Teaching and learning online in political science.

Abstract

We report results from an analysis that assesses the effectiveness of teaching political science courses online. We look at three questions: How can the Internet effectively enhance student learning? What are the effects of online teaching on gender equality in political science classrooms? And how does participation in online discussion affect student learning? We find that active learning is possible online, that women benefit from online learning, and students overall do better when they read postings in online discussions.

Introduction

Over the past decade, the Internet has profoundly changed college teaching across the disciplines, including Political Science. Many instructors now use their course websites to augment or facilitate face-to-face lectures--posting PowerPoint slides, making announcements, posting tests or quizzes, or opening discussion boards for student-to-student interaction. Yet when it comes to courses delivered entirely online, many instructors remain skeptical that students can learn as much as in a traditional classroom setting. After all, both teaching and learning online are quite different in the two instructional modes (see, e.g. Lee 2003; McCormack and Jones 1998) since instructors' lecturing and students' listening is replaced by other activities, such as instructors posting notes and structuring assignments, and students reading, discussing online (often using asynchronous discussions), and completing assignments. Consequently, little is known about the effectiveness of online teaching in political science. Our research has aimed at illuminating how students interact online and how online learning affects learner outcomes. Here we report the results of a series of projects on designing and assessing political science courses taught partially or entirely online. In particular, our research has addressed the following questions:

1. How can instructors use the Internet effectively to enhance political science education?

2. How do web-based courses affect gender equality in the classroom?

3. Do assignments that encourage students to participate actively in online discussions improve student learning?

The Internet--More than Just a Source of Information

If the Internet is to add to students' learning experience, online teaching should be designed so that it goes beyond "fact-finding" missions, that is, looking up discreet facts and information online and reporting it to the instructor. If students' active engagement in the classroom stimulates and enhances the learning process, as the literature suggests (see, e.g Bligh 2000), then active learning should also be part of distance learning and online assignments should be structured accordingly.

Active learning can take many forms in the virtual classroom. For instance, students can participate in online discussions, interactive assignments, or critical thinking exercises. Many of these activities are made possible by online course management software (such as Blackboard or WebCT), which greatly facilitates the organization of online courses and interaction among students. Large classes can be split into smaller virtual discussion groups that can he kept constant throughout the semester, thus creating a seminar-like atmosphere for the students resembling smaller learning communities in a large class. Quizzes can be graded automatically, and essays or research papers delivered to the website where they can easily be managed by the instructor. Online assignments and discussion questions can stimulate students' critical thinking capabilities and allow them to practice their higher-order thinking skills (Bloom 1956) through analysis and application of textbook or online material, material presented by the instructor, or by other students. Online courseware also allows for easy management of group work and student-to-student communication including real-time chats and asynchronous discussion boards as well as internal course e-mail (for examples of some political science assignments that require such higher-order thinking skills, see Hamann and Wilson 2003).

In sum, online teaching does not have to be reduced to posting lecture notes online or streaming videotaped lectures to students staring at a computer screen while trying to take notes, and students do not have to be passive and isolated recipients of information. Instead, web-based instruction can embrace a student-centered, active learning approach that engages the student in the learning process. However, how do we know whether this type of instruction "works," and what the effects on the learners are? The next section discusses the implications of online teaching on gender equality.

Gender Equality in Online Classes

Existing research suggests that discussions in traditional, face-to-face classrooms tends to be dominated by male students. Given the absence of many social clues such as age or appearance, online discussion would seem to hold much potential for democratizing the classroom, perhaps promoting gender equity in student participation and gender parity learner outcomes. Yet, the evidence for gender equality in online classes is more mixed--some studies show that male dominance continues in the virtual classroom, while others conclude that gender equality tends to be more pronounced in online classes (see, for example, Herring 1993; Hum 2002; Bhappu, Griffith, and Northcraft 1997). However, none of the studies specifically assess gender equality in online discussions in political science classes.

To gauge gender equality in participation in web-based political science courses, we conducted a content analysis of the online discussion board postings of 453 students enrolled in multiple sections of three different comparative politics courses taught by two different instructors. The students participated in asynchronous discussion groups of 5-13 students. Many of the learning modules included a discussion component guided by discussion questions posted by the instructor. Students had to post a minimum number of meaningful discussion contributions in order to receive full credit for the discussion portion of the module, but there were no requirements as to whether they had to respond to others' postings or as to the length of the posting beyond the fact that it had to be substantial and meaningful. We split each posting into individual statements and followed a coding protocol adapted from Henri (1992) to see whether and how students interacted with each other. Thus, we wanted to find out whether students posted statements that were either independent (statements not related to previous postings), direct (statements that refer directly to a previous posting), or indirect (statements that refer tangentially to a previous posting). We also coded each statement to see whether it was evaluative or cognitive, and whether the contribution was a "surface" one or in-depth (that is, advanced the discussion by offering new reasoning).

When comparing gender patterns, we found little difference between male and female students with respect to the number of statements per message, the frequency of evaluative and cognitive statements, or the frequency of surface and in-depth statements. To illustrate, the mean number of statements per posting for men was 7.7 compared to 7.5 for women; 18 percent of male students' and 21 percent of female students' statements were evaluate; 78 percent of males' and 75 percent of females' statements were cognitive, and 48 percent of all statements made by men were surface, compared to 51 percent by women. Yet, more pronounced differences existed with respect to the interactive aspect of discussions. Women were significantly more likely than men to post independent statements, that is, to post a monologue, rather than to engage in a dialogue with students who had posted previously: Fifty-six percent of all statements made by women were "independent", compared with 46% for all statements made by males.

A closer look at the gender composition of the groups reveals further interesting and perhaps unexpected results. When we clustered the discussion groups by gender composition, we found that the more gender-balanced the groups were, the more both men and women engaged in dialogue rather than posted monologues. In contrast, the more the groups were gender-tilted, the less interaction and the more monologues were posted. [1] In other words, while women overall engaged less in interaction with their peers than men, this behavior was particularly pronounced when few males were in the group: When the group was 100% female, 71% of all statements were independent and 26% were dependent. [2] In comparison, when the group was equally split between men and women, 43% of men's and 46% of women's statements were independent, and 51% and 47%, respectively, were dependent statements, that is, referred to postings by other students. What does this mean for gender equality? We could not identify any male dominance either in volume or quality, but patterns of interaction changed depending on the gender composition of the class.

A second aspect of the gender democratization hypothesis concerns learner outcomes rather than learning behavior. How do women do in online classes compared with men? To address this question, we conducted a comparison of the learner outcomes in two courses. Both classes were freshmen general education classes on American Government taught by the same instructor using the same textbook and covering the same topics. One section was a traditional, primarily lecture-based class, while the other section replaced half of the face-to-face class time with online assignments and activities. To assess learner outcomes, a questionnaire containing 18 questions on basic American government facts and concepts was distributed in both sections at the beginning and at the end of the semester. The questionnaire was administered by someone other than the instructor, who had no knowledge of the questions. We found that while women in both classes initially scored lower than their male peers, the knowledge gained by female students who completed the mixed-mode section was higher than for any other group. Women in the mixed-mode section increased their knowledge by 3 points, compared with 2.8 points for men in the same section. Both groups in the mixed-mode section increased their knowledge more than their peers in the traditional lecture class: Women gained 2 points, and men 1 point. Thus, both male and female students learned more in the section that was partially taught online than those in the lecture-based class, and female students in particular did considerably better in the mixed-mode format compared with women in the traditional class.

Active Learning Online and Learner Outcomes

Existing literature shows that online courses can be as effective in furthering students' learning than traditional, face-to-face courses in Political Science. Botsch and Botsch (2001) compare student learning in American Government courses taught in class and web-based and find little difference in learner outcomes; if anything, students--especially low-GPA students--learned slightly more than their peers in the traditional classroom. Why is that? We wanted to find out whether the active learning components present in many online courses could provide an explanation. In particular, existing literature shows that discussion can be a useful active learning tool that improves learner outcomes (e.g. Bender 2003, Davis and Hillman Murrell 1993, Bligh 2000). We apply this finding to the online environment and in most of our assignments, online discussions play a central role. Yet, little is known about whether discussions conducted in the virtual classroom have the same effect. Thus, we were interested in investigating whether political science students performed better in online political science classes when they were actively involved in discussions.

To assess whether discussions in the virtual classroom improve learner outcomes, we analyzed the relationship between discussion participation and course grades for an upper-level comparative politics course on Latin American Politics. We measured three components of online discussion: The number of direct statements made by each student (roughly paralleling whether a student responds to a different student in a traditional classroom); the number of "in-depth" statements (an indicator of whether the student is grasping the material); and the number of postings each student read (a measure of whether students are "listening" to what their peers have to say). [3] None of these items were required as part of the assignment and none influenced the course grade itself. Finally, we controlled for prior-term GPA since students who have performed better overall prior to taking the class are likely to also do better in any particular course.

Not surprisingly, our findings confirmed that students with a higher GPA entering the course did better in this class : A 1-point difference in GPA (about one letter grade) produces a 10-point difference in the course average, also equivalent to about one letter grade. Yet, beyond the GPA effect, we found that discussion participation made a difference in course grades. Somewhat unexpectedly, however, the type of statements (direct or indirect, in-depth or surface) students posted had a negligible effect on course grade. Thus, it seems that the type of postings students engage in has little impact on their performance in the course. However, looking at the number of postings read produced a different picture: reading peer postings was positively and statistically significantly related to course grade. More precisely, students who read more improved their course average by about 6 points, approximately half a letter grade. This effect was particularly strong for students who came into the course with a lower GPA. Low-GPA students who read many postings in the class did much better than those who read less. For each posting read, low-GPA students improved their course grade by .156 points. What does that mean? If a student with a GPA of 2.5 reads 100 postings, this student would earn a course grade of 71, while 250 postings would result in a course grade of 80, almost an entire letter grade higher.

It appears that discussion is a useful active learning tool for political science students. The literature has already established the value of discussions in traditional classrooms. Our study shows that the same is true for the virtual classroom. Interestingly, it is the least "active" aspect of discussions--reading peer postings (equivalent to listening to other students)--that makes the biggest difference for learner outcomes, and similarly interesting, it is mostly those students who generally perform less well who benefit the most from reading their peers' postings.

Conclusions

Does teaching political science courses online "work"? In other words, do students learn as much as in traditional courses? And what can instructors do to promote student learning in the virtual classroom?

We found that it is not the teaching mode as such that drives student learning. That is, students can learn in any environment. Students with different learning styles might learn better in different classroom settings; someone who learns best by listening is probably better served in a lecture-based class rather than in an online class. Nonetheless, the delivery mode itself appears to matter less for learning outcomes than the type of teaching and learning that goes on within a classroom, be it traditional or virtual. To illustrate, some instructors are less apt at delivering lectures in front of a large audience, and students will learn less in one of their lectures than in one delivered by a different instructor. The same is true for online instruction. Online classes can be structured in many different ways, and it is thus the way the course is taught rather than the instructional mode in an of itself that helps or hinders student learning. Our research has shown that active learning can also be applied effectively to online instruction in political science. Specifically, we have found that students learn well in the online contexts; we also found that women learn well in web-based environment and that there is no evidence that men dominate the interaction in online discussion groups and that most interaction occurs when discussion groups are gender-balanced. Our analysis of political science courses supports the gender "democratization" hypothesis. Lastly, we found that engaging students in discussions benefits all students, but primarily those with a lower GPA. In designing assignments, it thus seems to make sense to encourage students to engage in a dialogue that involves reading other students' postings. Thus, online teaching appears to present a valuable addition or even alternative to teaching political science in the traditional classroom.

References

Bender, Tisha. 2003. Discussion-Based Online Teaching to Enhance Student-Learning. Sterling, VA: Stylus.

Bhappu, Anita D., Terri L. Griffith, and Gregory B. Northcraft. 1997. "Media Effects and Communication Bias in Diverse Groups." Organizational Behavior and Human Decision Processes 70 (3): 199-205.

Bligh, Donald A. 2000. What's the Use of Lectures? San Francisco: Jossey-Bass. Bloom, Benjamin (ed.). 1956. Taxonomy of Educational Objectives: The Classification of Educational Goals--Handbook I: Cognitive Domain. New York: McKay.

Botsch, Carol S. and Robert E. Botsch. 2001. "Audiences and Outcomes in Online and Traditional American Government Classes." PS: Political Science and Politics 34(1): 135-141.

Davis, Todd M. and Patricia Hillman Murrell. 1993. Turning Teaching into Learning: The Role of Student Responsibility in the Collegiate Experience. ASHE-ERIC Higher Education Report No. 8. Washington, D.C.: The George Washington University, School of Education and Human Development.

Hamann, Kerstin and Bruce M. Wilson. 2003. "Beyond Search Engines: Enhancing Active Learning Using the Internet." Politics & Policy 31(3):533-553.

Henri, F. 1992. "Computer Conferencing and Content Analysis." In A. R. Kaye (ed.), Collaborative Learning Through Computer Conferencing: The Najadan Papers. Berlin: Springer Verlag, pp. 117-136.

Herring, Susan. 1993. "Gender and Democracy in Computer-Mediated Communication." Electronic Journal of Communication 3 (2).

Hum, Susan. 2002. "Performing Gendered Identities: A Small-group Collaboration in a Computer-Mediated Classroom Interaction." Journal of Curriculum Theorizing 18 (2):19-38.

Lee, Donna. 2003. "New Technologies in the Politics Classroom: Using Internet Classrooms to Support Teaching and Learning." Politics 23(1):66-73.

McCormack, Colin and David Jones. 1998. Building a Web-Based Education System. New York: Wiley & Sons.

Endnotes

[1] Our gender-tilted groups were heavily female-dominated; we had no group where males constituted a strong majority. Our groups were randomly assigned without manipulating gender composition.

[2] The total is less than 100 because we also coded for other types of statements, such as social comments.

[3] The course management software our university uses automatically counts how many postings each student has opened. We assume that most students will read most of the postings they open.

Kerstin Hamann, University of Central Florida 

Philip H. Pollock, University of Central Florida

Bruce M. Wilson, University of Central Florida

Hamann (Ph.D., Washington University) is Associate Professor of Political Science; Pollock (Ph.D. University of Minnesota) is Professor of Political Science, and Wilson (Ph.D. Washington University) is Associate Professor of Political Science, all at the University of Central Florida.