Recent accounts attest to the educational benefits of involving students in political science research projects. For example, Rosenthal (Reference Rosenthal1999) found that participation in a group research project can foster a range of positive outcomes among undergraduate students. Similarly, Ishiyama and Breuning (Reference Ishiyama and Breuning2003) concluded that undergraduate participation in collaborative research with faculty is positively associated with both learning and entry to graduate or professional programs.
Research at the precollegiate level, in turn, suggests that civics education should provide a greater focus on critical analysis of politics. In particular, Niemi and Junn recommend that civics instruction in high school incorporate learning about “theoretical foundations of government” as well as “basic quantitative skills that are involved in reading charts and tables” (Reference Niemi and Junn1998, 151). The authors also criticize the political science profession's “hands-off policy toward the high schools,” arguing that political scientists should pay more attention to precollegiate learning (Niemi and Junn Reference Niemi and Junn1998, 158; Hepburn Reference Hepburn1987).
With these ideas in mind, we consider the potential for science fair projects to provide precollegiate students with opportunities to learn about political science research and faculty advisors with opportunities for sharing their knowledge with precollegiate students. Science fairs have long been part of precollegiate education in the United States, but teachers, students, parents, and even social scientists may tend to assume that student science fair projects must address “hard sciences” such as astronomy, biology, chemistry, geology, and physics. To challenge this assumption and illustrate the possible benefits of political science fair projects, we discuss a case study of one such project. The project in question revolves around an experiment that tested the effects of a televised political comedy sketch on political knowledge, opinion, and interest among high school students. One of the authors is the high school student who completed the project; the other is the faculty member who advised the student on the project.
THE PROJECT
The project began as part of the student's ninth-grade general science course, for which she was required to design a science project and enter it in a science fair competition. The student was interested in studying whether—and, if so, how—portrayals of 2008 Republican vice presidential nominee Sarah Palin on Saturday Night Live (SNL) influenced high school students. In September and October of 2008, the NBC television comedy program aired a series of sketches in which Tina Fey parodied the candidate. The Palin sketches helped boost the program's audience to 10 million viewers (a 50% increase from 2007), including over 2 million between the ages of 18 and 34, and were viewed on the Internet millions of times (Carter Reference Carter2008).
When the student presented the project idea to her instructor, the teacher expressed initial skepticism about its focus on a social science, or “soft science,” topic. In particular, he questioned whether it was possible to conduct scientific research on human behavior. The student and her parents, however, were able to allay this concern by finding a faculty advisor at a local university. Working with this advisor, the student designed the project, conducted it, and analyzed the results.
The student's study built on the growing body of research regarding the political effects of “soft,” or entertainment-oriented, news media (Baum Reference Baum2003). Of particular relevance for the student's project were recent studies suggesting that political comedy television programs such as SNL, The Daily Show with Jon Stewart, and The Colbert Report with Stephen Colbert are not only popular among young adults (Brewer and Cao Reference Brewer, Cao, Baumgartner and Morris2008; Pew Research Center 2008), but also influential among them. For example, exposure to political comedy programs is associated with greater knowledge about campaigns among young adults (Cao Reference Cao2008) and can produce more negative opinions about candidates among college students (Baumgartner and Morris Reference Baumgartner and Morris2006).
If political comedy television programs can influence young adults, then it seems plausible that they would influence adolescents as well. Moreover, effects produced by these shows among adolescents could be substantively important, given that the “impressionable years” of adolescence constitute a particularly crucial phase of political socialization (Jennings and Niemi Reference Jennings and Niemi1981; Sears and Valentino Reference Sears and Valentino1997). To shed new light on this topic, the student tested three hypotheses about the effects of SNL among high school students. First, she hypothesized that exposure to an SNL sketch parodying a political candidate would lead to increased knowledge about the candidate by providing low-cost and easy-to-absorb information (see Cao Reference Cao2008). Second, she hypothesized that exposure to the mockery presented by such a sketch would produce more negative opinions about the candidate (see Baumgartner and Morris Reference Baumgartner and Morris2006). Third, she hypothesized that exposure to the sketch would foster greater interest in politics by making politics seem entertaining (see Cao and Brewer Reference Cao and Brewer2008).
Methods
To test the impact of SNL among high school students, the student conducted a posttest-only experiment in January 2009 (two months after the 2008 presidential election). The 91 participants were ninth graders at the student's high school, which was located in a suburban, upper-middle-income school district in the Midwest. Given that these participants were 14 to 15 years old when the study took place, they will be eligible to vote in the 2012 election. Adult voters in the school district tilted toward the Democratic ticket in the 2008 presidential election, with Barack Obama and Joe Biden defeating Republicans John McCain and Sarah Palin by a 21-point margin in the district's four municipalities (compared to a 7-point national margin).
The study took place in a classroom setting. Before conducting the experiment, the student obtained approval from the high school's institutional review board. Participants were randomly assigned to one of three conditions (randomization was done in groups, resulting in unequal sample sizes across conditions). Those in the first treatment condition (n = 25) watched a 77-second video clip in which CBS Evening News anchor Katie Couric interviewed Sarah Palin. In the clip, which originally aired on September 25, 2008, Couric asks Palin about a proposed federal bailout of the financial industry (see table 1). Participants in the second treatment condition (n = 18) watched a 90-second clip taken from a September 27, 2008, SNL parody of the Couric interview in which Amy Poehler (playing Couric) asks Tina Fey (playing Palin) about the bailout (see table 1). Participants in the control condition (n = 45) watched a 62-second video clip on an unrelated topic.
Table 1 Experimental Treatments—Transcripts of Television Clips
After watching the video clip for their condition, all participants completed a posttest. The questionnaire began with a series of questions about the video that participants had just watched. Subsequent questions measured the key dependent variables under study (see table 2 for question wording): political knowledge (specifically, knowledge regarding Palin's position on the bailout), political opinion (in the form of views regarding Palin), and political interest. Two additional questions measured participants' partisanship (59% identified as Democrats, 23% as Republicans, and 18% as neither—a pattern that presumably helps explain the tendency toward negative assessments of Palin across conditions; see table 2) and gender (50% female, 50% male).
Table 2 Political Knowledge, Opinion, and Interest, by Condition
Note. Table entries are proportions or means; standard errors for means are in parentheses.
a Proportion differed from proportion in control condition at p < .05 (one-tailed test);
b mean differed from mean in control condition at p < .05 (one-tailed test);
c mean differed from mean in control condition at p < .10 (one-tailed test);
d mean differed from mean in Couric interview condition at p < .05 (one-tailed test).
Results
Compared to the real Couric interview, the SNL parody was more familiar to the high school students who participated in the study: only 12% of participants who watched the Couric interview said that they had seen it previously, whereas 56% had previously seen the SNL parody (p < .01, two-tailed difference of proportions test). Participants rated the Couric interview clip as more informative (M = 1.08 on a 0 to 3 scale) than the SNL clip (M = .78), but not significantly so. They rated the SNL clip (M = 2.44 on a 0 to 3 scale) as significantly funnier (p < .01, two-tailed difference of means test) than the Couric clip (M = 1.24).
Participants in both treatment conditions evidently increased their knowledge about the candidate by watching the clip for their condition (see table 2). Only 14% of the participants in the control condition correctly responded that Palin favored the $700 billion bailout of the financial industry, whereas 32% of those who watched the Couric interview and 33% of those who watched the SNL parody provided the correct answer. Both proportions differed significantly from the proportion in the control condition (p < .05 in each case, one-tailed test). There was no evidence that participants learned more from one clip than from the other. Given the decline in television news viewership among the young (see, e.g., Mindich Reference Mindich2008), it may be reassuring that high school students can acquire at least some political information from alternative sources.
The results indicate that exposure to the Couric interview influenced some aspects of opinion about Palin (table 2): compared to those in the control group, participants who watched this interview saw Palin as less qualified (p < .05, one-tailed test) and reported a lower likelihood of voting for her in a future presidential election (p < .10, one-tailed test). In contrast, no significant differences in opinion about Palin emerged between participants in the SNL condition and those in the control condition (table 2). The absence of significant effects on opinion for the SNL parody could reflect the relatively small sample size for this condition. Then again, none of the contrasts between opinion in the SNL condition and opinion in the control condition even approached statistical significance. A more plausible explanation is that when high school students were asked to evaluate the candidate, they discounted the parody of Palin as humorous fakery. Another possibility is that the impact of the parody on opinion was blunted by students' familiarity with it (recall that over half of participants in the SNL condition had already seen the clip).
Lastly, the results indicate that exposure to the SNL parody did more to foster political interest than exposure to the real news interview. Compared to participants in the Couric interview condition, those in the SNL condition expressed greater interest in politics (p < .05, one-tailed test; table 2). On the one hand, this finding may reflect poorly on the capacity of traditional television news to foster political engagement among adolescents. On the other hand, the same result suggests that news parodies may, by making politics seem funny, also promote the future health of democracy.
In sum, the study yields evidence for two of the student's three hypotheses. To be sure, there are limits to what one can do in a science fair project at the high school level (let alone at earlier stages of the educational process). Accordingly, the design for the project in question was relatively simple and modest. The experiment used a small convenience sample rather than a larger and more representative sample, and the analyses excluded a host of individual-level and contextual factors that could have moderated the observed effects. Even so, the study generates new knowledge about the effects of political comedy television programming among high school students—a population that Niemi and Junn (Reference Niemi and Junn1998) argue deserves more attention from political science researchers.
REFLECTIONS ON CONDUCTING AND ADVISING POLITICAL SCIENCE FAIR PROJECTS
One observation that we derive from our experience is that collaboration between high school students and faculty advisors on political science fair projects can introduce students to the political science research process and, in doing so, to a critical and analytical perspective on politics. In the case at hand, the student learned about multiple aspects of political science theory and methods. In regard to the former, she learned how to draw a theoretical framework from previous research, derive hypotheses from this framework, and use her own results to evaluate both the hypotheses and the underlying theoretical framework. In regard to the latter, she learned how to design a true experiment, construct a survey, conduct basic statistical analyses, and present the results of these analyses using tables and charts. Thus, the project developed the kinds of theoretical and quantitative skills that Niemi and Junn (Reference Niemi and Junn1998) identify as crucial to the development of future citizens. Furthermore, it provided a student who did not see herself as a “science type” with an opportunity to learn about scientific methods in the context of a topic that excited her interest. It is difficult to draw generalizations about the educational effectiveness of such projects from a single case, but our experience leads us to be optimistic about the potential for other political science fair projects to achieve the same ends.
At the same time, we suspect that some topics and methods may be particularly suited for political science fair projects. In terms of topics, both our findings and our experience suggest that projects dealing with the role of entertainment media in politics may be especially appealing to precollegiate students. Just as watching the SNL clip inspired greater political interest among participants in the study, watching such clips and discussing them with peers inspired the student author to conduct a political science fair project. By the same token, topics revolving around other entertainment or recreational media popular among precollegiate students may provide fertile ground for political science fair projects. For example, students might be attracted to projects examining the political effects of social networking sites such as Facebook (which hosted numerous political groups during the 2008 presidential campaign) or video games such as Guitar Hero and Madden '09 (which included paid advertisements for 2008 presidential candidate Barack Obama).
In terms of methods, we believe the experimental approach to be a good fit for high school science fair projects. This approach is not only common to many hard sciences (and thus should be familiar to precollegiate science instructors), but it is also easy to use and powerful. Students can design simple experiments that require only modest resources (in the present example, a free video-sharing site along with pen-and-paper questionnaires), and even experiments with a small number of conditions and a small number of cases (three and 91, respectively, in the case at hand) can generate statistically significant findings. Moreover, experimentation allows for relatively simple and intuitive statistical analyses (e.g., comparisons of means and proportions across conditions).
Although we see much potential in political science fair projects, our experience suggests that students can face obstacles to conducting such projects. Specifically, they may face challenges in obtaining support and guidance from science instructors who lack familiarity with or interest in social science research. In the case at hand, the student (and her parents) first had to overcome her science instructor's skepticism about the legitimacy of social science research. In his initial meeting with students' parents, he discouraged projects that involved human behavior because he saw these types of projects as being too difficult and not “concrete enough.” Once the student received approval to proceed with her project, she faced difficulties in obtaining meaningful guidance from the instructor. In particular, the student believed that she and other students who conducted social science projects received less attention and feedback from her instructor than did peers who conducted more traditional hard science projects (e.g., projects addressing tooth decay, rocket propulsion, and cancer cell research).
The involvement of a faculty advisor can help students overcome these sorts of obstacles. In our case, the student would not have been able to conduct her project without the assistance of a faculty advisor. The involvement of the advisor was crucial to legitimizing political science research as “scientific” and thereby winning the instructor's approval for the student's project idea. Furthermore, the faculty advisor provided extensive feedback on theory, methods, and analysis after the student began designing her project. Thus, our experience suggests that faculty advisors can alleviate institutional skepticism regarding science fair projects that focus on politics and can provide guidance on such projects that may not otherwise be available to students.
The involvement of a faculty advisor in a political science fair project can benefit the advisor as well. The faculty author had never considered the possibility of advising a political science fair project before being asked by the student and her parents to do so. Having advised one such project, he would recommend the experience to colleagues as a way to become involved in precollegiate political science education and share knowledge beyond the university setting. Faculty advisors for political science fair projects may be surprised and pleased, as he was, by the extent to which high school students are capable of originating, designing, and conducting theoretically grounded and methodologically rigorous political science research (even if it may be unrealistic to expect most student-originated projects to generate novel insights into politics). Moreover, involvement in a political science fair project can provide the advisor with insight into political science education. In our case, the advisor gained a new appreciation of the extent to which political science research can confront a legitimacy problem at the high school level. Another, and happier, insight he gained is that political science faculty members can help to address this legitimacy problem by becoming involved in precollegiate education.
We would like to thank the reviewers of PS: Political Science and Politics for their useful suggestions and David Allen for his help in making the project happen. We would also like to thank the students at Nicolet High School, Glendale, Wisconsin, who participated in this study.
