The Political Context and Field Research

When questioned as to the greatest difficulties encountered in the field, respondents overwhelmingly reported issues that directly or indirectly were a result of the authoritarian political climate. While researchers' experiences vary widely, 45% of the respondents noted what one researcher broadly speaking called “the looming smell of the mukhabarat” (internal security or secret police). This was reflected in a variety of ways. Forty percent of the total respondents stated that they had experienced difficulties in obtaining interviews with key individuals. The most ubiquitous reasons cited were the political sensitivity of the topic and interviewees' unwillingness to speak openly due, most commonly, to political repression and, to a lesser extent, potential retaliation from opposition groups. In addition, some interviewees were simply inaccessible as they suffered ongoing harassment by the authorities or had fled the country due to political tensions. Once having obtained access to interviewees, 27% of the respondents stated this “looming smell” most acutely manifested itself in a pervasive “culture of suspicion,” as evidenced by interviewees' mistrust and nervousness in speaking frankly to researchers for fear of political repercussions. Eleven percent stated that they had been denied permits or authorizations required for conducting surveys or obtaining access to documents or photocopying documents. In fact, 22% of the researchers noted that they at one point had difficulties gaining entry to the countries of research or obtaining research visas due to the perceived political sensitivity of their topics by the host governments. Others reported that they had experienced the threat or actual seizure of their research data (5%), surveillance and monitoring by security (4%), arrest and/or detention (4%), and police harassment (2%). Finally, 13% of researchers expressed issues related to political violence—colleagues or friends being attacked, tear-gassing, and a general lack of safety—as among the greatest difficulties in conducting research.

This culture of suspicion is exacerbated by the international political climate. Twenty-seven percent of researchers specifically identified anti-Westernism (usually in the form of anti-Americanism) and the general suspicion and distrust of U.S. policies and perceived agendas as impeding their efforts to undertake field research. Researchers noted the common perception that American researchers in general may be connected with the CIA or other intelligence agencies. In conversations, colleagues also have noted the relatively new concern that they must demonstrate that they are not connected to academic programs or centers related to the study of terrorism (and presumably, counter-terrorism), such as the Department of Homeland Security's Terrorism Center at the University of Maryland.

For most researchers, these political concerns are compounded by logistical challenges largely due to the overall “Third World status” of the region. Thirteen percent of researchers complained that their greatest difficulties were related to a lack of reliable data and statistics, including archives, which impeded, amongst other things, the ability to define sample sizes. Twenty-nine percent of respondents were frustrated by the challenges related to locating interviewees and data—the dearth of good directories and phone books, unreliable Internet, outdated telephone numbers, lack of accurate addresses, restrictive opening hours, and the cancellation of appointments. As one researcher stated, all these hurdles mean that everything takes longer than you think.

As a consequence of these political and logistical considerations, respondents' choices of countries for field work are strongly informed by issues of feasibility. While respondents stated that their primary reason for choosing a country was based on the methodological dictates of their research questions, 47% of researchers chose, avoided, or abandoned countries based on feasibility issues related to the political climate, familiarity with a country (due, for example, to language training), and, to a lesser extent, language, contacts, and funding. When more than one country could fill the methodological criteria of their research, scholars were strongly influenced by the availability of research permits, their own language skills, and the degree to which they had established contacts in the field. The institutional mechanisms for conducting research, including even the prevalence of telephones, are not established for all types of research projects, particularly given time and financial constraints. Thirty-one percent of researchers stated that their country choices were based strictly on the dictates of their research questions. It remains to be determined the extent to which these research projects and/or researchers share certain commonalities. While general perceptions are that the region is an unsafe place for the researcher, only 16% of respondents specified issues relating to the political situation and safety as contributing to their country choices.

Not surprisingly, only a very limited number of researchers choose to conduct research in conflict zones. While the researcher may not be in direct physical danger, issues of validity become too great as challenges of representativeness, such as the impeded ability to travel and discuss politics freely, access to relevant individuals and materials, and the ethical concerns related to putting interviewees in danger, become acute. What this means, however, is that countries experiencing the most acute political upheavals are least studied in terms of in-depth field research. In this survey, five researchers (9%) have conducted research in Iraq—however, only two researchers (4%) are conducting field work in present day Iraq. As David Romano elaborates in his essay in this symposium, for these researchers the threat of being detained or harmed is very real—the challenges related to research, however, are not insurmountable.

Locating and Contacting Interviewees

For ethnographic research in particular, researchers need to overcome interviewees' suspicion and to build rapport and trust, and this is reflected both in how researchers locate and contact interviewees and in their interview techniques. Within the social sciences, “snowball sampling,” broadly defined as a technique for finding research subjects by which one subject gives the researcher the name of another subject, who in turn provides the name of a third, and so on (Atkinson and Flint 2001), is a non-probability sampling strategy used to locate subjects with certain attributes or characteristics necessary in a study of a particular population (Berg 2004, 36). Most commonly, researchers employ the term “snowballing” in reference to a method of contact, specifically a chain of referrals (usually via social networks) to secure interviews (Berg 2004, 161). In both cases, snowballing is used to obtain interviewees when they are few in number, “hidden,” or where some degree of trust is required to initiate contact (Atkinson and Flint 2001). In the absence of reliable data on samples, snowballing is especially useful. It is for precisely these reasons that the majority of researchers in the survey sample rely on snowballing as critical to gaining entry to organizations and interviewees (Bernard 2000, 179–180). Eighty-nine percent of respondents stated that they rely on snowballing—used by respondents in the looser sense of a chain of referrals—as a means of identifying their interviewees. In this way, researchers are able to refer to a common contact and thereby establish a degree of trust when setting up a new interview. Most revealingly, 71% of those who snowball (64% of the entire sample) stated they commonly begin their snowballing efforts with their personal friends and social networks.⁴

Once having identified their interviewees, the vast majority of researchers contact them directly via the telephone. While a significant number of researchers expressed strong reservations about the effectiveness of official letters of introduction (some even expressing concern that they are an impediment to securing an interview), 36% of the respondents contact potential interviewees by emails, letters, and, to a lesser extent, faxes. Of these, 30% (11% of the entire sample) use them only as a means of introduction (usually when they are still in their home country) and do not rely on letters/emails or faxes to establish their interview. Rather, they follow up with a phone call when they arrive in the country of research. Further research needs to be conducted as to for which population official letters of introduction and/or emails and faxes work best.

In the effort to gain the necessary trust for a successful interview, 44% of researchers use or have used a third party to set up their interviews. Of these, most commonly, 46% (20% of total sample) rely on friends. Others requested that a previous interviewee, a local academic, or their local research assistant contact the desired interviewee and secure an appointment. As discussed below, some researchers stated that their researchers and translators, by virtue of being from the country, initially were granted greater trust by interviewees.

A limited number of respondents stated that their field work indicated that the higher the potential interviewee in the political hierarchy, the more vital local contacts are in order to secure interviewees. However, the survey results demonstrate a perceived need and emphasis by all respondents to use contacts as much as possible when dealing with interviewees regardless of the position or socio-economic status of the interviewee. Further research needs to be conducted examining the variation in the need for contacts according to class and country in the region.

Conducting Survey Work

The same difficulties in securing interviews also presented obstacles in respondents' abilities to conduct reliable survey work in the region. While the majority of researchers in this survey undertook in-depth unstructured interviews largely as an effective technique to build rapport and trust, 25% of respondents have undertaken surveys (some of whom have conducted multiple surveys). Of these respondents, 78.5% have conducted qualitative surveys (or mixed qualitative and quantitative surveys) and 28.5% have conducted quantitative surveys. An additional two researchers attempted survey work but were forced to abandon their efforts due to the political sensitivity of their questions and respondents' resistance to answer questions.

Once again, respondents' greatest frustrations were the reluctance of people to complete the surveys and/or give potentially misleading answers—those in line with official rhetoric. Respondents expressed concerns in assessing the frankness of their survey responses and questioned the degree to which the fear of answering survey questions honestly may have invalidated the survey results. One researcher confirmed the difficulty in even carrying out a pilot study in order to test the validity and reliability of the qualitative survey questions' wording. In some cases, the researcher's task was not made easier by the censoring of questions and harassment by the authorities, despite having official permission to conduct a survey.

A second area of concern was the degree to which those who were being surveyed were familiar with the meaning of the concept as intended by the researcher given their relative newness and/or rarity in some of the countries of the region due to restrictions on freedoms of speech (construct validity). Amaney Jamal and Mark Tessler found this to be especially true for traditionally marginalized voices, such as women and/or the poor, who, due to lack of education and/or familiarity, required assistance in understanding the survey instructions. As they explain in their article in this symposium, this required that the researchers hire additional assistants to meet with survey respondents on a face-to-face basis. Other researchers noted survey comprehension difficulties due to language, translation, and cultural or socio-economic issues. Questions do not always translate clearly in varied contexts (question reliability). One respondent noted that he/she posed a simple question as to the number of adults living in the household and was faced with the issue of appropriately defining the word “adult.” Is a young woman who is married and has her first child at age twelve an adult? These issues of comprehension and frankness are more acute in a structured, closed-ended survey where researchers are bound by a set of questions that they are reluctant to alter for fear of invalidating the survey results.

Researchers noted tremendous variation across countries in their ability to conduct survey work, even when attempting the same survey. Indeed, survey experiences in the region were not only varied but, to a degree, surprising, with more than one researcher noting the difficulties of conducting qualitative and quantitative survey work in Jordan, a relatively free country, and one researcher expressing the relative ease with which he/she was able to conduct a large scale quantitative survey with over 400 respondents in conjunction with a local university in Tehran.

The majority of surveys (all of the qualitative surveys and some of the quantitative) were conducted by the researcher him- or herself or together with a small number of assistants. As Jamal and Tessler note, this is partially due to the fact that reliable polling institutions are few and far between in the region. Only relatively recently have a number of research and polling institutions with reliable credentials become established. Furthermore, some researchers strongly felt that the personal touch was needed to ensure the trust necessary to gain access to people. Indeed, others relied on networks, professional and social, to vouch for the credibility of the survey projects. Only one survey, a quantitative survey conducted by Jamal and Tessler, used a polling company. As they elaborate, an important key to the success of quantitative survey work is that it be conducted on a face-to-face basis and that it be contextualized in conjunction with qualitative interviews.

Ethical Concerns

In terms of actual interview techniques, they too are impacted by the political conditions. Indeed, the need to focus on trust-building as a result of the high risk associated with speaking openly about political issues in some countries incurred a consequent and unanticipated degree of ethical discomfort for some respondents. Sixty percent of all respondents expressed various ethical concerns while conducting field research. The vast majority of researchers felt confident that they were fulfilling their Institutional Research Board guidelines and requirements. One respondent noted concern that interviewees who were informed of the research project and goals and who agreed to being quoted did not understand their own risks of being identified. Another expressed concern that the researcher's reliance on personal contacts to arrange interviews indirectly meant that anonymity is compromised by the fact that interviewees either know or can trace each other. With the very real possibility that researchers are under surveillance by internal security, there are risks that interviewees may also come under surveillance, be subject to questioning or, given the political sensitivity of the topic, arrest. In both cases, the respondents were confident that they had minimized these risks.

Many researchers simply felt uncomfortable with what Sheila Carapico calls in her article in this symposium the “fly on the wall” approach to research. The most common ethical discomfort that researchers experienced was related to interviewees expecting or requesting services or favors—many of which were political in nature—in return for participating in the interview or for any friendship that had developed through the course of field work. Eighteen percent of the expressed concerns were of this kind. In one of the more complex cases, a researcher was asked to become involved in a human rights case—while the breech was egregious, the researcher's involvement would have been time-consuming and potentially biased his/her research. In other cases, researchers were asked to help interviewees obtain medical attention or immigration/employment in their home countries or for monetary assistance. While these latter cases are easier to decline, all researchers felt the uneasy debt of interviewees devoting time to the respondents' research. Respondents questioned the degree to which they should get involved and to what extent. At what point would it compromise their research?

Twelve percent of the concerns revealed researchers' discomfort with the fact that they disagreed with their interviewees' expressed opinions (and even found them repugnant), yet the probes they used to elicit information, ranging from silence to echoing to nodding, appeared to imply agreement. In some cases, researchers felt they gained the trust of the interviewee due to this perceived confluence of views. One researcher was even complimented for his/her “open-mindedness.” This applied to all researchers, but was largely experienced by those doing interviews with government officials and opposition party and movement members. This is certainly not unique to the Middle East, except, potentially, in terms of degree. Maintaining a professional distance and the requirements of fieldwork require that researchers listen to their interviewees. To have objected to the interviewees' opinions would have risked the curtailment of the interview or the self-censorship of the interviewee. While researchers understood the necessity of silence, they felt uncomfortable with the hypocrisy of the situation and what they perceived as the moral betrayal of their views.

Related to the two above-mentioned issues, 8% of the stated ethical concerns also dealt with interviewees' perceptions that the researcher had some form of influence over their home country's government and/or the researcher would defend the interviewee's views in the West. Researchers struggled with the fact that they were viewed as representatives of their government, nationality, and/or gender.

While researchers were confident that they protected the anonymity and confidentiality of their interviewees—who in most cases the researcher only visited a limited number of times—they were less sure about their friends. Eight percent of the ethical concerns dealt with the researcher's presence potentially putting his/her friends at some degree of risk. In one case, the researcher's friends were contacted and questioned by internal security, although they suffered no consequences (the researcher did not learn of these events until after the completion of the field work). Another researcher left the field when he/she felt this situation developing.

As Carapico discusses in her article, many researchers are beset with dilemmas concerning what to publish and, most importantly, who will be reading and to which uses they may put the data. These concerns included both the local authoritarian regimes but also their home governments. For which causes were they providing political fodder? For example, how detailed and critical does one want to be with a burgeoning civil society? Despite the weaknesses in civil society that the researcher uncovers, does one want to contribute indirectly to its suppression by an authoritarian regime?

Other ethical issues less directly related to the political situation and more familiar to researchers everywhere concerned researchers' uneasiness in publishing conclusions that are either unflattering to the interviewees or contradict the interviewees' opinions. Fourteen percent of the ethical concerns were related to interviewees' reactions to the published results of the fieldwork. Researchers noted that they had experienced interviewees feeling betrayed, offended, and angry with research results. This was particularly so when the researcher had developed friendships with interviewees as a result of repeated contact and the researcher's integration into interviewees' social networks as a result of snowballing or the nature of the research topic itself.

Finally, while researchers clearly expressed that the friendships they developed in the field were an important, rewarding, and enriching part of their field work experiences, 12% of the ethical concerns registered in the survey noted that researchers also struggle with maintaining a healthy degree of distance. This arises from researchers' realization that they are only temporarily in the field and will not have to pay the consequences for their actions (or inter-actions). Professionally, it also stems from the need to ensure that an interviewee/friend is aware of the fact that the researcher wears two hats—one of friend and one of researcher. Issues of confidentiality (and betrayal), as Carapico states, while not insurmountable, become more complicated when interviewees turn into friends.

Field Research and Gender

These results indicate that more important than any cultural differences, the political climate has the greatest impact upon field work in the region. Culture manifests itself most acutely in issues related to gender. However, the survey results, in line with field research elsewhere, confirm that gender is one of many variables that can influence field work and, more importantly, it may or may not be advantageous (Barnett and Cason 1997, 88). Female researchers were relatively equally divided as to the role gender plays in their field research—if at all. Fourteen percent unequivocally stated that gender was an asset and 17% stated that gender was a mixed bag—that it could work to your advantage or disadvantage. Significantly, 31% of female respondents did not mention gender at all—in other words, it is an issue of little or no import.

Most female researchers found that being female was helpful when studying issues related specifically to women, gender, and/or required interviews with women. As many countries in the Middle East are socially segregated according to sex, men are largely barred from women's realms. Indeed, one male respondent (the only one to refer to gender) noted his difficulties in obtaining interviews with women. Yet women found that even in the most conservative of states, they were not barred from the male realm. As Jillian Schwedler discusses in her article in this symposium, Western women enjoy the advantages of being a “third gender.” Other women stated that women researchers are often seen as less threatening and, as a consequence, interviewees are more likely to let their guard down.

Thirty-eight percent of the female respondents stated that they confronted difficulties in conducting field work as a result of their sex and local gender norms. Only 7%, however, noted that they had less access to male interviewees as a result of being female. One respondent remarked that she was not taken seriously as a young female researcher. Most of the respondents stated that these gender-related problems were more-or-less confined to a degree of sexual harassment on the streets or having to conform to local gender norms socially—and less to do with the actual interviews or field work per se. Indeed, two of the above-mentioned ethical discomforts that researchers reported were a result of feeling pressure to conform to local gender norms: that female researchers were not always entirely truthful about their personal and social lives and that they altered their clothing styles for specific interviews while reverting back to their personal styles when the interview was over. In both cases, these researchers felt a degree of hypocrisy. It is important to note that many of the female researchers' gender-related challenges were also attributed to the age of the woman and the country in which the research was being conducted.

Interview Techniques

In terms of interview techniques, such as whether or not to use a tape recorder or a translator, respondents had similar concerns to those well documented by researchers elsewhere; however, these are heightened in the context of the prevailing authoritarianism. Researchers were cognizant of potentially adding to the mistrust and suspicion of interviews. One researcher noted that some interviewees went so far as to announce that they would only restate the “official line” should a tape recorder be used. As one researcher put it, “the tape recorder instantaneously transforms an informal and perhaps more informative interview into a formal platform for political statements.” While researchers everywhere take this risk, it becomes more pronounced in authoritarian regimes where interviewees fear both local and often foreign governments' use of and response to the individual answers and the congregate data.

Despite these concerns, 56% of respondents (including those with strong foreign language skills) have used or use tape recorders during interviews. Of these respondents, 19% (11% of the entire sample) regularly record their interviews. Researchers noted that tape recording interviews was often reassuring to interviewees (indeed, some requested the interviews be taped) in that they were more confident they would not be misquoted. This is particularly the case for government and party officials. Other respondents stated that interviewees are often flattered when the interviewer regards them as significant enough to record. Still others noted that the interviewee's self-consciousness as a result of being taped can often work to the interviewer's benefit—particularly when the interviewees are government officials or party representatives—as interviewees can be more deliberate and exact in their responses. Advocates of tape recording interviews strongly stated, however, that the most successful use of tape recorders occurs when researchers already have the trust of the interviewee or the associated political risks of doing an interview under an authoritarian regime have otherwise been mitigated.

Similarly, while respondents were concerned about the impact of a translator upon the frankness of the interviewee, the dynamic flow of the interview, and the potential bias inserted by the translator (also of concern to researchers elsewhere), 42% of respondents have used or use translators during interviews; of these, 22% (9% of the entire sample) employ them regularly. In one case, the potential pitfalls of using a translator were exacerbated by the fact that the translator did not dress appropriately and, as a result, affected the quality of the entire interview. However, respondents also noted the greater ease with which translators could secure interviews, the translator's knowledge of potential interviewees, and the fact that the translator was often more trusted by the interviewee than the researcher him/herself. Others noted that they could concentrate on the interviewee's responses more closely when a translator was doing half the job and that they benefited from post-interview discussions with the translator. As with tape recorders, respondents noted that the success of using a translator depends on the relationship between the translator and the researcher, and whether the interviewee knows and/or trusts the translator.

Graduate Training and Preparation for the Field

Finally, in terms of the preparation or training researchers received for field work during their graduate degrees, researchers' responses reflect larger trends in political science where only 20 out of the United States' top 30 departments in political science (66%) offer qualitative methods courses to their graduate students and only two departments (6.6%) require courses in qualitative methods with eight more offering them as optional (33% in total) (Bennett, Barth, and Rutherford 2003, 377). Only 25% of the respondents unequivocally stated that they had been well-trained by their political science or related departments in qualitative methods during graduate school.⁵

While 11% of the respondents questioned whether field research methods can be taught in the classroom, the survey indicates that there is much training that can be done. In addition to basic methods and techniques and how to adapt them to the peculiarities of the field, the survey results indicate a need to prepare researchers for the “gray zone”—those ethical discomforts that regularly arise as part of interaction between human subjects and the demands of research.