Contentious issues in research on trafficked women working in the sex industry: study design, ethics, and methodology.

Trafficking of women and children for work outside their countries of origin in an increasingly globalized sex industry is a significant issue for public health professionals, international law enforcement and human rights agencies, international labor monitors, and groups concerned with women's and children's welfare (Coalition Against Trafficking Woman [CATW], 2003; Human Rights Watch, 2002; Levenkron & Dahan, 2003; Vanderberg, 1997; Watts & Zimmerman, 2002; Zimmerman & Watts, 2003). There are huge profits gained from women trafficked for sex work (WTSW): the turnover, estimated at between $7-10 billion a year, is seen as the best cost/risk-benefit ratio of all criminal activity (Levenkron & Dahan; USAID Office of Women in Development, 1999; U.S. Department of State, 2003). However, the revenues from trafficking reported are "guestimates," because they are based on estimates of the number of transactions between WTSW, clients, and traffickers. Furthermore, Interpol calls trafficking the fastest-growing crime category today (Sulavik, 2003). The profit from trafficked women is vast compared with the $54 million over two years that the U.S. government invested worldwide to try to stop trafficking (U.S. Department of State).

Moving women between countries for the purposes of work in prostitution dates back to Roman and Biblical times and was a major concern among social reformers of the late 19th century who fought against the "White slave trade." However, the nature of contemporary trafficking enterprises has changed both in volume and method. Growth of the internet has provided new methods of recruiting, procuring, and supporting this clandestine movement of people and expanding the demand for exotic or foreign women for sexual services (von Struensee, 2000). Furthermore, 21st century paraphernalia such as the internet and cell phones facilitate communication and organization between source and destination along the trafficking routes. Like fast food, name-brand soft drinks, and sporting goods, "fast sex" has taken full advantage of the nature of our cyber-world to market women's and children's sexual services by generating a supply of women, generally from economically disadvantaged countries, who work illegally in foreign countries to meet this demand. Thus, the former "White slave trade" today has a wide variety of trafficking routes from diverse source countries and to many destinations, and with various modes of transportation (plane, boat, foot, etc.).

The scope of human trafficking is hard to measure, but it is estimated that from 700,000 to 2 million women (United Nations [UN], 2000), with some estimates as high as 4 million women and children, are trafficked across borders to work in the sex industry each year (Estes & Weiner, 2001; Raymond, 2001; U.S. Department of State, 2002). Estimating the numbers of WTSW is difficult since not all of those trafficked for prostitution were recruited for this occupation. Although most of the women and children are recruited for work in prostitution, sex tourism, or the mail-order-bride business (Watts & Zimmerman, 2002), many are trafficked to work in the garment industry, to join family members, or to work in domestic services; but they may find themselves pressured to provide sexual services as part of their duties (Richard, 1999). At the destination, some women are duped into sex work, and others voluntarily leave low-paying, dead-end jobs for the lure of higher-paying opportunities in prostitution.

Despite the disparities of these estimates, no data source gives details on how these "guestimates" were derived (Kelly, 2002). Accurate estimates are difficult to obtain because the movement of people occurs almost completely in secret. For each person who comes to the attention of border police, immigration, or health or welfare services, there are probably 10 to 20 persons who do not; thus, they remain an invisible labor force. WTSW lack citizens' rights, often do not have passports, and are wary of authorities for fear of deportation (Feingold, 2003). Even when trafficked women are identified, often brothel owners will move them between brothels in order to avoid police intervention and to meet the demand for new and exotic women among the other prostitution outlets under their control. Thus, the determination of the numbers of women and children who are trafficked depends on the definition of trafficking that is used by agencies; problematic access to a mobile population; the data source; and which types of trafficked women are included in the estimates (i.e., only those who know before being trafficked that they are going to work in prostitution, or all illegal residents who work in sex work in destination countries).

The UN Trafficking Protocol defines a trafficked person as someone who is transferred or transported across national and international borders, by means of threat or coercion, for the purposes of economic exploitation in prostitution, forced labor, slavery, or the removal of organs (UN, 2000). Whether or not the trafficked person consented to the transfer is irrelevant if there was any pressure, financial or otherwise, applied to the person or her guardian. The transfer of a child under 18 years for economic exploitation is considered trafficking even in the absence of coercion. All women transported across borders to provide sexual services share experiences such as falling victim to deception and economic indebtedness during the trafficking process. Deceptive tactics may range from outright abduction and kidnapping to enticement through the representation of apparently favorable conditions to deceptive offers of marriage or legitimate employment. It is common for women to enter a "work contract" of six months or a year, and then work without pay for several months or never receive any remuneration if they are resold to new owners or bosses. These contracts are usually verbal accords and often do not articulate the type of work or conditions the women will find at destination. Often the recruiter is someone known or trusted, such as a neighbor, friend, or family member (Raymond, 2001). Although the majority of Soviet women in a study conducted by one of the authors reported knowing in advance that they were going to work in the sex industry, most of them were unaware of the conditions, obligations, and work demands of their prospective employers (Chudakov, Ilan, Belmaker, & Cwikel, 2002).

Even when trafficked women are identified, few request help or cooperate with humanitarian aid or welfare agencies. For example, in 2001 and 2002, a UN Mission on Special Trafficking Operations Program (STOP) conducted 720 "raids" in Bosnia and Herzegovina, where they interviewed 2,120 women and girls working in clubs. They reported that 25% (530) of interviewees were trafficked, but only 230 requested assistance from the STOP mission. However, non-government organization (NGO) experts working with the UN mission concluded that the statistics collected on these women were "woefully unreliable" (Human Rights Watch, 2002).

A study of trafficked women conducted in five countries--Indonesia, the Philippines, Thailand, Venezuela, and the United States--found nine factors responsible for the worldwide increase in human trafficking (Raymond, 2001):

* Under economic policies of globalization, many services that used to be state-supported, such as education, health care, and social welfare, are now being transferred to private hands, increasing the economic burden on families who must pay for these services. This is particularly true in the countries of the Former Soviet Bloc, where basic health and educational services were provided under Soviet rule, but now need to be financed out-of-pocket in free-market economies. Often, the women and children are sent abroad to earn foreign currency to pay for these essential services. For example, women we interviewed reported entering a trafficking contract to finance expensive health care for family members or their own university education, items that had once been provided by the state (Cwikel, Chudakov, Paikin, Agmon, & Belmaker, 2004).

* The sex industry is becoming more globalized, with recruitment and transport being conducted in larger and more sophisticated trafficking networks. Sex industry advertising is accomplished over the internet, offering further opportunities to provide international sex business (Hughes, 2000, 2001, 2004; Jeffreys, 2002; von Struensee, 2000).

* The male demand for sex services is a hard market to saturate, suggesting that "the way in which sex has been tolerated as a male right in a commodity culture is all part of this demand" (Raymond, 2001, p. 2; Batros, 2004; Yea, 2004).

* The social structure in most of the world is built on women's inequality and economic dependence on fathers and husbands and male relatives. This inequality has allowed an almost endless supply of women who are desperate to earn money, particularly in developing countries and emerging industrialized countries such as the former Soviet Union.

* The commodification of women's bodies as sexual objects, and therefore for sale, is common (Long, 2004).

* Child sexual abuse, in particular, puts young women in a vulnerable state that may be exploited in order to pressure women to work in prostitution (United Nations Children's Fund [UNICEF] 2003; Widom & Kuhns, 1996).

* The stereotype that "the exotic is the erotic" has fueled the demand for foreign women to enter prostitution, further inflating the demand for trafficked women (Batros, 2004). This has been a traditional marketing angle in the sex industry, dating back to Roman times when the hetaerae, or foreign women, commanded the highest prices for sexual services. Today, there is an even broader selection of source countries for recruitment.

* War or a military conflict has fueled the demand for women to be brought to places of conflict so they can provide sexual services for troops. Where a permanent military presence is established, there are always brothels and prostitutes in the vicinity and places for the troops to rest, relax, and be entertained (Mirkinson, 1997).

* Restrictive immigration policies do not offer working opportunities with legitimate travel documents for those who want to work in non-professional jobs. (Harcourt, 2004; Raymond, 2001).

Countries such as Australia, New Zealand, the United States, and European Union member nations have tried to formulate international and national policy responses to the burgeoning movement of people across borders (Galiana, 2000; Jeffreys, 2003; Kantola & Squires, 2002; Maxim Institute, 2003; Outshoorn, 2002). In 2000, the UN Trafficking Protocol focused on preventing and combating trafficking and developing ways to deal with the problem at the national and international levels by providing housing, training, legal aid, information, and psychological and medical aid to WTSW (UN, 2000).

The definition of WTSW used in this article is consistent with the UN Trafficking Protocol definition. It acknowledges that when deception, threats, violence, and fraud are used, even in cases where women have consented to entering a work contract, the abrogation of their human rights constitutes trafficking. In addition, it includes indentured labor, forcing women to repay debt through sex work, and withholding information about work conditions, wages, medical care, or legal documents. Policy debates often get bogged down attempting to achieve consensus of the definition and absolute conditions for trafficking status. Unfortunately, this can result in stagnation and inaction when proactive strategies are urgently needed (Kelly, 2002). We consider the minimal conditions for trafficking to be illegal working status when a third party has been involved either in transport or setting of work conditions in any branch of the sex industry (including a range of venues and sexual services; e.g., bars, dancing clubs, hotels, brothels, street work, and work in pornography). In accordance with the UN Trafficking Protocol, these conditions obviate the question of whether the person entered the trafficking condition voluntarily or involuntarily.

Zimmerman and colleagues (2003), in their European Union study of the health effects of trafficked women, outlined the stages in the trafficking journey--from pre-departure, transit, working in the destination country, possible deportation to the country of origin, and re-integration. Not all women follow these five stages, since some may become illegal sex workers without ever being trafficked (e.g., visit as a tourist and then decide to work in the sex industry in order to stay in the destination country) or come as a WTSW but attain legal residence in the destination country. Furthermore, any research project is likely to interview women at only one or two of the five stages. The bulk of our data were drawn from observations and interviews with women at the working or deportation stages.

Zimmerman and colleagues formulated the World Health Organization (WHO) ethical and safety guidelines to sensitize those working and researching with trafficked women, particularly with regard to the safety risks that may threaten a trafficked woman if she participates in an interview with authorities or researchers (Zimmerman & Watts, 2003; Zimmerman et al., 2003). Most of the principles outlined in the WHO guidelines are recommended in all research settings (e.g., ensure anonymity, prepare referrals in case of distress, obtain informed consent, use research materials to promote the health and safety of the subjects). However, some principles recommended by Zimmerman et al. are specific to working with trafficked women, including (a) knowing the risks associated with trafficking, (b) adequately preparing interpreters, interviewers, and researchers for secondary traumatization that may occur following exposure to stories of abuse and human rights violations, and (c) avoiding re-traumatizing women with invasive questions about traumatic events (Zimmerman & Watts).

In this article, we outline a series of methodological and ethical issues to be considered when conducting research in the area of human trafficking. We draw on our experiences in the field while collecting data from WTSW and traffickers and on our interactions with immigration and law enforcement officials in Australia and Israel (Cwikel, Ilan, & Chudakov, 2003; Hoban, Gordan, & Maltzahn, 2003). We believe that most current data are compromised by methodological problems and that an open discussion of these issues will generate better data in the future. First we present issues that should be considered in the design of research on WTSW; then we outline a series of ethical and methodological problems in the conduct of such research. With the presentation of each issue, we present our suggestions or solutions.

STUDY DESIGN ISSUES

Sources of Research Funding

We encourage the development of coalitions to fund and support research that addresses the problems of WTSW. Funding for research on WTSW on a country-by-country or regional basis is problematic because many governments do not consider trafficking a priority area for action or, consequently, for funding (Galiana, 2000). Often, governments are unwilling to fund research that reaches out into the world of trafficked women in the sex industry and establishes non-threatening contacts with women who are frightened of authorities and what they represent (i.e., law enforcement, deportation, loss of income and earning status). Herein lies the problem: without dedicated, multi-focus outreach programs that provide health and support services for WTSW and research that is anonymous, confidential, and impartial, we will never be able to draw adequate samples with reliable or proxy indicators that can give data on which to base policy decisions. The organizations that are best situated to access and conduct research with WTSW are pro-women's rights or anti-violence groups, but funding sources may not see them as appropriate partners for research. However, without the data to show the number of WTSW and the impact of interventions, it is almost impossible to argue for further research funding.

Furthermore, funding obligations that exclude "illegal non-citizens" often block government programs that might encompass applied research that includes WTSW. Thus, opportunities for documenting the health and welfare of WTSW are missed. For example, Reproductive Health and Education in the Sex Industry (RHED) in Victoria, Australia, which provides health education, counseling, and support for women working in the sex industry, receives considerable funding from Department of Health Services, Victoria, but does not target or encourage access by women who work illegally in the sex industry (Hoban et al., 2003). Also, the agencies that administer programs for this population often fail to establish an applied research component. Such in-house applied research programs, which are inexpensive to administer, might provide opportunities to collect valid data from primary sources that would allow these organizations to evaluate the impact of their activities, identify emerging public health, immigration, and welfare problems, and monitor the enforcement of sex industry and trafficking legislation (Dean, 2004).

Notable international reports include the report drafted by International Organization for Migration (IOM) on trafficking in Southern Africa (Martens, Pieczkowski, & van Vuuren-Smyth, 2003; UNICEF, 2003) and the European study on health issues among WTSW (Zimmerman et al., 2003). However, the numbers of trafficked women included in these studies are small for any location or country (Martens et al., 2003; Zimmerman et al.). The development of better research partnerships and collaborations might have allowed these researchers to access a larger and more representative sample of WTSW.

Development of a Critical Research Stance

Most researchers enter a WTSW study with preconceived notions about the type of people they will encounter. Researchers may be unaware that these ideas have been fueled by sensationalist media coverage of exceptional cases of murder or abuse, as well as the framing of WTSW as victims. Although there are cases of horrendous abuse, including the murder of WTSW (Galiana, 2000; Levenkron & Dahan, 2003; Zimmerman et al., 2003), this stereotype may not fit the profile of many women who have been trafficked and are subject to more mundane pressure, control, and veiled threats (Kelly, 2003). As Zimmerman and Watts (2003) stated, "It should not be assumed, however, that all women who have been trafficked are traumatized, consider themselves victims, detest their captors, or wish to escape or go home" (p. 3). Furthermore, we concur with Kelly's stance that the difference between being forced or voluntarily entering the trafficking world is irrelevant when considered against the weight of the human rights and worker's rights violations that occur among these women.

Many researchers may feel pressure from colleagues and activists to declare "whose side" they take in relation to prostitution and trafficking. Many positions and approaches relating to discourse on trafficking exist, including moralist, crime and border control, labor and occupational, public health, migration, human rights, and feminist (Global Alliance against Trafficking in Women [GAATW], 2001; Kelly, 2002; Levenkron & Dahan, 2003).

Many of these positions have become polarized, making it more difficult to engage parties in public discussions about WTSW. For example, prostitution abolitionists such as the Coalition Against Trafficking in Women (CATW), the most influential anti-trafficking organization, equate human trafficking to "forced prostitution" and consider it a form of violence against women and "sexual exploitation." In contrast, GAATW, another feminist organization influenced by the sex workers' rights movement, makes a distinction between "trafficking in women" or "forced prostitution" and "voluntary prostitution" (CATW, 2003; Doezema, 2000). A third position comes from sex workers' rights groups, who differentiate between women forced to work as prostitutes by a third party (i.e., a pimp), particularly where violence or deceit is used, and women forced or pressured to make that decision based on economic reality (Alexander, 1998).

These disparate and highly politicized positions on prostitution and human trafficking make it difficult for researchers to gather data and engage parties in discussions about WTSW. Sometimes the researcher's position is assumed a priori by the funding source, but regardless, any stance will limit the ability of the researcher to explore and collect data using other paradigms. Our recommendation as researchers is to avoid alignment with any organization in order to retain neutrality and critical awareness in the discourse and the research design. It may seem easier, cheaper, and more convenient to access WTSW through established movements such as GAATW or sex workers' organizations, but researchers need to avoid the politics of all groups and seek out the quiet or silenced voices.

Use of Multiple Research Paradigms and Methods

As in other research, the approach or research paradigm of WTSW studies will influence findings. If the paradigm guiding the work stems from trauma theory, human rights, migration, sexology, or legal aid, this will influence the type of questions asked and how the respondents will perceive them. For example, several of the major reports on human trafficking were written by female lawyers who represent the legal view and their connections with police, crimes, courts, witness protection, and human rights groups (Galiana, 2000; Levenkron & Dahan, 2003; von Struensee, 2000). On the other hand, some feminist scholars have conducted clinical work with victims of rape, domestic violence, incest, and abuse and have drawn on these frameworks in collecting and analyzing data (Kelly, 2002; Raymond, 2001; Watts & Zimmerman, 2002; Zimmerman et al., 2003). If researchers expect that the experiences of WTSW include abuse, torture, and trauma, their findings may be biased. However, if the goal is a more balanced and objective picture, this will be reflected in the design, instruments, and methodology selected. Thus, we recommend using multiple theoretical perspectives in developing the research design.

Capturing an accurate and comprehensive picture of the ways in which trafficked women enter the sex industry requires the understanding that in addition to being recruited in their home countries, women sometimes choose to engage in sex work after illegal entry into a country or when they are in breach of their visa conditions. These women may have a variety of motivations, personal and family histories, and personalities. This finding reflects work done by other researchers on women in the sex industry in diverse settings, such as in Uganda (Gysels, 2002), the republics of the former Soviet Union (Aral & St. Lawrence, 2002; Aral et al., 2003), and Utrecht and Bangkok (Wijers & Lap-Chew, 1997).

We recommend using creative research and investigative methods and accessing unlikely key informants, including brothel workers, managers, and traffickers or their representatives. One of the authors utilized a multi-method approach and incorporated research instruments such as historical timelines conducted during interviews with the WTSW and her family and friends (when possible), multiple data sources including information from family and friends, brothel management and workers, and, when possible, accessed official migration documents under the Freedom of Information Act. This process allowed the researcher to develop a portrait of the trafficking experience and created an opportunity to increase validity and reliability of the data.

Identification of the Uses for Data

Researchers need to consider the intended use of the data that is being collected and to be wary of a hidden agenda (e.g., of law enforcement, labor courts, women's rights organizations, pro-prostitution, anti-prostitution, or abolitionist groups) for gathering the information. An honest and ethical approach to WTSW requires researchers to state the purpose of their study and what they intend to do with the data. WTSW, like most vulnerable populations, are wary of the way information may be used for fear they could be identified and deported. This leads to the use of pseudonyms in order to protect their own and their families' identities. Women's vulnerability may also lead to situations where women are afraid to provide information about their trafficking and migration history. Alternatively, they may provide researchers and authorities with false statements to protect themselves, their families, and those who "have been kind to them"; the latter may include traffickers and agents in the source country (Hoban, 2005). Women awaiting deportation are aware that governments are interested in assisting women who will testify against their traffickers, yet they are often unwilling to put themselves or others at risk, and therefore forgo preferential treatment, including access to sheltered housing, medical attention, psychological counseling, or special visas for trafficked women.

Therefore, we suggest stating the purpose of the study in plain language and providing a clear, frank statement of why the study, and WTSW's input, is necessary. This sounds like standard practice. However, to recruit WTSW into studies, researchers often feel pressure to give informants extra information in order to gain their trust. WTSW may be reluctant to be interviewed unless the researchers can show what the women can gain, who will read and take notice of the information, and which government agencies, welfare and support groups, and feminist organizations are involved. Researchers need to show impartiality by providing information about the organization conducting the research and their interest in the research subject. The researcher should not disclose any personal affiliations with feminist, welfare, or human rights groups if they are not relevant to the study.

Protection of Respondent Identity

We recommend the creation of mechanisms at the design stage to protect the identities of informants during interviews or observations. In general, we do not suggest recording interviews by tape or video since this compromises the interviewee's identity. In lieu of electronic recording, we recommend conducting interviews using two interviewers, one conducting the interview and the other recording information. However, in some situations, taped interviews with informed consent procedures may be appropriate. If so, researchers should provide the WTSW with transcripts of the interviews for feedback and editing as soon as possible after the encounter. This practice will improve the reliability of the data, assure interviewees that their identities are protected, empower WTSW to give voice to their experiences, and develop trust between WTSW and researchers.

Despite the difficulties associated with in-depth interviews of women with illegal status, we found that many women were grateful for the opportunity to tell their stories to concerned, neutral listeners. After being trafficked as purely sexual commodities, many women found interviews with researchers cathartic and helpful for gaining perspective. As researchers, we utilized the contacts with women to inform them of services that they could access to help them with health and legal issues.

ETHICAL ISSUES

There are four basic moral principles in public decision-making, whether it be in medical practice or in research: (a) respect for an individual's capacity for autonomy and independent decision-making; (b) nonmaleficence, or the Hippocratic dictum in medicine of "above all, do no harm" (primum non nocere); (c) beneficence, or the pursuit of benefit from actions in balance between risks and costs; and (d) pursuit of distributive justice that ensures the greatest benefit will accrue to the largest number of persons (Beauchamp & Childress, 2001). The perilous and tentative legal status of most trafficked women puts these ethical principles to the test. Below we provide specific examples of these problems with suggestions on how to resolve them.

Complications in Informed Consent Procedures

Informed consent implies that the person is able and capable of giving consent, consent is voluntary, and the full information as to the benefits, risks, and uses of the research are provided. Our experience has shown that informed consent procedures for WTSW require modifications for a number of reasons: (a) women are reluctant to sign documents and thus verbal informed consent may need to be the standard procedure, (b) generally women do not use their real names for fear of reprisals or being deported, so they may use their work pseudonyms instead, and (c) most women are unfamiliar with concepts like informed consent and research protocols and they may not fully understand the process that is explained to them, even if they are being interviewed in their native language. Thus, the ability to match any interview to a particular woman who is in need of medical, legal, or psychological advice may be seriously compromised.

We suggest either having women sign informed consent procedures using their pseudonyms or allowing for verbal, not written, informed consent. Furthermore, it is necessary to explain in their language why the project is being conducted, for what the data will be used, and who will have access to it. Also helpful is leaving the respondents with access information such as the name and phone numbers of researchers and NGOs that have counselors who speak their language. In order to match interviews for follow-up, we suggest keeping password-protected double lists of their real names and dates of birth together with the pseudonyms recorded on the questionnaire or the questionnaire number. We used this procedure in a substudy of sexually transmitted infections among trafficked sex workers (Cwikel, Latzer, Latzer, & Press, 2005). We explained to the women that we needed to re-contact them in order to supply them with medical test results. Other strategies to ensure data confidentiality are also pertinent: coding the names and access phone numbers, training all interviewers and data entry personnel on the importance of the confidentiality of the data, and allowing only authorized research personnel to access the data once collected. If tape recorders are used for interviews, researchers should delete interviews once they have been transcribed.

Problematic Access to WTSW

Access to women during the various stages of trafficking is severely limited for researchers. To the best of our knowledge, no data has been collected on women before they enter into the trafficking process in their countries of origin. Since trafficking involves the illegal migration of women, access may be through a third party, such as the traffickers, brothel owners, agents, madams, or voluntary workers such as the chaplain or human rights advocates. Anonymous sexual health clinics or agencies that provide services to women in the sex industry are in an ideal position to collect reliable data, but they rarely use this opportunity to do so. Women may be interviewed once they have left the sex industry or are detained awaiting deportation, while they are being primed to testify against traffickers, or during trials. Access to women in the latter circumstances is problematic since contact may be possible only through prisons, immigration authorities, or lawyers.

We suggest trying to gain access to women in the most representative settings with a clear understanding of the limits and biases of the sample. In other words, if women are accessed through the traffickers or brothel owners, it is necessary both to develop a sense of reciprocal utility and trust and to acknowledge that such informants may have an alternative motive for assisting the researcher, such as gaining legitimacy with legal authorities. We understood that some brothel owners were interested in widening access to health care for the women working for them, and thus, in some cases we provided medical examinations and referrals to health care when needed. At the same time, we understood that the brothel owners might be screening the women that they were allowing us to access, and we tried to cross-validate our findings using alternative samples (e.g., women in detention or women waiting to testify).

Loss of Respondents for Follow-Up

Aside from the ethical problems of loss to follow-up in the short-term, lack of real identification, such as the absence of personal identification documents, hampers researchers' understanding of the range of women's experiences. A comprehensive view of the cycle of trafficking is possible only if women are followed over the various stages of the trafficking cycle, possibly using in-depth ethnographic techniques, such as social and network mapping and life histories (Long, 2004). Women are understandably reluctant to be identified as formerly trafficked people or sex workers if they have migrated to other areas or have reintegrated into their home countries following completion of their contracts or deportation. However, we will be able to understand the full impact of their trafficking experiences only once such longitudinal research has been conducted.

Recent initiatives in source countries have led to the establishment of projects to reach out to women who have been trafficked and returned to their home countries or are at risk of trafficking. For example, the Angel Coalition, a consortium of 43 Russian and Former Soviet Union NGOs that was founded in 1999, has been active in helping women reintegrate into their home countries, providing them with education, vocational retraining, and loans in a micro-enterprise lending program (The Angel Coalition, 2003). Other regional projects include the Kvinnoforum (Women's Forum) in Stockholm, Sweden, which has been active in training and creating networks in Scandinavia (Kvinnoforum, 1999). Similar networks have been formed by GAATW from Bangkok, Thailand, and across the Asian-Pacific region (GAATW, 2001). We suggest that researchers conducting longitudinal and follow-up research can collaborate with these NGOs in source, transit, and receiving countries.

Inability to Intervene in Illegal Acts

Due to the sometimes precarious situations in which trafficked women are interviewed, researchers may find themselves present during illegal events. The perpetrators of these crimes may be brothel owners or law enforcement officials, immigration officers, or other government authorities. Researchers sometimes witness violent or illegal acts or find themselves in illegal brothels during police raids, but choose not to intervene (Bailey, 2002). On these occasions, researchers may prefer to maintain their autonomy and independence and choose to be an observer. In addition, interviewing women who have clearly been trafficked but are still in the sex industry presents a dilemma, because women's trafficked state means that someone has facilitated their entry into the country and the women are working without legal residence. In some countries, such as in Australia and Israel, the women, their traffickers, and brothel operators are in breach of law and all parties are subject to arrest.

When researchers are collecting data with the consent of brothel owners on their premises, there is no recourse other than to record the woman's status without intervening. Thus, we can offer no workable suggestions to this conundrum other than for researchers to develop awareness of the limits of their ability to intervene and to act accordingly. Any intervention may result in the deportation of a trafficked woman who is working illegally; on such occasions, the researchers may do more harm than good. Subjective reports from brothel-based WTSW and women awaiting deportation suggest that the psychological state as measured by depression scales was far worse for women awaiting deportation than for women who were working (Cwikel et al., 2004); thus, incarceration and imminent deportation may be more threatening and psychologically disturbing than the most severe working conditions.

Although it is rare for researchers to witness physical violence, we often heard stories of physical, sexual, and emotional abuse. We also observed women suffering from the shock and distress associated with finding themselves in trafficked situations beyond their control. As researchers, we noticed women who looked under 18 years old, but reported that they were over age 18. Therefore, sex work by minors was not directly observed in a way that could be validated. In situations where researchers note the appearance of abuse, such as visible injuries to the body, with the woman's consent it may be possible to alert the brothel proprietors to the woman's situation and provide first aid themselves, or if the condition is more serious, offer to seek the assistance of confidential health services for sex workers.

Researchers who venture into brothels may observe unsafe working environments. For example, we observed women who were providing sexual services despite reporting severe pain and symptoms of reproductive tract infections that required treatment (Cwikel et al., 2003; Hoban et al., 2003). In one case, a group of WTSW who had been interviewed at one brothel by researchers in Israel insisted that the brothel owner give them direct access to health care for their symptoms.

WTSW often have limited rights to refuse clients who are violent, demand unprotected sex, or put their safety at risk. For example, licensed brothels in Victoria, Australia, are required by law to provide safety measures such as emergency call buzzers in the rooms. However, these occupational health and safety standards do not always exist in brothels where trafficked women provide sexual services. Women who provide sexual services outside brothels are at particular risk of violence when they are either working or being transported. Again, the researchers may be in a position to facilitate women's access to confidential health services for treatment of illnesses. However, they are rarely able to intervene in situations outside the research domain, such as promoting condom use or reporting occupational health and safety violations. To do so could place the researcher in an unsafe position.

Preservation of Research Credibility

We emphasize that researchers should make no promises as to how the information will be used, outside what has been stated in the informed consent procedures. This is important because WTSW and their supporters often see research as a means of pursuing their own agenda, and this may lead researchers into situations outside their realm of expertise or in violation of ethical research practices. Key documents and interviews with WTSW should use the woman's first language to ensure accuracy of data and context of the experiences. If the researcher does not speak the WTSW's native language, it is essential that a trained translator is used. Poor interpreting and translation of official documents will lead to false information that may be detrimental to the woman's welfare in the future (Hoban, 2005). Both being consistent about how data will be used and allowing WTSW to understand interviews and documents in their own language will help researchers to maintain credibility.

METHODOLOGICAL ISSUES

Constructing Representative Samples

Due to the hidden and illicit nature of the migration of people for sex work, many researchers have noted the difficulties associated with accessing WTSW for their studies (Galiana, 2000; Kelly, 2002; Raymond, 2001). However, studies of WTSW that rely on one data source may be biased. There will never be an accurate census of trafficked persons, because of the ramifications of their illegal status: they are mobile, may have illegal papers or lack documents such as national identity cards or social security numbers, and will not answer phones or mail. In general, it is safer for them to keep a low profile. However, using multiple data sources and methods to triangulate data, it is feasible to "guestimate" the number of trafficked women in a locale. Keeping in mind that any source probably represents only a fraction of the actual population, we suggest the following methods:

* Collect data using government statistics (e.g., immigration data such as refugee and migration review tribunal applications, airport, harbour, and cross-border statistics, immigration detention centre and deportation data). These data are often ignored and could be disaggregated by gender, age, education, ethnicity, country of origin, and visa status.

* Incorporate epidemiological data obtained from programs that provide services (e.g., STI clinics) to women in the sex industry.

* Examine the classified advertisements in newspapers for travel, leisure, and sex services, and include phone-sex providers, dancing restaurants, immigration and travel agents who arrange international marriages and education visa assistance, and travel guides for male sex tourists.

We have learned from our fieldwork that diverse advertisements offering different sex services (with different business names, phone numbers, and registration numbers) may belong to the same agency that has learned to respond to client demand for different "tastes." The advertisement may explicitly say "new women just arrived from X foreign country," or "student on short-term holiday from X country" that is a known origin country for trafficked women (e.g., new girls directly from Asia, as in the reproduced advertisement in Figure 1).

Figure 1. An Advertisement for "Exotic" Sex Workers (The
Chinese Melbourne Times, 2002).

"BLOSSOMING FLOWERS"
18-25 years
Ethabell, 18 years from Sichuan Province,
People's Republic of China
JoJo, 18 years. Her first time in Australia.
She is a student and new to the business
BiBi, 19. Slim body
All women available at weekends.

Foreign language advertisements may be more explicit about the fact that foreign women are working in the sex industry than they are in advertisements in the English newspapers. Batros (2004) found this to be true for many licensed brothels in Victoria, Australia, where owners and managers were aware of the Prostitution Control Act (1994) regulations regarding sex industry advertising. However, they chose to ignore the regulations because they were aware that the designated government authorities were not monitoring sex industry advertisements.

* Samples drawn from telephone interviews or surveys with representative samples will be unlikely to include trafficked women who do not have a listed phone number (Grulich, de Visser, Smith, Rissel, & Richters, 2003; Rissel, Richters, Grulich, de Visser, & Smith, 2003). However, telephone surveys, such as the report on Sex in Australia by Rissel and colleagues (2003) on the use of commercial sex workers, could easily have included a nested question for those who have paid for sex services in the past year, to ask whether the sex worker appeared to be a citizen (e.g., did she appear foreign or have trouble communicating in the local language?). When national or large-scale surveys are being planned, such as Rissel et al.'s (2003) survey in Australia and the Global Sex Survey 2004 in 41 countries with 350,000 participants (Durex, 2004), this is an ideal opportunity to include such questions that obtain consumer or sex worker prevalence data.

* Organizations that help migrant workers or victims of violence are often in contact with WTSW who have been subject to abusive or exploitative situations (e.g., a woman who works for six months without any pay), and thus could collect data on their clients. If their data is collected systematically and analyzed over time, this might give an indication of the numbers of WTSW in a locale, keeping in mind that these are generally the extreme cases (Zimmerman et al., 2003).

Thus, it is possible to aggregate across data sources as a way of developing more reliable estimates of the numbers of WTSW in any country. We would give priority to representative survey data and data collected directly from WTSW themselves who are able to report how many other women were working at their work site. In the latter case, we caution researchers from using secondary sources of information (e.g., WTSW asking other women in the sex industry to provide information about WTSW; drawing on secondary informants such as legal representatives, chaplains, and immigration detention center and deportation center staff), since these sources are unreliable and there is no way of validating the data. These numbers can be weighted by the number of women approaching NGOs for help (considering that these are usually a small fraction of the population) to give more reliable estimates. We chose to replicate our study of women working in brothels (and approached through brothel owners) with a sample drawn from the detention center in order to validate our findings on the nature of the sex industry (Cwikel et al., 2004).

Handling Media Interest

One of the disconcerting forces in the trafficking discourse is the need to find sensational stories that grab the attention of readers and legislators. Thus, horrifying stories of abuse, torture, and exploitation are sought after by journalists and the media. The more complicated and ambiguous realities of many trafficked women's lives and the context of their trafficking experience do not make good stories. Investigative reporting is a specialized skill, often reserved for experienced journalists. Most copy reporters do not have the experience, time, or support from their editorial boards to invest in researching politicized or dangerous issues. For example, the media has not highlighted the nature of political or law enforcement involvement in trafficking operations or the extent to which women and their families are cognizant that they are entering a trafficking situation for the purpose of sex work.

From our experience (both authors have had news stories published about their research), we know it is difficult to manage the media and prevent research findings from turning into a "hot story" that trivializes human experiences and the complicated aspects of the trafficking experience. We suggest that there are ways to work more effectively with reporters while recognizing the work constraints that they face: (a) choose a senior reporter whose reputation is known and who has displayed journalistic integrity and responsibility, (b) prefer reporters who work for national, not local, newspapers, (c) prepare a press release that summarizes the research findings in a comprehensive but succinct fashion that will improve the accuracy of reporting, and (d) insist on editorial rights prior to publication.

Some journalists are fond of staging rescue operations that make great copy for newspapers. This behavior may do more harm than good and violate ethical principles (e.g., above all, do no harm). For example, Zimmerman et al. (2003) described a journalist who staged a rescue without understanding the trafficked woman's circumstances and without protecting her identity (p. 23). A more recent example was a rescue operation of three underage prostitutes in Bali that was engineered by a welfare worker in association with journalists (O'Brien, 2004). The girls were returned to their families in rural Bali. Despite the well-meaning intentions of the rescuers, it is not clear whether these girls will return to the sex industry or what consequences will follow.

Handling Risky Information

Those who gather information on the mechanisms that WTSW use to enter and work in a destination country have evidence, sometimes from primary sources, that may put the researcher, WTSW, and their families in danger. In our research, we learned about the various ways that traffickers ensure their ability to work with impunity through reciprocal ties with government agencies, such as police and immigration officials.

Our aim is to collect reliable data, but we must be aware that source material may be fed to outsiders, possibly by the traffickers, for strategic and manipulative reasons. Collaboration with researchers by those engaged in trafficking should raise questions about the ulterior motives of those informants. On the other hand, our experience showed that sometimes informants act as advocates for WTSW. While working outside the law, traffickers and key informants occasionally take advantage of their contacts with researchers to access health care or legal aid in situations where trafficked women have been victimized, are ill, or are in danger. On other occasions, sex industry informants provide researchers with information about WTSW or business' associations with government authorities such as immigration officials because they want to implicate their business competitors in the trafficking network. There are no set solutions to many of these issues, but there still may be value in the information derived from such sources. We encourage researchers to be aware of the myriad of reasons that informants participate in studies on human trafficking.

DISCUSSION

Public debate is pushing for empirical evidence on the magnitude of the problem of WTSW. We believe the time has come to start collecting data on the clients of trafficked women and their owners/bosses. This side of the equation remains obscure. Do men who pay for sex recognize that the women providing sexual services may be working illegally, under indentured conditions that often are no better than slavery, often with woefully inadequate working protection (Kelly, 2002)? Men, who make up the bulk of the consumers of WTSW, deserve a voice in the public discourse but may be put off by strident anti-prostitution forces, sex workers' unions, or the highly publicized political and criminal nature of the problem. In a general population telephone survey in Australia, one in six men (15.6%) had ever paid for sex and 1.9% had done so in the past year. This payment overwhelmingly (97%) went to women (Rissel et al., 2003). One in 10 men who had paid for sex had done so overseas, which could be a factor fueling the import of foreign women for the sex industry.

Despite our research contacts with pimps, agents, and brothel owners, we are not aware of any published research on their motivations, attitudes, and behaviors. However, like trafficked women who are working outside the law, this group can and should be studied. This would offer us a better understanding of other factors fueling the trafficking deluge and would assist efforts to describe the characteristics and magnitude of human trafficking. For example, Shelley (2003) analyzed six different models of business that provide the infrastructure for the global trafficking industry, giving insight to how the businesses use different types of network structures. Batros's (2004) study of the licensed sex industry in the city of Yarra in Victoria, Australia, explored the business operations and marketing strategies used by brothel operators to provide a variety of new and "exotic" women to meet the demand by male clients of the sex industry. Kotnik's (2004) study, also in Yarra, used qualitative and quantitative methodologies to explore community perceptions of foreign women in the licensed and unlicensed sex industry in Australia, in particular their attitude towards WTSW. Similar research can illuminate the supply-and-demand side of human trafficking.

One way to estimate the scope of the trafficking industry is to monitor classified advertisements and distribution of promotional handouts in casinos, bars and night clubs, billboards, and in phone boxes and underground entertainment venues, that reflect the money available for advertising and the businesses utilizing the media to promote their services. From these estimates, it is relatively simple to gauge the amount of profit generated in the industry and to identify the major money-makers on the scene. Information can also be gathered from applications for brothel and escort service licenses in areas where the industry is decriminalized. Size estimates can be derived from reports made by women in the sex industry on their work conditions: how many clients they see in a 24-hour period, the number of days they work per month, and the average charge per client.

In Australia, for example, in 1999 a brothel owner provided insight into human trafficking in Australia. This case revealed the trafficking of between 20 and 40 Thai women, and the informant made at least $1.2 million from the services of the 40 women (Ford, 2001). One newspaper reported that Victorian brothels earned around $1 million a week from the sex slave trade (Forbes, 1999) in an environment where the sex industry is licensed. The inadequacy of law enforcement against traffickers also ensures that fear of punishment is outweighed by the motivation for profit. The U.S. State Department noted that "penalties for trafficking of humans for sexual exploitation are often relatively minor compared to those for criminal activities like drug and gun trafficking" (U.S. Department of State, 2000, p.4).

In conclusion, we have outlined the most important problems in the conduct of high-quality research on trafficked women. We believe that the only recourse for those interested in mapping the extent of the problem is to fund inter-sectoral studies that take advantage of different access points and data sources. The situation is reminiscent of the story of five blind men trying to describe an elephant. Each one touches the animal in a different place: the flexible trunk, the smooth tusks, the huge, rough flank, the stocky legs, and the spiky tail. Each provides a completely different verbal report of what they feel, and yet no one alone can adequately describe what the elephant is really like. Combine the stories, the descriptions, the numbers, and the calculations, and the real picture of trafficking will emerge.

Note. This paper was written while the first author was a visiting scholar at the Key Centre for Women's Health in Society, The University of Melbourne, and we are grateful to our colleagues for their encouragement and support. Furthermore, the authors acknowledge the helpful suggestions of anonymous reviewers.

Manuscript accepted June 15, 2005

REFERENCES

Alexander, P. (1998). Prostitution: Still a difficult issue for feminists. In P. Delacoste & P. Alexander (Eds.), Sex work: Writings by women in the sex industry. 2nd Edition. (pp. 184-230). San Francisco: Cleis Press.

The Angel Coalition. (2003). Trafficking from Russia & the CIS. Retrieved May 11, 2003, from the World Wide Web: www.angelcoalition.org/trafficking.html.

Aral, S. O., & St. Lawrence, J. S. (2002). The ecology of sex work and drug use in Saratov Oblast, Russia. Sexually Transmitted Diseases, 29, 798-805.

Aral, S. O., St. Lawrence, J. S., Tikhonova, L., Safarova, E., Parker, K. A., Shakarishvili, A., et al. (2003). The social organization of commercial sex work in Moscow, Russia. Sexually Transmitted Diseases, 30, 39-45.

Bailey, J. (2002). Conversations in a brothel: Men tell why they do it. Sydney: Hodder.

Batros, D. (2004). The customer is always right: Meeting the demands of Victorian brothel clients. Honours Thesis. Deakin University: Melbourne, Australia.

Beauchamp, T. L., & Childress, J. F. (2001). Principles of biomedical ethics. Oxford: Oxford University Press.

The Chinese Melbourne Times. (2002, November 8). Classified advertisement.

Chudakov, B., Ilan, K., Belmaker, R. H., & Cwikel, J. (2002). The motivation and mental health of sex workers. Journal of Sex & Marital Therapy, 28, 305-315.

Coalition Against Trafficking in Women [CATW]. (2003). Retrieved May 11, 2003, from the World Wide Web: www.catwinternational.org/index.php.

Cwikel, J. Chudakov, B., Paikin, M., Agmon, K., & Belmaker, R. H. (2004). Trafficked female sex workers awaiting deporation: Comparison with brothel workers. Archives of Women's Mental Health, 6, 62-68.

Cwikel, J., Ilan, K., & Chudakov, B. (2003). Women brothel workers and occupational health risks. Journal of Epidemiology & Community Health, 57, 809-815.

Cwikel, J., Latzer, T., Latzer, S., & Press, F. STIs among trafficked sex workers, a short report. (2005). Unpublished data.

Dean, A. (2004, Feb. 25). "Sex slaves" abandoned. The Melbourne Times, p. 7.

Doezema, J. (2000). Loose women or lost women? The re-emergence of the myth of white slavery in contemporary discourses of trafficking in women. Gender Issues, 18, 23-31.

Durex. (2004). Global Sex Survey 2004. Did it for your country, yourself and for a good cause. Global Sex Survey Report: Durex.

Estes, R. J., & Weiner, N. A. (2001). The commercial sexual exploitation of children in the U.S., Canada and Mexico. Philadelphia, PA: University of Pennsylvania School of Social Work.

Feingold, D. (2003). Interview with David Feingold. UNESCO. Retrieved May 11, 2003, from the World Wide Web: http://www.eaglel.american.edu/~aa5971a/singerman_class/Feingold.htm.

Forbes, M. (1999). Our secret slave trade. The Age. Melbourne, Australia.

Ford, M. (2001). Sex slaves and legal loopholes: Exploring the legal framework and federal responses to the trafficking of Thai "contract girls "for sexual exploitation to Melbourne, Australia. pp. 1-74. Melbourne: Project Respect.

Galiana, C. (2000). Trafficking in women: Working paper. Brussels: The European Parliament.

Global Alliance Against Trafficking in Women [GATTW]. (2001). Human rights and trafficking in persons: A handbook. Bangkok: GAATW.

Grulich, A. E., de Visser, R. O., Smith, A. M., Rissel, C. E., & Richters, J. (2003). Sex in Australia: Sexually transmissible infection and blood-borne virus history in a representative sample of adults. Australian and New Zealand Journal of Public Health, 27, 234-241.

Harcourt, W. (2004). Shifting responsibilities: Displacement, global change and health. Presentation in panel on "Global movements: Impacts on the health of migrants and refugees" at 18th World Conference of Health Promotion and Health Education, Melbourne, Australia.

Hoban, E., Gordon, E., & Maltzahn, K. (2003). "We've been to war together": Experiences of trafficked women in the Australian social, legal, and political system. Draft report for the Myer Foundation. Melbourne: International Social Service, Good Shepherd Youth and Family Service, Project Respect.

Hoban, E. (2005). Club 417: A case study of sexual slavery in Victoria, Australia. Unpublished data.

Hughes, D. M. (2000, Spring). The internet and sex industries: Partners in global sexual exploitation. Technology and Society Magazine.

Hughes, D. M. (2001). The impact of the use of new communications and information technologies on trafficking in human beings for sexual exploitation. A study of the users. The Group of Specialists on the Impact of the Use of New Information Technologies on Trafficking in Human Beings for the Purpose of Sexual Exploitation (EG-S-NT) Committee for Equality between Women and Men (CDEG) Council of Europe.

Hughes, M. (2004). The use of new communications and information technologies for sexual exploitation of women and children. In C. Stark & R. Whisnant (Eds.), Not for sale: Feminists resisting prostitution and pornography, pp. 38-56. Melbourne: Spinifex.

Human Rights Watch. (2002). Hopes betrayed: Trafficking of women and girls to post-conflict Bosnia and Herzegovina for forced prostitution. Human Rights Watch, 14, 1-36.

Jeffreys, S. (2002, April-May). Women trafficking and the Australian connection: Do legalized sex industries encourage the trafficking of women and children? Are they partly to blame for the explosion in this trade over the last decade? Whose needs are being served by the legitimization of sex work? Arena Magazine, 44-47.

Jeffreys, S. (2003). The legalisation of prostitution: A failed social experiment. Retrieved May 11, 2004, from the World Wide Web: www.sisyphe.org.

Kantola, J., & Squires, J. (2002). Discourses surrounding prostitution policies in the UK. Paper presented at PSA Annual Conference, Aberdeen.

Kelly, E. (2002). Journeys of jeopardy: A review of research on trafficking in women and children in Europe. London: Child and Woman Abuse Studies Unit, University of North London, International Organization for Migration.

Kelly, L. (2003). The wrong debate: Reflections on force is not the key issue with respect to trafficking in women for sexual exploitation. Feminist Review: Exile and Asylum--Women Seeking Refuge in "Fortress Europe," 73, 139-144.

Kotnik, E. (2004). Breaking into Australia under false pretenses: Community attitudes and perceptions towards trafficked women in the sex industry in Australia. Honours Thesis. School of Health & Social Development. Melbourne, Deakin University.

Kvinnoforum. (1999). Crossing borders against trafficking in women and girls: A resource book for working against trafficking in the Baltic Sea Region. 2nd edition. Sweden: Kvinnoforum.

Levenkron, N., & Dahan, Y. (2003). Women as commodities." Trafficking in women in Israel 2003. Haifa: Hotline for Migrant Workers, Isha L'Isha--Haifa Feminist Center, and Adva Center.

Long, L. D. (2004). Anthropological perspectives on the trafficking of women for sexual exploitation. International Migration, 42, 5-31.

Martens, J., Pieczkowski. M., & van Vuuren-Smyth, B. (2003). Seduction, sale and slavery: Trafficking in women & children for sexual exploitation in Southern Africa. IOM--International Organization for Migration, Regional Office for Southern Africa, Pretoria, South Africa.

Maxim Institute. (2003). Prostitution Reform Bill (2003)--Summary report on the effects of prostitution legalisation in Victoria, Australia. Auckland, New Zealand.

Mirkinson, J. (1997). The global trade in women. Earth Island Journal, 13, 30-31.

O'Brien, N. (2004, Feb 21-22). Bali child sex slaves rescued. The Weekend Australian, pp. 7, 19.

Outshoorn, J. (2002). Legalizing prostitution as sexual service: The case of the Netherlands. Copenhagen: EPCR.

Raymond, J. G. (2001). A comparative study of women trafficked in the migration process: Patterns, profiles and health consequences of sexual exploitation in five countries (Indonesia, the Philippines, Thailand, Venezuela and the United States). MINCAVA. Retrieved April 22, 2003, from the World Wide Web: www.mincava.umn.edu/traffick.asp.

Richard, A. O. (1999). International trafficking in women to the United States: A contemporary, manifestation of slavery, and organized crime. Washington, DC: Center for the Study of Intelligence (CSI).

Rissel, C. E., Richters, J., Grulich, A. E., de Visser, R. O., & Smith, A. M. (2003). Sex in Australia: Experiences of commercial sex in a representative sample of adults. Australian and New Zealand Journal of Public Health, 27, 191-197.

Shelley, L. (2003). Trafficking in women: The business model approach. Brown Journal of World Affairs, 10, 119-131.

Sulavik, C. (2003). Facing down traffickers: Europe takes on its fastest-growing criminal enterprise. Newsweek, Aug 25/Sept. 1, International Edition, 27-28.

United Nations. (2000). Protocol to prevent, suppress and punish trafficking in persons, especially women and children supplementing the United Nations Convention against transnational organized crime. Retrieved April 10, 2003, from the World Wide Web: www.undcporg/odccp/trafficking_human_beings.html.

United Nations Children's Fund [UNICEF]. (2003). Trafficking in human beings, especially women and children, in Africa. Florence: United Nations Children's Fund Innocenti Research Centre.

U.S. Department of State. (2002). Victims of trafficking and violence Protection Act 2000: Trafficking in persons report. Washington, DC: Department of State.

U.S. Department of State. (2003). Trafficking in persons report, June 2003. Retrieved from the World Wide Web: www.state/gov/g./tip/rls/tiprpt/2003.

USAID Office of Women in Development. (1999). Women as chattel: The emerging global market in trafficking. Gender Matters Quarterly, 1, 1-8.

Vanderberg, M. (1997). Trafficking of women to Israel and forced prostitution. Jerusalem, Israel: Israel Women's Network.

von Struensee, V. (2000). Globalized, wired, sex trafficking in women and children. Murdoch University Electronic Journal of Law, 7, 1-10.

Watts, C., & Zimmerman, C. (2002). Violence against women: Global scope and magnitude. Lancet, 359, 1,232-1,237.

Widom, C. S., & Kuhns, J. B. (1996). Childhood victimization and subsequent risk for promiscuity, prostitution, and teenage pregnancy: A prospective study. American Journal of Public Health, 86, 1,607-1,612.

Wijers, M., & Lap-Chew, L. (1997). Trafficking in women, forced labour and slavery-like practices in marriage, domestic labour and prostitution. Utrecht and Bangkok: Foundation Against Trafficking in Women (STV), The Global Alliance Against Trafficking in Women.

Yea, S. (2004). Push and pull factors in trafficking of women: Case studies. Paper presented at the People Trafficking, Human Security and Development Conference. Australian National University. Development Studies Network: Canberra. 1-2 September 2004.

Zimmerman, C., & Watts, C. (2003). WHO ethical and safety recommendations for interviewing trafficked women. Geneva Switzerland: World Health Organization.

Zimmerman, C., Yun, K., Shvab, I., Watts, C., Trappolin, L., Treppete, M., et al. (2003). The health risks and consequences of trafficking in women and adolescents. Findings from a European Study. London: London School of Hygiene & Tropical Medicine.

Julie Cwikel

Ben Gurion University of the Negev, Israel

Elizabeth Hoban

Deakin University, Australia

Address correspondence to Prof. Julie Cwikel, Director of Ben Gurion University's Center for Women's Health Studies and Promotion, POB 653, Ben Gurion University of the Negev, Beer Sheva, 84105, Israel; e-mail: jcwikel@ bgumail.bgu.ac.il.