Saturday, June 24, 2006

Poor People Believe In God, Rich People Don't?

An inverse relationship exists between levels of education and income and belief in God. Generally speaking, Americans who have attained higher levels of education and who have higher household incomes are less likely to be certain that God exists than those who are on the other end of the socioeconomic spectrum:

Only about three-quarters of Americans with post-graduate educations or with incomes above $150,000 a year are certain that God exists or believe that God probably exists and have only a little doubt. That compares with more than 9 out of 10 of those with high school educations or less, and about 9 out of 10 of those making less than $50,000 a year in family income.

Some of this may reflect the generally more skeptical nature that often comes with more formal education. Those who are college graduates may have been socialized to be more cautious about agreeing with any statement of fact. There are also many theories about the functions of religion that may come into play here; those who have fewer material possessions or who are more oppressed in this life may be more likely to depend on the belief in a God and the potential for a blissful life after death as a coping mechanism for dealing with their current life circumstances.

Women are more religious than men on many measures, including self-reported importance of religion in one's daily life and active church attendance and church participation. It is not surprising, therefore, that we find a modest gender difference in terms of belief in God. Ninety-one percent of women are in the "basic believers" category, compared to 85% of men.

These results are based on telephone interviews with randomly selected national samples of approximately 1,000 adults, aged 18 and older, conducted at three points in time: Oct. 24-26, 2005, Nov. 17-20, 2005, and May 8-11, 2006. For results based on any one of these samples, one can say with 95% confidence that the maximum error attributable to sampling and other random effects is ±3 percentage points. For analyses involving the aggregate of all three polls, the maximum error attributable to sampling and other random effects is ±2 percentage points. In addition to sampling error, question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of public opinion polls.