Large Marriage Gap Evident on Moral Issues

by Jeffrey M. Jones, Gallup Poll Managing Editor

In the last several presidential elections, some political observers have pointed to a "marriage gap" in voting intentions. Specifically, married Americans tend to vote Republican, while unmarried Americans tend to choose Democrats. But the marriage gap extends beyond basic voting intentions -- an aggregate of Gallup's 2001-2003 Moral Values polls shows considerable divergence between married and unmarried Americans on a variety of moral issues*.

Each May, Gallup asks Americans whether a series of issues are morally acceptable or unacceptable. For the most part, opinion has been stable from year to year (one exception is attitudes toward homosexuality have varied somewhat in the past few years).

On most of the issues that touch on marriage in some way, married Americans tend to hold much more conservative moral views than do unmarried Americans (living together, divorced, or never married), with the possible exception of their views on extramarital affairs.

Percentage Saying Act Is Morally Acceptable, by Marital Status

Issue

 

Married

(N=1,608)

Living together

(N=176)

 

Divorced

(N=387)

Never married

(N=496)

Sex between unmarried man and woman

49%

84%

59%

68%

Married men and women having an affair

6%

12%

8%

8%

Divorce

58%

73%

73%

68%

Having a baby outside marriage

43%

73%

52%

56%

The gaps in opinion are especially large between those who are married and those who are unmarried but are living with someone. For example, 49% of married Americans think sex between an unmarried man and woman is morally acceptable, compared with 84% of those living with a partner who say the same. Forty-three percent of married Americans think having a baby outside of marriage is acceptable, compared with 73% of those living together. Divorced Americans and those who have never married tend to be significantly more liberal on marriage issues than married Americans are, but not to the same degree as those who are living with a partner.

But attitudes toward moral issues also extend beyond questions of marriage to a wide variety of moral issues such as abortion, stem cell research, and medical testing on animals. In these cases, however, the differences are not nearly as large as those found on issues dealing with marriage.

Percentage Saying Act Is Morally Acceptable, by Marital Status

Issue

 

Married

(N=1,608)

Living together

(N=176)

 

Divorced

(N=387)

Never married

(N=496)

Abortion

36%

52%

38%

51%

Homosexual behavior

38%

58%

39%

52%

Death penalty

68%

62%

65%

60%

Doctor-assisted suicide

48%

58%

51%

52%

Medical testing on animals

67%

61%

62%

55%

Buying/Wearing animal fur

64%

51%

62%

54%

Cloning animals

30%

34%

28%

37%

Cloning humans

6%

10%

6%

13%

Suicide

12%

17%

15%

14%

Stem cell research

54%

63%

48%

59%

Specifically, the gap in perceived morality on marital issues between married Americans and those living with a partner averages 22 percentage points. The gaps between married Americans and divorced Americans and those who never married average 9 percentage points and 11 percentage points, respectively. On non-marital issues, the average gap between the views of those living with a partner and those who are married is nine points. Those who were never married show a slightly smaller gap of seven points, and divorced Americans show little difference with married Americans on these moral issues.

Gap in Perceived Moral Acceptability of Acts, Married vs. Unmarried Americans

Issue

 

Married

(N=1,608)

 

Living
together

(N=176)

 

Divorced

(N=387)

Never married

(N=496)

Sex between unmarried man and woman

49%

+35

+10

+19

Married men and women having an affair

6%

+6

+2

+2

Divorce

58%

+15

+15

+10

Having a baby outside marriage

43%

+30

+9

+13

Average gap on marital issues

+22

+9

+11

Abortion

36%

+16

+2

+15

Homosexual behavior

38%

+20

+1

+14

Death penalty

68%

-6

-3

-8

Doctor-assisted suicide

48%

+10

+3

+4

Medical testing on animals

67%

-6

-5

-12

Buying/Wearing animal fur

64%

-13

-2

-10

Cloning animals

30%

+4

-2

+7

Cloning humans

6%

+4

0

+7

Suicide

12%

+5

+3

+2

Stem cell research

54%

+9

-6

+5

Average gap (absolute value) on non-marital issues

9.3

2.7

7.4

Influence of Marriage on Moral Values

Do these data suggest that being married tends to make people's moral outlooks more conservative? Indeed, when they are asked to indicate their political ideology, half of married Americans say they are conservative, compared with 29% of those who are living together with a partner, 39% who are divorced, and 29% who have never married. Among those who are living with a partner or have never married, liberals outnumber conservatives. In the general public, there are generally twice as many conservatives as liberals, and among married Americans, there are three times as many conservatives as liberals.

Marital status is related to ideology as well -- only 4% of conservatives are living together with a partner, compared with 7% of moderates and 11% of liberals who are.

However, in many instances the presence of a relationship does not necessarily imply a causal connection. Other factors may have more to do with the relationship between marriage and conservative moral outlooks. Previous Gallup analyses have shown that moral values are strongly related to religiosity, political ideology, and in some cases, age; factors that also tend to be related to marital status and/or views on marriage.

To get a sense of how strongly being married influences someone's views on moral issues, one can analyze the relationship taking into account the effect of characteristics such as age, ideology, and religiosity. A statistical technique known as logistic regression can be used to attempt to predict whether a person will find an issue morally acceptable or unacceptable based on a variety of factors. In this example, the predictive factors include gender, age, education, race, marital status, religiosity (as measured by church attendance), political partisanship, and political ideology. The results show whether being married independently influences one's views on moral issues, as well as the relative importance of the various factors in determining how questions of morality are perceived.

The following table summarizes the results**. For each issue, the table indicates whether being married (versus not being married, combining divorced, never married, and living together into one category) independently affects finding the issue morally acceptable, taking into account the effects that other characteristics have. Secondly, the table shows the relative importance (by rank order) of the various characteristics on one's views.

Predictors of Attitudes on Moral Issues

 

Issue

Marriage independently related to views on issue?

 

Most important predictors of view on issue

Marital issues

Sex between unmarried man and woman

Yes

Religiosity, Age, Ideology, Education

Married men and women having an affair

Yes

Religiosity, Race, Age, Gender

Divorce

Yes

Religiosity, Ideology, Age, Education

Having a baby outside marriage

No

Religiosity, Ideology, Age, Gender

Non-marital issues

Abortion

No

Religiosity, Ideology, Education, Partisanship

Homosexual behavior

No

Religiosity, Education, Ideology, Age

Death penalty

Yes

Race, Partisanship, Ideology, Age

Doctor-assisted suicide

No

Religiosity, Education, Race, Ideology

Medical testing on animals

Yes

Gender, Education, Partisanship, Married

Buying/Wearing animal fur

No

Gender, Education, Partisanship, Religiosity

Cloning animals

No

Gender, Education, Religiosity, Race

Cloning humans

No

Gender, Education, Religiosity, Age

Suicide

No

Religiosity, Education, Age, Ideology

Stem cell research

No

Religiosity, Education, Ideology, Gender

Taken together, the results suggest that being married has an influence on how one views marriage-related issues, including extramarital affairs, sex between people who are not married, and divorce (but not having a baby outside of marriage). However, while marriage relates to one's views on these matters, this relationship is nowhere near as strong as that of one's religious commitment, age, or political ideology.

Being married typically is not related to one's views on other moral issues, with just a couple of exceptions (the death penalty and medical testing on animals). But on these issues as well as others, religiosity, ideology, and education are the most important predictors of one's views.

Bottom Line

While married people view moral issues -- especially those that relate to marriage -- quite differently from those who are not married, these differences are due largely to people's religious commitment, political ideology, age, and education. The marriage gaps observed in these data result from a small, independent influence of one's marital status on how he or she views the issues, but result even more from the probable influence that factors such as religiosity and ideology have on people's attitudes toward marriage. Surely, one's religiosity and ideology also affect one's personal decision whether to get married at all, as well as willingness to stay in a troubled marriage.

This interplay of religion, ideology, and views toward marriage could help explain the counterintuitive finding that couples who live together before marriage have higher divorce rates than those who do not cohabitate. That is, the mere act of living together may not account for a greater likelihood of divorce. Rather, it could stem from the observation that people who are unmarried and live together are more likely to be more liberal politically and less religious and also to hold more liberal views on marriage and divorce. Those who are more religious and more conservative -- whose views on marriage are more conservative and probably more grounded in religion -- are far less likely to live together prior to getting married.

*Data are based on a combination of three Gallup Polls, each of approximately 1,000 adults aged 18 and older. Interviewing dates were May 10-14, 2001, May 6-9, 2002, and May 5-7, 2003.

For the total sample of 3,029 national adults, the maximum margin of error attributable to sampling is ±2 percentage points. For the sample of 1,608 married Americans, the maximum margin attributable to sampling error is ±3 percentage points. For the sample of 176 unmarried Americans who are living together with a partner, the maximum margin attributable to sampling error is ±8 percentage points. For the sample of 387 divorced Americans, the maximum margin attributable to sampling error is ±5 percentage points. For the sample of 496 Americans who have never married, the maximum margin attributable to sampling error is ±5 percentage points.

**Analyses are done using logistic regression analysis, with each of the morality issues as a dependent variable (coded 0 as not acceptable, 1 as acceptable), and the demographic characteristics described above as the independent variable. The table indicates whether the married variable (married or unmarried) is statistically significant at the (p < .05) level. The table also lists in rank order the relative impact of the predictor variables (as determined by the average change in the probability of finding an issue morally acceptable across the range of the variable's values). Please note that in many cases additional variables are also statistically significant predictors of one's view on the moral issue being analyzed.

Get Articles in Related Topics:


Gallup http://www.gallup.com/poll/9124/large-marriage-gap-evident-moral-issues.aspx
Gallup World Headquarters, 901 F Street, Washington, D.C., 20001, U.S.A
+1 202.715.3030