I read Anil G. Jacob's piece in the January 2006 issue ofPS (“Asian American Political Participation:Research Challenges for an Emerging Minority,” 103–106) with amixture of admiration and consternation. He has my admiration forgetting published as a graduate student, and that too inPS—a feat that I'm sure will stand him goodmeasure when he is on the job market. However, he has myconsternation because he makes slight of a field that has,especially in the last five years, produced a good deal ofscholarship on a myriad of questions relating to Asian-Americanpolitical behavior.

By far the most significant form of gerrymandering is partisangerrymandering. It is painfully ubiquitous and yet can be covert,hidden in plain view. More importantly the advent of the computerand the growing polarization of the electorate have enhanced itsimportance. Already the House of Representatives has becomevirtually safe for most incumbents and in recent electionsexperienced less turnover than the Senate, which has far fewermembers and only one-third of its members up for reelection.

In recent years, a growing chorus of voices in the political sciencediscipline has become concerned with the balance among alternativemethodological approaches in empirical research and publications.Accordingly, these scholars have issued a call to further developand refine rigorous methods for qualitative studies, in contrast tostudies that rely on quantitative methods and formal modeling(Bennett, Barth, and Rutherford 2003;Yanow 2003; George and Bennett 2005). My own interest and reason forconducting the following survey is rooted in the observation, evenfrustration, that the literature on qualitative research methodslargely focuses on democratic and not on authoritarian regimes orthe Middle East in particular. Research in the Middle East isclearly essential and has received increased attention since theterrorist attacks of 9-11, but given how critical this work is andwill continue to be in coming years, what are the challenges? Doqualitative methodological tools need to be adapted for research inthe Middle East? And, if so, how?Theauthor would like to thank Anna Szczurk for her help indesigning the survey and the contributors to the symposium andMelissa Gabler for their help in the research and writing of thearticle. The author also is grateful to the survey respondentsfor the time and effort they devoted to answering thequestions.

Field research on politics in the Middle East poses far less problemsfor Western female researchers than many Westerners assume. Thequestions most commonly asked of me by colleagues include whetheranyone will speak to an American woman (yes), and whether I need towear a veil (no, but see below). Female researchers do face manychallenges, but most have less to do with gender than with examiningsensitive political issues in highly repressive environments. Ofthose that do concern gender, many vary by degree rather than typefrom the issues faced by female scholars conducting field researchelsewhere. I draw heavily on my own overwhelmingly positiveexperiences with ethnographic research in Jordan and Yemen—totalingmore than four years over the past decade, including nearly a yearof research conducted post-September 11—and those of many friends inoutlining the following issues facing Western female politicalscientists conducting field research in the Middle East.I would like to thank Laryssa Chomiak,Janine Astrid Clark, Sam Fayyaz, and Lisa Wedeen for theirthoughtful comments on earlier versions of thisarticle.

Because it is a slice of one's life as well as a method for gatheringdata and insights, and because it entails socio-cultural immersionand interpersonal trust, field research confronts politicalscientists with a special set of normative issues not encountered inlibraries, laboratories, or home countries. This essay weighs fourplausible yet fallible positions on the ethics of fieldworkoverseas, with particular reference to American political scienceresearch in the Arabic-speaking world in the 21stcentury. I am raising questions I cannot really answer:specifically, amidst American military engagement and a good deal ofspying by governments all around, can we, or ought we, practicescholarly detachment, reciprocity, activism, or, perhaps,espionage?

This essay describes two very different survey projects thatinvestigate the political attitudes, values, and behavior patternsof ordinary men and women in the Arab world. One is the ArabDemocracy Barometer, an American-Arab collaborative project beingcarried out in five countries. The other is a survey in Palestineconducted as part of the fieldwork for a doctoral dissertation.These projects illustrate the emerging opportunities for politicalattitude research in the Arab world. The essay begins, however, witha brief reflection on the history of political surveys in the Arabworld, which is necessary to appreciate the significance of theopportunities now emerging.

In many people's minds, the Middle East stands out as the world'smost dangerous place. I often remark to my colleagues and friends,however, that I feel safer doing field research in most MiddleEastern countries than I would in much of Africa or Latin America.To begin with, few parts of the Middle East suffer from high ratesof random crime. Rather, the region's violence is mostly politicalin nature, and, with the right approach, a researcher can takeseveral steps to minimize risks. I have conducted research in theKurdish regions of Turkey, Iran, Syria, and Iraq (before and afterthe 2003 war), in Lebanon, and in Israel-Palestine during the firstIntifadah. In this article, I discuss various strategies and issuesone faces while “in the field” in regions such as these.The author would like to acknowledgeresearch support provided by the Canadian Department of NationalDefence, the Inter-University Consortium for Arab and MiddleEast Studies (ICAMES), and the University of Montreal's Centred'Études et de Recherches Internationales.

All manner of political observers are fascinated by “gaps” in votingbehavior. Whether it is the now-famous gender gap, the newlydiscovered religion gap, or the once-prominent generation gap, sharpdifferences in partisanship and voting behavior often emerge aroundcommonplace demographic characteristics such as gender, worshipattendance, and age. These gaps are not just intrinsicallyinteresting; they also offer a potent way to understand electionresults. Like batting averages in baseball, such simple statisticsoffer the power of language in describing the political world.

Gender differences in political behavior are a source of ongoinginterest for political pundits, campaign advisors, and students ofAmerican politics. In a closely divided nation, even small shifts inthe political choices of men and women can have significantelectoral consequences. In politics, the gender gap refers tomale-female differences in party identification and voting behavior.As a practical matter, the contemporary gender gap (with women morealigned with the Democratic Party) emerged in 1964 with the electionof Lyndon B. Johnson. The gender divide has grown incrementally overtime, hitting a contemporary high of 14 percentage points in 1996(see Figure 1). Since then, the size ofthe political difference between men and women has declinedsomewhat; the 2004 gender gap in party identification and votingfell to 9.5 and 7 points, respectively. The 2004 vote gap was halfof what it was at its apex in 1996, and while one hesitates to maketoo much of a single point on this larger time series, a sustainedcontraction of the gender gap could pose rather dire circumstancesfor the Democratic Party, or, conversely, great fortune for theRepublicans.

As the 2004 presidential election approached, journalists discoveredan apparently new political phenomenon: the “religion gap,” or thetendency of the most religious Americans to espouse conservativepolitical beliefs and prefer Republican candidates. As a typicalnews story put it, “Want to know how Americans will vote nextElection Day? Watch what they do the weekend before. If they attendreligious services regularly, they probably will vote Republican bya 2-1 margin. If they never go, they likely will vote Democratic bya 2-1 margin” (Thomma 2003). The religiongap was illustrated frequently and clearly by public opinion polls.A Pew Research Center (2003) study, for example, found the Americanpublic sharply divided between the Republicans and Democrats on theeve of the 2004 campaign, with frequency of worship attendanceemerging as a powerful predictor of party preference. In fact, mostevidence of the religion gap was linked to differences in worshipattendance.

Political divisions based on class would seem to be central tocontemporary American politics. Inequalities in the distribution ofincome and wealth have grown steadily over the last 30 years,creating greater differences in opportunity and challenging thelongstanding American belief in equality of opportunity (Hochschild1995). The two major parties differsharply in their concern for class issues, making class politicaldivisions all the more likely.

The time does seem to be ripe for revisiting the urban-rural divide,with an eye toward understanding what lies beneath it. The last twopresidential elections have revealed an urban-rural cleavage that ishard to ignore. Most observers now recognize that the “red” vs.“blue” Election Night maps really mask an urban-rural dividewithin states—a gap that has increased inrecent years according to a range of definitions. In 2004, thedifference in Democratic support between the most and least populouscounties in the nation exceeded 25 points (see Figure 1). Survey marginals are comparable, althoughrural populations are not well represented in most national polls.Tabulations from the 2004 American National Election Studies (ANES)indicate a 20-point gap in presidential preference betweeninhabitants of counties with more than a million people and those innon-metro counties of less than 25,000.

Weighting survey data based on social background characteristics suchas age, education, gender, and race is a standard practice in publicopinion polling. By adjusting the composition of their samples toconform to known demographic characteristics of the population,pollsters greatly increase the accuracy of their results. However,with the exception of the Zogby Poll, most major pollingorganizations, including the Gallup Poll, have strongly opposedweighting their data based on party identification. This oppositionis understandable. Party identification is not a fixedcharacteristic of the electorate. It is a political attitude thatcan vary over time. As a result, most pollsters believe that thereis no way of accurately determining the underlying distribution ofparty identification in the population for weighting purposes.

In November 2000, exit poll interviews with voters in Floridaindicated that Al Gore won the state. As a result, many televisionnetworks declared Gore the winner of Florida, a pivotal state towinning the presidency in 2000. Only a few hours later, the firstvote tallies from the Florida Secretary of State's office revealedthat George W. Bush was in fact leading in Florida. After 45 days ofrecounts and lawsuits, it was clear that the exit polls were wrong;Bush had won the state by the narrowest of margins. As a result ofthe flawed exit poll the media and pollsters scoured and reanalyzedthe methodology used in 2000 to prepare and correct for the 2004presidential election. The old system, Voter News Service (VNS) wasscrapped entirely, and Edison-Mitofsky Research was chosen toimplement a new and more accurate national exit poll in 2004 by aconsortium of news organizations retained by the Associated Presscalled the National Election Pool (NEP). What happened? Exit pollresults from Edison-Mitofsky showed John Kerry ahead in Ohio,Florida, and New Mexico—all states which he lost to Bush in2004.Author names are listedalphabetically. The co-authors were also the co-principalinvestigators of the Loyola Marymount University 2005 LosAngeles Mayoral Exit Poll. Thanks to Salvador Paniagua and HavenPerez for their tremendous research assistance in implementingthis project and to the more than 120 student researchers whoparticipated in the exit polling and data entry. Robert Aguinagaand Antonio Gonzalez of the Southwest Voter Registration andEducation Project also provided valuable assistance inimplementing the poll. Mark Blumenthal, of mysterypollster.comwas instrumental in tracking down exit poll archives.

Conferences are an integral part of academic life. They generatefocal points for intellectual exchange and encourage professionalsocialization. They help scholars working on similar issues toidentify others who might be part of their specialist communitywhile also creating opportunities to learn about research thatcolleagues are doing in different areas of the discipline. In otherwords, conferences provide an opportunity for scholars to presenttheir work in progress, to see what others are working on, and tonetwork with their peers.The authorswould like to thank PS's anonymous reviewersfor their comments and suggestions.

No—this is not what it seems. This is not a discourse on statistics,but it is a discussion of a type of methodology—that of how, whenthey are writing, political scientists can better approach theiraudience. Put differently, this article is related to the moregeneral proposition that scholars need to demonstrate not only thatthey can conduct research but also that they can communicate iteffectively to a wide audience. And, although critical of somesocial science writing, it carries a suggestion for increasing yourrange of publication options.

In recent years, a number of prominent political commentators haveraised concerns about the lack of ideological diversity on collegecampuses (Shapiro 2004; Black 2004; Kors and Silvergate 1999; Kimball 1998). Amongother accusations, they claim that liberal college professors mayactually penalize students for expressing conservative opinions byassigning them lower marks on exams and assignments (Horowitz 2003; Hebel 2004).Their concern is not without merit. Researchers have found that,when evaluating a colleague's research, college professors are morecritical of work that contradicts their own views (Mahoney 1977). It is logical to assume that the samebias influences professors' evaluations of students' arguments. Itis also reasonable to expect that students, charged with theimportant task of evaluating their professors, are vulnerable totheir own ideological biases.We thankMarkus Kemmelmeier and Janet M. Box-Steffensmeier for theirhelpful suggestions and valuable guidance. We also acknowledgethe valuable support of research assistants Jessica Defenderfer,Cristina Ciocirlan, and Kathleen Winters. Finally, we would liketo acknowledge a debt of gratitude to the dozens of politicalscience faculty throughout the country who took the time todistribute our survey to their undergraduate students. Thisproject would not have been possible without their generousassistance. This research was funded through a Faculty ResearchGrant from Elizabethtown College.

Beginning in March 1942, three months following the Japanese attackat Pearl Harbor, and lasting until as many as 16 months followingthe end of World War II, slightly more than 120,000 persons ofJapanese ancestry were excluded, detained, and held in “relocationcenters” by the United States government, ostensibly because theywere considered a threat to national security. Nearly 70% wereAmerican citizens by birth; the rest were Japanese nationals whowere legally barred from naturalization because of the dejure racist policies of the time (Daniels, Taylor, andKitano 1991). Despite this treatment, over1,200 individuals volunteered to serve in the U.S. armed forceswhile several thousand others were drafted from the relocationcenters. Most served in a segregated unit in the European Theater,while others served as interpreters in the Pacific Theater, allwhile their families remained behind barbed wire in relocationcenters. These individuals served with great distinction within someof the most highly decorated units of the U.S. Army (Crost 1994).

Requiring a short research paper in introductory political sciencecourses can prove challenging for instructors, in part because ofthe diversity of student grade levels and majors in such courses.This is particularly the case with Introduction to ComparativePolitics, as, at most universities, this course is not only requiredfor the political science major, but often also fulfillsmulticultural requirements and can be used to complete lowerdivision social science requirements. In deciding whether to includea research paper assignment in this course, one must weigh thebenefit of catching majors early in order to begin building strongresearch skills, as well as the benefit of immersing students in keyissues affecting the larger world, against the challenge ofunderdeveloped writing and research skills among the typical studentin such a course. Another factor to consider is the grading load ofreading large numbers of papers, given that most introductorycourses have high enrollments. When choosing to require a researchpaper under these circumstances, then, one must also commit toworking with students to hone their research and writing skills. Ina collaborative, mutually beneficial partnership, college andlibrary faculty can work together to identify ways to integrateinformation literacy concepts into course content and to designeffective learning assignments addressing these important skills.This approach was undertaken by the authors, who revised asignificant portion of the Introduction to Comparative Politicscourse at the University of West Florida to integrate informationliteracy activities in support of a semester-long, content-basedresearch and writing assignment. This article explores usinginformation literacy exercises as a tool to produce higher qualityresearch papers in lower division political science courses.

Undergraduate classroom presentations of the scientific method in thesocial sciences usually involve a lecture on the nature of thescientific enterprise: the discovery of the “truth” (if such a thingeven exists) about the relationship between some set of concepts,events, or phenomena by means of generating falsifiable hypothesesand testing them against data. The tendency of faculty, howeverunderstandable, to convert the presentation into an abstractdiscussion of the philosophy of science often obscures the basicissues for students. I present here a simple, active learningclassroom exercise which sensitizes students to the difficulties ofhypothesizing about and testing unobservable phenomena by appealingto students' intuitive understanding of the nature of knowledge andreality. I first describe the purposes and objectives of theactivity in more detail, then discuss implementing this specificactivity. Suggestions for activity adaptation and development byother instructors follow, and a brief discussion of sources andresources concludes.