Opinion data about abortion have been collected for more than thirty years. Isolating results to one or two questions or one or two surveys, in this instance, can cause one to completely miss the complexity of this controversial issue. In this instance the reliability checking should be across survey houses and question wording.

Here is a dichotomous question:

With respect to the abortion issue, would you consider yourself to be pro-choice or pro-life?

51%
Pro-choice
42%
Pro-life
3%
Mixed/Neither (Vol.)
3%
Don’t know what terms mean (Vol.)
1%
No opinion

Survey by the Gallup Organization for CNN and USA Today. July 22-July 24, 2005 and based on telephone interviews with a national adult sample of 1,006.

When there are three options for respondents, public opinion tends towards a middle ground on this issue as demonstrated in these three different questions:

Do you think abortion should be legal under any circumstances, legal only under certain circumstances, or illegal in all circumstances?

23%
Legal under any circumstances
53%
Legal only under certain circumstances
22%
Illegal in all circumstances
2%
No opinion

Survey by the Gallup Organization, May 2- May 5, 2005 and based on telephone interviews with a national adult sample of 1,005.

 

Which of the following best represents your views about abortion—the choice on abortion should be left up to the woman and her doctor, abortion should be legal only in cases in which pregnancy results from rape or incest or when the life of the woman is at risk, or abortion should be illegal in all circumstances?

55%
Should be left up to the woman and her doctor
29%
Legal only in cases of rape/incest/life of the woman is at risk
14%
Illegal in all circumstances
2%
Not sure

Survey by NBC News, Wall Street Journal. May 12- May 16, 2005 and based on telephone interviews with a national adult sample of 1,005.

Which of these comes closest to your view—abortion should be generally available to those who want it, or abortion should be available but under stricter limits than it is now, or abortion should not be permitted?

35%
Generally available
37%
Available but with limits
25%
Should not be permitted
3%
Don’t know/No answer

Survey by CBS News, March 21- March 22, 2005 and based on telephone interviews with a national adult sample of 737.

Now, observe what happens when the question offers more options, attempting to measure intensity of opinion:

Do you think abortion should be legal in all cases, legal in most cases, illegal in most cases or illegal in all cases?

20%
Legal all cases
36%
Legal most cases
27%
Illegal most cases
14%
Illegal all cases
3%
No opinion

Survey by ABC News/Washington Post, April 21- April 24, 2005 and based on telephone interviews with a national adult sample of 1,082.

And with an introduction of suggested circumstances:

Which comes closest to your view on abortion? Abortion should always be legal. Should be legal most of the time. Should be made illegal except in cases of rape, incest and to save the mother’s life. Abortion should be made illegal without any exceptions.

24%
Always legal
19%
Most of the time legal
41%
Illegal with exceptions
12%
Always illegal
4%
Don’t know

Survey by The Los Angeles Times, January 15- January 17, 2005 and based on telephone interviews with a national adult sample of 1,033.

In the example of abortion, the data, like the issue, are extremely complicated. In addition to these items, there are series of questions about which circumstances one might consider permitting abortion, including timing, financial and health considerations, methods, etc. These are strong arguments for taking the time to become acquainted with the whole picture—multiple sources and different questions and time periods—before preparing meaningful secondary analyses.

 

Example 1 – Outlier Effect