Sociology Now, Census Update

should set a deadline for withdrawing U.S. military forces from Iraq in order to avoid

further casualties. Others say knowing when the United States would pull out would

only encourage the anti-government insurgents. Do you yourself think the United

States should or should not set a deadline for withdrawing U.S. forces from Iraq?”

In other words, the Postasked, should we get out just to save American

lives, even if that would be a victory for the terrorists? What was surprising is that

47 percent of the respondents still said that some timetable is better than no

timetable.

How about the placement of the question in the survey? Respondents are much

more likely to respond honestly to the shoplifting question if it’s near the end of the

survey. When sensitive or embarrassing questions come early, respondents are put off,

wondering how intimate the questions are going to get. After they get a little prac-

tice by answering questions about their gender, race, age, and occupation, then they

are able to handle the tough questions more readily.

Secondary Analysis of Existing Data. Given the enormous amount of time and

money it takes to conduct a survey from scratch, many sociologists rely on the

survey data previously collected from others. They may perform secondary analysis

of already existing data. Secondary analysis involves reanalyzing data that have

already been collected. Often this new analysis asks different questions of the data

than the original researcher asked.

Others may need to use existing historical data. After all, if you’re interested

in political debates in seventeenth-century France, you can’t very well conduct

a survey or interview the participants. Still others use content analysis to explore what

people actually mean when they give the sorts of responses they do.

122 CHAPTER 4HOW DO WE KNOW WHAT WE KNOW? THE METHODS OF THE SOCIOLOGIST

2000 Presidential Election

Sociological research is often used to gauge the political attitudes and behaviors of groups or of

the general public. You often hear about polls predicting voting behavior, and after elections we

have data about which candidate got how many votes. What we don’t have is the demographic

breakdown of who voted for whom. With a random, representative national survey, we can find

out how voting behavior varies along such lines as gender and race. If you didn’t vote in the

2000 presidential election, consider who you would have voted for. So, what do you think?

See the back of the chapter to compare your answers to national survey data.

4.2

What

do

you

think

❍Gore

❍Bush

❍Nader

❍Someone else

❍Didn’t vote

If you voted in the 2000 presidential elections, did you vote for Gore, Bush, Nader or someone else?

?