Biblical literalism correlates with thinking extramarital sex is wrong

The Audacious Epigone has a post up where the title says it all, Extramarital sex wrong? Gays and supporters of same sex marriage less likely to think so. But I was curious how MARHOMO, attitudes toward gay marriage, stacked up against other independent variables in relation to XMARSEX, attitudes toward extramarital sex. Here is what XMARSEX is representing:

What is your opinion about a married person having sexual relations with someone other than the marriage partner?

1 – ALWAYS WRONG
2 – ALMOST ALWAYS WRONG
3 – SOMETIMES WRONG
4 – NOT WRONG AT ALL

Here’s the logit regression from the GSS:

Like my previous posts you can see attitudes about the Bible have a huge effect. Let’s look at how BIBLE and XMARSEX relate specifically:

Cells contain: -Column percent -Weighted N		BIBLE
Cells contain: -Column percent -Weighted N		1 WORD OF GOD	2 INSPIRED WORD	3 BOOK OF FABLES	4 OTHER	ROW TOTAL
XMARSEX	1: ALWAYS WRONG	90.1 4,689	80.3 6,096	59.8 1,447	71.4 118	80.3 12,350
	2: ALMOST ALWAYS WRONG	5.2 272	13.3 1,009	23.0 556	17.4 29	12.1 1,865
	3: SOMETIMES WRONG	3.2 169	5.2 396	12.8 309	8.3 14	5.8 888
	4: NOT WRONG AT ALL	1.4 73	1.2 94	4.5 109	2.8 5	1.8 280
	COL TOTAL	100.0 5,202	100.0 7,595	100.0 2,421	100.0 166	100.0 15,383

Gabriel says:

April 8, 2009 at 8:45 pm

your logit regression includes a lot of variables that are coded as multi-value categorical (eg REGION) or ordinal (eg DEGREE). the categorical one is a big no-no and the ordinal one is debatable. to do it right the categorical ones need to be broken into dummy sets or at least collapsed into a dummy (eg REGION could be “south” vs. else). likewise the ordinal variables should be broken into dummy sets or recoded as continuous (eg DEGREE could be approximate years of education).
David Hume says:

April 12, 2009 at 3:04 pm

thanks. i’ll try not to be so sloppy in the future!

Comments are closed.