GraphCheating on your taxes is almost as bad as cheating on your spouse.

Drinking excessively is worse than smoking marijuana.

Engaging in homosexual behavior and having an abortion are equally fraught.

Telling a lie to spare someone’s feelings is worse than gambling.

Sex between unmarried adults is more objectionable than overeating (but not by much).

Those edicts aren’t drawn from some new millennium user’s guide to morality. Rather, they represent the collective judgment of the American public when asked in a Pew Research Center survey to assess the moral dimensions of different kinds of behaviors.

Survey respondents were read a list of ten behaviors and asked whether, in their personal opinion, each one is “morally acceptable, morally wrong, or not a moral issue.”

The survey, by design, covered a wide range of activities, in part to avoid signaling to respondents that inclusion on the list was meant to convey a presumption of moral unacceptability.

The activity that drew the most widespread moral disapproval, 88%, was “married people having an affair,” while the one that drew the least was “overeating” – although a sizable minority (32%) said that activity was morally wrong.

The survey did not measure intensity of feelings. It’s possible, therefore, that the difference between the 79% who say it’s morally wrong to cheat on one’s taxes and the 88% who say the same about cheating on one’s spouse is greater (or smaller) than those numbers indicate. Judgments about right-and-wrong are by nature profound, and – in real life – often nuanced and situational. By contrast, this survey questionnaire is a blunt instrument.

Even so – and in admittedly coarse strokes – the alignment of responses to the ten questions paints an interesting portrait of contemporary American morality.

About that 1040

As April 15 approaches and tens of millions of Americans prepare their tax returns, they may be interested to know that eight-in-ten of their fellow citizens (79%) consider not reporting all income on one’s taxes to be morally wrong, while just 5% consider it morally acceptable and 14% say it’s not a moral issue.

Moral disapproval is one thing, behavior another. Earlier this year the IRS reported that in 2001 (the last year for which it had conducted such research) there was a gross “tax gap” of $345 billion, resulting from an overall non-compliance rate of about 16 percent. Of that gap, the biggest missing slice, some $197 billion, was from underreporting of income on individual income tax returns; most of that missing sum, in turn, resulted from underreporting of business income on those individual returns, the IRS found.

About a Different Kind of Cheating

The only behavior on the Pew list that draws more moral condemnation than cheating on one’s taxes is cheating on a spouse. Some 88% say it is morally wrong for married people to have an affair, while 3% say it is morally acceptable and 7% say it is not a moral issue.

Here again, condemnation is one thing, behavior another. The General Social Survey (which is funded by the National Science Foundation) has been asking about adulterous behavior in numerous surveys since 1991.In 2004, it finds that 15% percent of those ever married say that they have had sex outside of their marriage, and that more (currently or formerly married) men (20%) than women (12%) report this behavior. (Needless to say, on a topic as sensitive as adultery, it is possible that people are not always honest when asked in a survey questionnaire about their personal behavior).

In the Pew survey, there are also some gender differences in moral judgments about adultery. Some 90% of women, compared with 85% of men, say adultery is morally wrong. Men (3%) are no more inclined than women (2%) to say it is morally acceptable, but 10 percent of men say extra-marital sex is “not a moral issue,” compared with just 5% of women who feel that way.

Men and Women, Young and Old Differ On Abortion and Homsexuality

Two moral issues that have had the greatest political resonance in recent years – homosexuality and abortion – divide the broad public in almost exactly the same way, but are seen differently by some sub-groups in the population.

Men are more morally disapproving than women of homosexuality, but both genders have similar views about abortion. Likewise, the old and the young judge the morality of these two behaviors in different ways. On the question of homosexuality, the old are more disapproving than the young. But on the question of abortion, there is no clear difference between the old and the young.

Catholics are more disapproving of abortion than they are of homosexuality. Married people are more disapproving of abortion than are those not currently married, but there is no clear difference between the married and unmarried on homosexuality.

Despite these sub-group differences, the two behaviors wind up being judged in nearly identical ways by the full population. About half of those surveyed say abortion (52%) and homosexual behavior (50%) are morally wrong, while an identical 12% say that each of these activities is morally acceptable. Another one in three (33%) say homosexuality is “not a moral issue.” Some 23% also say that about abortion, with an additional 11% volunteering an answer to the effect that “it depends on the situation.” (Of all ten behaviors tested, abortion drew the most volunteered responses of that nature.)

Differences by Age, Income, Religiosity, Ideology and Party Line Up with Differing Responses

These are the some of the traits associated with responses to the battery of ten questions:

  • The groups with a majority saying that eight or more of the behaviors are morally wrong include: conservatives, frequent church-goers; white evangelical Christians, and those ages 65 and older.
  • Majorities of three groups – weekly church-goers, white evangelical Christians and those ages 65 and older – say that nine of the 10 behaviors are morally wrong. Among these groups, overeating is the only behavior not judged by a majority to be morally wrong.
  • Groups with a majority saying that no more than two of these behaviors are morally wrong include college graduates, people with family incomes of at least $75,000 a year, and people who seldom or never attend religious services. For all those groups, adultery and income tax cheating are the only two behaviors that a majority judge to be morally wrong.
  • There is a partisan divide in how people judge these behaviors. A majority of Republicans say seven of the ten behaviors are morally wrong; while a majority of Democrats and independents say just three of the behaviors (adultery, underreporting taxable income; drinking excessively) are morally wrong. Independents are the least inclined of the three partisan groups to view the behaviors as morally wrong and most prone to see them as “not a moral issue.”
  • Some demographic factors – including gender and marital status – are not strongly correlated with views on these questions. However, there is regional pattern in the responses, with Southerners more likely than people living in the Northeast or West, and slightly more likely than those living in the Midwest, to describe the 10 behaviors as morally wrong. There are too few respondents in the survey for a substantive analysis by race and ethnicity.
  • People who decline to call a behavior morally wrong don’t necessarily believe it is morally acceptable. In fact, for all ten items on the list, those who say a behavior is “not a moral issue” outnumber those who say it is morally acceptable, generally by a sizable margin. A majority of all adults (58%) believe that overeating is “not a moral issue,” while pluralities say this about gambling (42%) and sex between unmarried adults (37%).

About the Survey

Results for this survey are based on telephone interviews conducted with a nationally representative sample of adults, ages 18 years and older, living in continental U.S. telephone households.

  • Interviews conducted February 1- 5, 2006 • 745 interviews for the half sample (1,502 interviews for the full sample)
  • Margin of sampling error is plus or minus 4 percentage points for results based on the total sample at the 95% confidence level. The margin of sampling error is higher for results based on subgroups of respondents.

Survey interviews conducted under the direction of Princeton Survey Research Associates International.

In addition to sampling error, bear in mind that question wording and practical difficulties in conducting surveys can introduce error or bias in the findings of opinion polls.

Read the full report for more details.