survey. Requests for reprints should be sent to Donald Granberg, Center for Research in Social

Behavior, 111 E. Stewart Road, University of Missouri ?Columbia, Columbia, MO 65211.

Population and Environment, Vol. 5(2), Summer 1982 0199-0039/82/1400-0075$02.75?1982 Human Sciences Press 75

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

76

POPULATION AND ENVIRONMENT

In each year, a regression was done with an approval of abortion scale as the dependent variable and sex, age, race,

education, occupational prestige, income, region, and urbanism as

the independent variables. Of these, education was the strongest and the most consistent predictor but, across years, R averaged only .31, indicating that demographic and social status variables accounted for a small percentage of the variation in abortion attitudes (Granberg & Granberg, 1980). This conclusion was

consistent with an earlier analysis by Mileti and Barnet (1972) who

argued that, in order to gain a more complete understanding of abortion attitudes, it will be necessary to take a more social

psychological approach. A second problem in the social science research on abortion is

that it has focused on people to whom the abortion controversy may not be very salient. Representative samples of U.S. adults, some specific occupational groups (e.g., nurses, social workers,

clergy), or undergraduate students have served as subjects in most

studies. The people most committed and involved in the pro- and anti-abortion movements have been generally neglected. Yet, citizens participating in social movement organizations probably have an effect on history out of proportion to their numbers and, therefore, they also deserve the attention of those who would understand the development of a social controversy.

This article reports a study of members of two abortion related social movement organizations, the "Missouri Citizens for Life" (MCL), a group committed to the protection of prenatal life and the "Abortion Rights Alliance" of Missouri (ARA), a group committed to retaining the right of women to choose to have a le

gal abortion. These groups, which have almost antithetical politi cal objectives, are compared on certain social psychological fea tures: values, attitudes, and beliefs.

PROCEDURE

In February, 1980 a mailed survey was sent to representative samples of 377 ARA members and 377 members of MCL. The mailing included a letter from the head of the person's organization, a second

one from the researcher, a four-page questionnaire (printed on both

sides), and a business reply envelope. Usable questionnaires were returned by 292 ARA and 232 MCL members for response rates of about 77 and 62 percent, respectively. Eighty-one percent of ARA and 64

percent of MCL respondents were women. An estimate based on the names in the original mailing lists were 75 and 50 percent women,

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

77

DONALD GRANBERG

respectively, so it appears that in both organizations, women were somewhat more likely to respond than men.

Both of these groups were of higher socioeconomic status than the

average of ILS. society, but relatively speaking, ARA members were

somewhat higher in status than MCL members. Seventy-seven percent of ARA members were college graduates compared to 45 percent of MCL members. About nine percent of MCL members had family incomes of

$50,000 or higher per year compared to 24 percent of the ARA members. The correlations between group membership and education and income were 37 and .18, respectively. The groups also differed slightly in the age categories with 46 percent of MCL and 30 percent of ARA members in the 31-45 age range. ARA members were slightly more likely than those in

MCL to be in the other age categories: 30 and under (23 to 18%), 46-64 (32 to 29%), and 65 and over (16 to 7%). The two groups differed much more

strongly on religious affiliation with 77 percent of the MCL group being Catholic, 16 percent Protestant, four percent Other, and three percent Agnostic or Atheist. By contrast, the ARA group was comprised of only two percent Catholic, 17 percent Jews, 40 percent Protestant, 11 percent

Other, and 30 percent Agnostic or Atheists. Other procedural information, information on pretesting and source

of items, speculation as to why the different response rates occurred, a

copy of the questionnaire, and a full report of initial findings are avail able elsewhere (Granberg, 1980).

RESULTS

Values

Values are abstract states, goals, or means for achieving goals which can be evaluated or ranked according to their desirability or

relative importance. The most comprehensive analysis of human values was that of Rokeach (1973) who distinguished between terminal and instrumental values. Because it seemed to be the

most carefully developed set of values available, the 18 terminal values identified by Rokeach were used in the questionnaire. Also included were the 13 qualities children could develop, as used in the NORC General Social Surveys of 1973, 1975, 1976 and 1978.

Analyses of these NORC surveys suggested the hypothesis that pro abortion people rank curiosity as more important and obedience as less important than anti-abortion people (Granberg & Granberg, 1980). These 13 qualities are similar to the instrumental values identified by Rokeach, and they were used because of their

potential for eventually making comparisons with representative

samples of U.S. adults in addition to the comparison between pro and anti-abortion activists reported here.

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

78

POPULATION AND ENVIRONMENT

Pretesting suggested that the ranking of such sets of values was a difficult and cumbersome task for many people but a pro cedure which nonetheless was capable of producing meaningful and significant results. In view of these considerations, it was

decided that each questionnaire would have one or the other set of

values. Thus, about half the ARA group (randomly determined) ranked the 18 Rokeach terminal values and half ranked the 13

NORC qualities, and the same division of the two forms was made

for the MCL respondents. Table 1 gives the average rank for the 18 terminal values as

evaluated by MCL and ARA members. The MCL and ARA groups differed significantly on 12 of the 18 terminal values. However, by far the largest difference was on the relative ranking of salvation. In fact, MCL respondents tended to rank "salvation" as more

important, on the average, than any of the other 17 values, while

the ARA members tended to rank it as least important. In addition to salvation, the MCL respondents tended to attach significantly

more importance to family security, national security, and inner

harmony than did the ARA members. ARA members tended to rank an exciting life, pleasure, equality, a sense of accomplishment, a

world of beauty, social recognition, a comfortable life, and

freedom as significantly more important than did members of

MCL. The two groups did not differ significantly in their average

ranking of mature love, a world at peace, self-respect, true friend

ship, happiness, and wisdom.

Table 2 presents the average rankings of the MCL and ARA

groups for the 13 qualities used by NORC to measure the values

adults would like to see developed by children. In Table 2, there is no one value which is evaluated as differently by the two groups as

salvation was in Table 1. There were still eight significant differ

ences, however, on the NORC qualities, the largest difference

being on obedience, which was ranked higher by MCL respondents. These respondents also ranked honesty, self-control, and gender appropriate behavior as relatively more important than did ARA

members. ARA members tended to evaluate curiosity, being a

good student, having good sense and sound judgment, and being

responsible as more important than MCL members. The five

qualities on which the two groups did not differ significantly were

having good manners, trying hard to succeed, being neat and

clean, getting along with other children, and being considerate of

others.

Because the ARA members tended to have a significantly

higher level of formal education, and because education is known

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

79

DONALD GRANBERG

Table 1

Average Ranking of 18 Terminal Values by Members of Missouri Citizens for Life and the Abortion Rights Alliance

VALUE

Citizens

for

Life

(N=106)

Abortion

Rights Alliance

(N=121)

t-test

1. A comfortable life 13.4 12.0 2.92** .19

2. Inner harmony 5.1 6.2 -1.98* .13

3. An exciting life 16.0 13.4 5.94*** .36

4. Mature love 7.3 7.0 0.57 .04

5. A sense of accomplish ment 9.1 6.7 4.98*** .31

6. National security 10.3 11.9 -2.90** .19

7. A world at peace 7.6 6.8 1.49 .10

8. Pleasure 16.0 14.2 5.61*** .34

9. A world of beauty 13.6 11.7 4.29*** .27

10. Salvation 3.2 15.2 -19.20*** .79

11. Equality 11.2 8.4 5.23*** .33

12. Self-respect 5.1 4.7 1.11 .07

13. Family security 6.6 8.2 -3.04** .20

14. Social recognition 16.0 14.9 3.05** .20

15. Freedom 6.7 5.7 2.03* .13

16. True friendship 8.5 7.8 1.72 .11

17. Happiness 8.0 8.8 -1.44 .10

18. Wisdom 7.0 7.2 -0.49 .03

Note: Each person was asked to rank these values from 1 (most important) to

18 (least important). Therefore, in the columns giving the average

rank for each value, the lower the number, the greater the importance

given to that value. The r given in the right hand column (without the

sign) indicates the strength of the relationship. * =

p <.05

** = p<.01

*** = p<.001

to be associated with the ranking of values, the above comparisons were repeated using the college graduates only. This was the education category which contained the largest number of people. The differences generally remained when this control was added. Those in the ARA group who were college graduates differed sig

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

80

POPULATION AND ENVIRONMENT

Table 2

Average Ranking of 13 Instrumental Values by

Members of Missouri Citizens for Life and the Abortion Rights Alliance

Citizens Abortion t-test r

for Rights

Life Alliance

VALUE (N=113) (N=142)

The child:

1. Has good manners 9.0

2. Tries hard to succeed 8.7

3. Is honest 2.5

4. Is neat and clean 10.2

5. Has good sense and sound judgment 4.7

6. Has self control 5.4

7. Acts like a girl if she is a girl, or acts like a boy if he is a boy 11.5

8. Gets along well with other children 6.7

9. Obeys parents well 5.4

10. Is responsible 4.2

11. Is considerate of others 4.0

12. Is interested in how and

why things happen 8.3

13. Is a good student 10.5

9.5

8.2

4.0

10.8

2.9

6.5

12.1

7.1

8.9

3.6

4.1

4.7

-1.73

1.33

-5.64***

-1.94

5.59***

-3.17**

-2.16*

-1.09

-10.08***

2.32*

-0.10

9.02***

6.30***

.11

.08

.33

.12

.34

.20

.13

.07

.55

.15

.01

.50

.37

Note: Each person was asked to rank the qualities from 1 (most important) to

13 (least important). Therefore, in the columns giving the average rank

for each quality, the lower the number, the greater the importance attached

to the quality. The r given in the right hand column (without the sign)

indicates the strength of the relationship. * =

p <.05

** = p <.01

*** = p <.001

nificantly from those in the MCL group who were college

graduates on nine of the 18 terminal values. The nine significant differences were, in each instance, in the same direction as for the

groups as a whole, and the ranking of salvation was once again by far the best predictor of group membership among these terminal

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

81

DONALD GRANBERG

values. As to the 13 NORC qualities listed in Table 2, the conclu sions regarding those in the MCL and ARA groups who were

college graduates were the same as for the groups as a whole with

one exception. College graduates in the two groups did not differ

significantly in their average ranking of being "responsible."

Attitudes

Like values, attitudes are (or are reflected by) evaluative,

affectively charged judgments. However, attitudes pertain to some

relatively specific object, practice, or concept. Table 3 reports

comparisons between the MCL and ARA groups on 12 items per

taining to the legal availability of abortion. One might wonder why such items were included in the survey as the responses might be

regarded as a foregone conclusion. However, the 12 items in Table

3 include the items (A,B,C,E,H and K) used routinely by the NORC between 1965 and 1980. These items, although they form the basis

of numerous articles, have never been tested for validity. One

basic way of validating attitude questions is to administer them to

groups whose position on an issue is generally known (Crano &

Brewer, 1973, p. 253; e.g., Singleton & Christiansen, 1977). Second, it was not known in advance how the groups would respond to the

extreme items (D,E,F and L). Third, this section served as an intro

duction to a following section in which respondents attributed atti

tudes to U.S. adults on these same items.

Not surprisingly, the two groups differed significantly on each

of the 12 items in Table 3, but there was some variation (.45 to .99) in the capacity of the various items to distinguish between the

groups. The items which most successfully differentiated between

the ARA and MCL group members are items H, J and K, which are

among the "soft," discretionary or social psychological reasons for

abortion: having a low income and not being able to afford more

children, being a teenager whose parents would not consent, and

being single and not wanting to marry the father. Three other items

(B, A and G) were almost as capable of differentiating the two

groups: being married and not wanting more children, if there is a

strong chance of a serious defect in the baby, and when the woman wants an abortion but her husband would not consent. On

these six items, there was nearly complete agreement within

groups and nearly complete disagreement between groups. On two

of those items, there was a slight deviation from the group norm.

About five percent of MCL respondents approved if there was a

strong chance of a serious defect in the baby, and about five per

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

82

POPULATION AND ENVIRONMENT

Table 3

Percentage Approving of Legalized Abortion Being Available Under 12 Conditions

by Missouri Citizens for Life and Abortion Rights Alliance

"Please indicate whether or not you think it should be possible for a pregnant woman to obtain a legal abortion under the following conditions :

% saying Citizens

for Life

Yes

Abortion

Rights Alliance

A. If there is a strong chance of serious defect in the baby 5.2 100.0

B. If she is married and does not want any more children 0.4 98.6

C. If she became pregnant as a result of rape 18.3 100.0

D. If tests determine that the

baby would not be the sex desired by the parents 0.0 53.6

E. If the woman's own health is seriously endangered by the

pregnancy 20.9 100.0

F. If it is necessary to save the woman's life 69.3 100.0

G. If the woman wants to have an abortion but her husband would not consent to her having an abortion 0.9 94.7

H. If the family has a yery low income and cannot afford

any more children 0.4 99.3

I. If she became pregnant as a result of incest 19.0 100.0

J. If she is a teenager and her parents would not consent to her having an abortion 0.4 99.3

K. If she is not married and does not want to marry the man 0.4 99.3

L. If she wants to have an abortion for whatever reason 0.0 84.4

.95

.98

.85

.59

.82

.45

.93

.99

.84

.99

.99

.84

520

520

514

508

514

503

516

522

511

520

522

513

Note: In this table, the percentages are based on those answering either yes or no.

Those leaving the item blank or who gave an uncodable response were excluded from

these calculations. Therefore the % saying no would simply be 100 minus the figure

given in the columns giving the percentage of yes responses. The number of cases

on which the comparison is based is given in the right hand column and varies

slightly from item to item.

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

83

DONALD GRANBERG

cent of ARA members disapproved if the husband would not consent.

On the remaining six items, there was complete (or nearly

complete) agreement within one of the groups but some relatively

larger division within the other group. On four items (C,E,F and I), ARA members unanimously felt that abortion should be legally available. MCL members, however, were somewhat divided on

these same items. Eighteen percent of the MCL respondents felt

abortion should be legally available to a woman who became preg nant as a result of rape and 19 percent if pregnancy resulted from

incest. If the woman's health was seriously endangered by the

pregnancy, 21 percent of the MCL group felt a legal abortion

should be available and this percentage rose to a majority of 69

percent if the abortion were necessary to save the woman's life. It

is clear that most MCL members are not completely opposed to

legalized abortion under all conceivable circumstances. Of

course, with modern medical techniques, the situation where an

abortion is necessary to save the woman's life is (as many MCL

members pointed out) very rare. At the same time, it isn't strictly a

hypothetical situation either so long as some pregnant women

develop cancer of the uterus or have ectopic pregnancies.

When the 12 items in Table 3 were submitted to a factor

analysis with varimax rotation, only one factor had an eigen value

of more than 1.0, and it accounted for 79 percent of the variance.

When the six NORC items were factor analyzed, again only one

factor emerged and it explained about 91 percent of the variance.

This factor structure is noticeably different from when the same

procedure is applied to responses of a representative sample of the

U.S. adults. Among the latter, two factors have been consistently

reported, the first factor being comprised of the "soft" reasons and

the second of the "hard" reasons (Arney & Trescher, 1976;

Granberg, 1978). The activist data do not yield these two factors, but rather they merge into one general abortion attitude factor.

In addition to the 12 NORC types of items in Table 3 which

pertain to the legal availability of abortion, MCL and ARA

members were also asked to approve or disapprove of six abortion

related policy proposals. Table 4 presents these items and the

responses of MCL and ARA members. The two groups responded in a significantly different way to each item, but there was substan

tial variation in how well these attitude items differentiated the

two groups. The best discriminator was "a Constitutional Amend

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

84

POPULATION AND ENVIRONMENT

Table 4

Percentage Approving of Six Policy Alternatives Pertaining to Abortion

by Missouri Citizens for Life and Abortion Rights Alliance

% Approving

r N

Policy Alternative

Citizens

for

Life

Abortion

Rights Alliance

A. A Constitutional Amendment which would grant each state the right to legislate as it desires regarding abortion 34.9 4.5

B. A Constitutional Amendment which would give the fetus or unborn child the same

rights of personhood as are now held by people who have been born 96.5 0.3

C. Requiring public hospitals and other hospitals receiv

ing government funds to

provide abortions, as long as abortion is legal 3.5 96.9

D. A Constitutional Amendment which would forbid abortions

except when the physical health of the woman is

seriously endangered by the

pregnancy 56.9 0.3

E. A Constitutional Convention called to propose an amend ment to the Constitution

pertaining to abortion 80.9 7.4

F. Use of federal money to fund abortions for poor

people as long as abortion is legal 3.0 98.3

.39 506

.97 520

.93 520

.65 508

.75 498

.95 520

NOTE: In this table, the percentages are based on those answering approve or dis

approve. Those leaving the item blank or who gave an uncodable response were

excluded from these calculations. Therefore, the % saying disapprove would

simply be 100 minus the figure given in the columns giving the percentage of ap

proval. The total number of cases on which the comparison is based is given in

the right hand column and varies slightly from item to item._

ment which would give the fetus or unborn child the same rights of

personhood as are now held by people who have been born/'

Proposals to use federal money for abortions for poor people and

to require hospitals to provide abortions divided the two groups

nearly as well. About 81 percent of MCL but only seven percent of

ARA members approved of a Constitutional Convention called to

deal with abortion. Practically none of the ARA members

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

85

DONALD GRANBERG

approved of a Constitutional Amendment which would forbid abortion except when the physical health of the woman is seriously

endangered by the pregnancy, and this proposal won the approval of only 57 percent of MCL members, presumably because the

wording is not strong enough to suit them. The idea of amending the Constitution to allow the states to legislate as they want regard

ing abortion, in effect, a return to the status quo ante the Supreme Court Decisions of 1973, was favored by only 35 percent of MCL

and about four percent of the ARA members.

These six items were also factor analyzed and, once again, a

single factor emerged with an eigen value of more than 1.0, and it

explained over 70 percent of the variance. This suggests that the six

items in Table 4 can be combined into an abortion attitude scale.

This six-item abortion policy scale correlates .93 with both the six

and 12-item scales from Table 3.

On two of the items (D and L), there was complete agreement within the MCL group that abortion should not be available, but

there was some disgreement among ARA members. Eighty-four percent of ARA members felt that a legal abortion should be

available to a woman if she wants one for whatever reason, but

only 54 percent approved of abortion being available for women

when tests determine the baby would not be the sex desired by the

parents. Sixty-four percent of those who said no to the sex of the child item later said yes to the "for whatever reason" item. This

represents a logical fallacy, and it was recognized as such by some

of the ARA members who approved of the latter item (L) but not the former (D). Currently, abortion for reason of the prospective sex of the offspring is very rare in the U.S.

The items in Table 3 do not exhaust the conditions or circum

stances which are relevant in the abortion controversy. For in

stance, the stage of the pregnancy (e.g., which trimester) affects the judgments not only of philosophers and the U.S. Supreme

Court but also those made by a representative sample of U.S. adults (Blake, 1971). At the same time, the items in Table 3 success

fully discriminated between members of the two groups, and the items can therefore be combined into a valid scale measuring

people's attitudes toward legalized abortion. If the basic six NORC items are combined into an approval of legalized abortion scale, the point biserial correlation with group membership is .98. If all 12

items are combined into a scale, the correlation is .97. On the six

item scale, 72 percent of MCL and 98 percent of ARA members fall

at the extreme possible scores. On the 12-item scale, the corre

sponding percentages are 30 and 56.

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

86

POPULATION AND ENVIRONMENT

The possibility was considered that the absolutists within each

group (who say yes to all items or no to all items) might differ from the others in the group in terms of demographic or status

characteristics. Within the MCL group, women (34%) and Catho lics (35%) were more likely to be absolutists than men (23%) and

Protestants (20%). Also, those with less than a high school educa tion (33%) were more likely to be absolutists than college grad uates (22%). Within the ARA group, men (57%) and women (55%)

were about equally likely to be absolutists, but the Atheists (78%) and Agnostics (65%) were more likely to be absolutists than Jews

(51%) or Protestants (42%). There was not much variation among the various education and income groups in the percentage of absolutists.

In concluding this section, some very large differences between the two groups have been observed. If one views joining the organization and being active in it as social behavior, one

observes here a very high degree of attitude-behavior consistency, in contrast to the relatively low level of consistency often reported in the social psychological literature. The implication is that

attitude-behavior consistency increases as involvement or

commitment increases (Sherif & Sherif, 1969).

Beliefs

Beliefs are nonevaluative cognitions, that is, assertions, about

the nature of the physical, biological, and social world and its

environment, its past, present and future, and about relationships which exist between events and phenomena. As such, beliefs may be true or false, testable or untestable. Although beliefs, as such, are nonevaluative in nature, in reality they are often correlated

with attitudinal preferences and values.

The literature and pronouncements of pro- and anti-abortion

groups reveal that they make very different, though not necessarily contradictory, sets of assertions. Because the two groups focus on

different aspects of abortion, it was not known whether they would

reject the claims made by their opponents or merely regard such

claims as not dealing with the part of the controversy in which they were most interested. Therefore, it was decided to include in the

survey a set of belief type statements which could be judged as to

whether they were true or false (Sherif & Jackman, 1966). Many of

these belief statements were assertions made by one of the con

tending sides in the abortion controversy. Table 5 presents the comparisons on 18 items between MCL

and ARA members, including the mean values for each item (from

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

87

DONALD GRANBERG

? 2 for definitely false to +2 for definitely true), the N, and the correlation between group membership and response to the item.

As on the attitude items, there was a significant difference

between the groups on each of the 18 belief items, but there was

also substantial variation in the strength of this relationship, rang

ing from .93 for "Abortion is murder" to .32 for "In an ideal

society, there would not be any abortions or any need for them."

MCL were more likely than ARA members to believe that the

actual number of abortions in the U.S. had increased since 1973, while the ARA members were more likely to believe that abortion

has become safer since 1973. Not surprisingly, MCL members were

much more likely to believe that abortion is murder, that abortion

involves killing a living creature, that a person's life begins at

conception, that the fetus is a person, that the heart of the fetus is

formed and beating by the fourth week of pregnancy, and that

every abortion is a tragedy. MCL members were also more likely to

affirm that the number of abortions increases if the government

pays for them, that there have been over a million abortions per

year in recent years in the U.S., and that in an ideal society there

would not be any abortions.

ARA members were significantly more likely than those af

filiated with MCL to believe that the U.S. has a problem of popula tion increase now and will have this problem 20 years from now, that

the number of teenage mothers would increase if abortion were il

legal, and that fewer abortions would be performed if contra

ception were more widely available. ARA members were also more

likely to feel that women with an unwanted pregnancy have to

choose between the lesser of two evils. MCL members may reject that description because they feel that resolving an unwanted

pregnancy by giving birth and giving the baby up for adoption would not be "evil." Surprisingly, ARA members were more likely to believe that the anti-abortion forces are more organized and active than the pro-abortion forces. Being on the inside, MCL

members may be more aware of conflict and disorganization which may exist within the anti-abortion forces.

A special type of belief is one's expectation or prophecy about the future. Social psychological theory and research lead to

the hypothesis that such a belief is influenced by one's pre

ferences, so that the people tend to expect preferred outcomes

(Marks, 1951; Rothbart, 1970; Matlin & Stang, 1978). This hypo thesis was supported, as MCL were significantly more likely than

ARA members to believe that in five years there will be more

restrictions on abortion than at present.

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

88

POPULATION AND ENVIRONMENT

A second special type of belief is one's attribution of an atti

tude to another person or regarding the distribution of attitudes in a group. If a Senator claims, on whatever basis, that the majority of constituents favor a constitutional amendment to outlaw

abortions, this is or reflects an attributional type of belief. In the

present study, in addition to being asked their own attitudes on the 12 items in Table 3, people were also asked to estimate the per

centage of a current representative sample of U.S. adults who

would say yes and no to each item. It was hypothesized that ARA

respondents would attribute more pro-abortion attitudes to U.S. adults than would those in the MCL group. This effect has been referred to as the pro-us illusion, the assumption of similarity to

self, an assimilation effect, or most recently as the false consensus

phenomenon (Travers, 1941; Wallen, 1943; Sherif & Hovland, 1961; White, 1970; Granberg, 1972; Granberg & jenks, 1977; Ross, Green & House, 1977). The relevant evidence is presented in Table 6.

As hypothesized, on each of the 12 items, the estimate of pro abortion sentiment among U.S. adults is significantly higher for the

ARA than for the MCL members. The strength of the relationship between group membership and attribution varied from .45 for "if

she is married and does not want more children" to .11 for "if it is

necessary to save the woman's life." The comparison in Table 6 is

between groups with no control for the person's own attitude.

Recall that there was less than complete consensus within groups on several of the attitude items. The statistical tests were also

done in three other ways: grouping people on the basis of their

attitudes irrespective of group membership, comparing only MCL

members who said no with ARA members who said yes, and

comparing the estimates of people who said yes or no within each

group. The results from such comparisons are consistent with

those of Table 6 and are not reported here to avoid redundancy. Was one group more accurate than the other? An estimate of

how a current sample of adults really would respond to these items

comes from the NORC General Social Survey done in March of

1980. These results are given in Table 6. The average estimate by ARA members was more accurate in four cases, MCL members

were more accurate in three cases, and the remaining five items

were not included in the NORC survey. So neither group had a

noticeably more accurate impression of how people in the U.S.

feel about legalized abortion.

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

89

DONALD GRANBERG

DISCUSSION

In this article, very large differences have been reported be tween respondents affiliated with MCL, a "pro-life" group, and

those affiliated with a "pro-choice" group as to their values,

attitudes, and beliefs. The value differences are noteworthy because the concept of value is one of the major tools social scientists have to analyze social movements. Sociologists have often distinguished between revolution and reform oriented move

ments or between value and norm oriented movements. The

question is whether the movement is trying to radically alter the structure of society or merely trying to modify some existing practice so as to conform more closely with some enduring value.

Subsequent analyses of these data should attempt to compare the

results of these two movement organizations to surveys of repre sentative samples of U.S. adults. The objectives of such analysis

might be to make judgments whether the value pattern of either

group departs radically from the mainstream of U.S. society and

whether one group is more similar to the "silent majority" of U.S. adults than the other.

Although there were many significant differences on values

between MCL and ARA members, by far the value that dif

ferentiated the groups most successfully was salvation. In view of

this, it would be hard to deny realistically that there is a substantial

religious factor in the abortion controversy. If one accepts the

premises of the religiously oriented "pro-life" person, that there is

a soul and that the soul enters the body at the time of conception, it is certainly understandable that they would be concerned about

the eternal fate of an embryo or a fetus.

The section on attitudes accomplished two things. First, it

offers evidence for the construct validity of attitude items which

have been and will undoubtedly continue to be used extensively in

social scientific analyses. Second, it demonstrated the lack of

complete consensus within these activist and relatively extreme

groups. There are absolutists within each group, but there are also

many who make differentiations.

A common procedure for activists on both sides has been to

make a set of belief type assertions, accompanied by appropriate documentation or illustration. The presumption seems to have

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

90

POPULATION AND ENVIRONMENT

ITEM

Table 5

Judgments of Truth and Falsity of 18 Belief Items

by Missouri Citizens for Life and Abortion Rights Alliance

Average Judgment by:

Citizens Abortion

for Rights Life Alliance r

B.

C.

D.

Since the Supreme Court Decisions of 1973, the actual number of abortions in the U.S. (legal & il

legal) has increased +1.9

For the women who obtain

them, abortions have become medically safer since the Supreme Court Decisions of 1973 +0.1

Abortion is murder +1.8

Abortion involves the

killing of a living creature +2.0

The unborn child or fetus is a person +1.9

If the federal government pays the cost of abortions, the number of abortions increases +1.7

Women with an unwanted

pregnancy have to choose between the lesser of two evils -0.9

Currently, the U.S. has a

very serious problem of

population increase -1.4

Twenty years from now, the U.S. will have a

very serious problem of

population increase -1.1

Currently, the anti-abortion forces are more organized and active than the pro abortion forces +0.2

K. In an ideal or Utopian society, there would not be any abortions or

any need for them +1.2

(Continued on next page)

G.

I.

J.

+1.0

+1.9

-1.8

-0.7

-1.7

+0.6

+0.9

+0.1

+0.4

+1.4

+0.3

.47

.69

.93

.76

.96

.51

.53

.54

.56

.49

.32

465

520

513

519

518

521

497

519

519

518

511

been that attitudes are intimately related to beliefs in this area and

that, if one can change the latter, the former may change. The data

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

91

DONALD GRANBERG

ITEM

Table 5 (continued)

Judgments of Truth and Falsity of 18 Belief Items

by Missouri Citizens for Life and Abortion Rights Alliance

Average Judgment by:

Citizens Abortion

for Rights Life Alliance r

L.

M.

P.

Q.

Five years from now, there will be more restrictions on abortion in the U.S. than is presently the case +0.5

The number of teenage mothers would increase if abortion were illegal +0.2

If contraceptive information and equipment were more

widely available, fewer abortions would be per formed -0.3

A person's life begins at the moment of conception +1.9

The heart of the fetus is formed and beating by the fourth week of pregnancy

In the past few years, there have been over a million abortions in the U.S. per year

Every abortion is a tragedy

+1.5

+1.5

+1.7

-0.3

+1.4

+1.2

-1.3

+0.2

+0.6

-1.3

.38

.49

.55

.89

.53

.48

.81

520

517

518

521

488

498

506

Note: The range of possible responses to these items was -2 definitely false,

-1 probably false, 0 don't know, +1 probably true, +2 definitely true.

in Table 5 are consistent with, but not definitive proof for, this way

of thinking. On many of the belief items, the within group consensus and the between group dissensus were both strong. The

belief items in Table 5 represent very different ways of thinking about the world. The abortion phenomenon is defined very

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

92

POPULATION AND ENVIRONMENT

Table 6

Percentage of U.S. Adults Approving of Legalized Abortion

(Based on 1980 Survey by the National Opinion Research Center,

University of Chicago) and as Estimated by Missouri

Citizens for Life and Abortion Rights Alliance

Actual % % of U.S. Adults Saying

U.S. Adults Yes as Estimated by:

saying Yes Citizens Abortion

(NORC, 1980) for Rights Life Alliance r

A. If there is a strong chance of serious defect in the

baby 83.1 59.5

B. If she is married and does not want any more children 47.1 35.4

C. If she became pregnant as a result of rape 83.4 70.3

D. If tests determine that the

baby would not be the sex desired by the parents NA 20.0

E. If the woman's own health is seriously endangered by the pregnancy 90.1 68.3

F. If it is necessary to save the woman's life NA 84.5

G. If the woman wants to have an abortion but her husband would not consent to her

having an abortion NA 39.0

H. If the family has a very low income and cannot

afford any more children 51.7 46.9

I. If she became pregnant as a result of incest NA 72.5

J. If she is a teenager and her parents would not

consent to her having an abortion NA 41.7

K. If she is not married and does not want to marry

the man 48.4 44.1

L. Tf she wants to have an abortion for whatever reason 41.1 35.3

75.5 .45 474

54.3 .47 469

83.6 .40 474

26.3 .17 468

79.8 .34 474

87.7 .11 473

51.6 .32 470

60.0

83.0

53.9

58.1

50.2

.34 472

.29 473

.30 472

.34 469

.34 471

Note: NA means the data are not available since this item was not included in the

NORC survey of 1980.

differently by the two groups. However, even though it often seems that the two sides are talking about different aspects of the

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

93

DONALD GRANBERG

same phenomenon, which are not always necessarily contra

dictory, the judgments of truth and falsity in Table 5 suggests that

group members tend to actively reject the assertions made by the

other side when they make a judgment.

Previously, it was not known whether the false consensus

phenomenon in the attribution of attitudes had sufficient

generality to extend to active members of issue oriented social

movement organizations. The evidence in Table 6 strongly suggest that it does extend to such group members. The question now is

whether this is merely an interesting cognitive bias or a matter of some consequence for the experiences people feel and how they act. If one feels that policy makers are not acting in a way con

sistent with one's preference, and one senses that most other non

elites share the preference, one might have a different experience and behave differently than if one felt that policy makers are not

acting in line with one's preference and that most other non-elites

do not share the preference. As to why the false consensus phenomenon occurs, two possi

bilities can be mentioned briefly. First, it may be an outcome of the

attempt to maintain psychological consistency. If one has a

positive attitude toward the U.S. and its citizens, it would be psy

chologically distressing to sense that they tended to disagree with

oneself on a matter of vital concern. Second, it may be a matter of

psychological and sociological selectivity. People are exposed to a

selective, nonrandom sample of facts, ideas, and other people and, when making attributions, they probably do not take adequately

into account the biased nature of this information and these

interaction patterns.

REFERENCES

Arney, W. & Trescher, W. Trends in attitudes toward abortion, 1972-1975. Family Planning

Perspectives, 1976, 8 (No. 3), 117-124.

Blake, J. Abortion and Public opinion: The 1960-1970 decade. Science, 1971, 777, 540-549.

Crano, W. & Brewer, M. Principles of research in social psychology. New York: McGraw

Hill, 1973.

Granberg, D. Authoritarianism and the assumption of similarity to self. Journal of Experi mental Research in Personality, 1972, 6, 1-4.

Granberg, D. Pro-life or reflection of conservative ideology? An analysis of opposition to

legalized abortion. Sociology and Social Research, 1978, 62, 414-429.

Granberg, D. Missouri Citizens for Life and the Abortion Rights Alliance: Initial Report

of Comparisons of Two Opposing Social Movement Organizations. Unpublished manu

script, 1979.

Granberg, D. & Granberg, B. Abortion attitudes, 1965-1980: Trends and determinants.

Family Planning Perspectives, 1980, 72(5), 250-261.

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions

94

POPULATION AND ENVIRONMENT

Granberg, D. & Jenks, R. Assimilation and contrast effects in the 1972 election. Human

Relations, 1977, 30, 623-640.

Marks, R. The effect of probability, desirability, and "privilege" on the stated expectations of children. Journal of Personality, 1951, 79, 332-351.

Matlin, M & Stang, D. The pollyanna principle: Selectivity in language, memory, and thought

Cambridge, Massachusetts: Schenkman, 1978.

Mileti, D. & Barnett, L. Nine demographic factors and their relationship to attitudes toward

abortion legalization. Social Biology, 1972, 79, 43-50.

Rokeach, M. The nature of human values. New York: Free Press, 1973.

Ross, L., Greene, D., & House, P. The "false consensus effect": An egocentric bias in social

perception and attribution process. Journal of Experimental Social Psychology, 1977,

13, 279-301.

Rossi, A. Abortion laws and their victims. Transaction, 1966, 3, (No. 6), 7-12.

Rothbart, M. Assessing the likelihood of a threatening event: English Canadians' evalua

tions of the Quebec Separatist Movement. Journal of Personality and Social Psy

chology, 1970, 15, 109-117.

Sherif, C. & Jackman, N. Judgments of truth by participants in collective controversy. Public

Opinion Quarterly, 1966, 30, 173-186.

Sherif, M. & Hovland, C. Social judgment: Assimilation and contrast effects in communi

cation and attitude change. New Haven: Yale University Press, 1961.

Sherif, M. & Sherif, C. Social psychology. New York: Harper & Row, 1969.

Singleton, R & Christiansen, J. The Construct Validation of a Shortform Attitudes Toward

Feminism Scale. Sociology and Social Research, 1977, 67(3), 294-303.

Travers, R. A study in judging the opinions of groups. Archives of Psychology, 1941, No. 266, 1-73.

Wallen, R. Individuals' estimates of group opinion. Journal of Social Psychology, 1943, 77, 269-274.

White, R. Nobody wanted war: Misperception in Vietnam and other wars. Garden City, New

York: Anchor Books, Doubleday, 1970.

This content downloaded from 35.50.40.65 on Tue, 03 Nov 2015 02:29:28 UTC All use subject to JSTOR Terms and Conditions