General Purpose, Questions, and Sample

In a Presidential election year, politics is a timely topic, but there were other reasons for making this the focus of the spring 2008 HCSS.  During the 1990s many commentators pointed to a new threat to democracy in America: Civic disengagement.  As signs of this crisis, Robert Putnam’s (2000) influential book Bowling Alone showed a steady decline during the last half of the century in Americans’ associational memberships, political interest and participation, church attendance, newspaper readership, and other indicators of civic involvement.  Social researchers who have examined this problem among college students—the future leaders of the nation—have tended to focus on two forms of engagement: voluntarism and politics.

According to various sources, beginning in the 1990s rates of volunteering increased, but political participation declined among college students (Longo and Meyers, 2006).  Young people showed diminishing levels of political knowledge and interest in public affairs (Zukin et al., 2006), and by the 2000 election, voter turnout among 18-24 year olds reached an all-time low.  A turnaround may be occurring in the new millennium, however.  College students were actively involved in the 2004 General Election, and voter turnout in the 18-24 age group rose substantially.  In last year’s HCSS, we found a very high rate of volunteering at Holy Cross; in fact, the rate far exceeded college students nationally.  So, we wondered whether we would find a similarly high level of political involvement, especially during a Presidential election year.

A survey on politics and current issues this spring also repeats the 2004 spring HCSS, thereby enabling us to compare Holy Cross students at two points in time.  Would we find greater interest in the 2008 campaign than in 2004?  Are students’ attitudes on key issues today similar to their views in 2004?

The survey asked students about the following:

1. Political participation, such as registering to vote and campaigning for candidates for political office.
2. Political ideology and political party preference.
3. Parents’ party preference and interest in politics.
4. Students’ interest in politics and public affairs.
5. Candidate preference in the 2008 Presidential election.
6. Involvement in student government.
7. Political knowledge.
8. Trust in government, satisfaction with the direction of the country, and approval of the President.
9. Opinions on current issues, including the war in Iraq, the economy, abortion, gay marriage, illegal immigration, health insurance, and the environment.
10. Volunteerism.

The bulk of the questions on political participation and interest were drawn from the National Election Study (NES), which conducts biennial surveys of the American electorate.  Questions on current issues were modeled after items used by Gallup and other polling agencies.  A few other questions were from the General Social Survey (GSS).

The survey took place between February 20 and March 19.  The target population and sampling frame consisted of students enrolled and on campus, thereby excluding those who were studying away or abroad or who had taken a leave of absence.  This amounted to 2,628 students, from whom we randomly selected a sample of 380.  A total of 340 interviews were completed, yielding a response rate of 89.5 percent.

Of the 340 respondents, 54.7 percent were female, 85.3 percent were white, 98.5 percent ranged in age from 18 to 22 years old, and 78.5 percent identified themselves as Catholic.  By academic class, 24.1 percent were first-year students, 35.6 percent second-year, 17.6 percent third-year, and 22.6 percent fourth-year.

Based on a sample of 340, the estimated margin of random sampling error in the spring 2008 survey is 5.5 percent, which means, for example, that if a reported percentage is 45, the percentage for all Holy Cross students is very likely (95 chances in 100) to fall between 39.5 and 50.5.  This estimate does not include error that may occur when respondents misinterpret questions, don’t remember correctly, or answer untruthfully (measurement error).  Nor does it include error due to nonresponse—differences between sampled individuals who participated in the survey and those who did not (unit nonresponse) or between those who responded to a question and those who chose not to answer (item nonresponse).