Risk for Researchers Studying Social Deviance or Criminal Behavior

Researchers often encounter dangerous situations while conducting social research. The concept of risk to researchers refers to the possible harm that may occur to researchers while in the field or after leaving a research project. This study explores issues experienced by social scientists engaged in research on social deviance or criminal behavior. The goal of this research was to discover the types of risk experienced by social scientists and any mediating factors affecting the experience of risk. An online survey was conducted to gather data on issues experienced by social scientists. This study found that researchers experienced a variety of risks within the categories of physical/health, emotional, legal, and personal/professional. Each of the survey options for risk were reported by at least one respondent; however, the greatest number of risks reported were of an emotional or personal/professional nature. There were no mediating factors found to be significant in relation to the experience of risk. This was a surprising finding especially for the variable of gender as it is suggested that gender plays a role in the experience of difficulties.


Introduction
Researchers face many dangers (e.g., injury, illness, emotional distress, violence, death) when entering the field. While there is a growing recognition of the risks involved in research, the full range of risk experienced is not yet fully realized. There have been a few studies (Worley et al. 2016;Kenyon and Hawker 1999;Johnson and Clarke 2006;Dickson-Swift et al. 2007;Bloor et al. 2007) that shed light on risk and provide examples of the emerging recognition of the issue of risk faced by researchers. However, most information available on the occurrence of risk is anecdotal. Furthermore, it is unknown if researcher characteristics, environment, or topic under study affect the occurrence of risk.
There are a variety of fieldwork and non-fieldwork-based methods used to collect data within the social sciences. Researchers go into crack houses, half-way houses, and prisons. Research can be conducted within a researcher's place of business or other areas under his/her control. There are even some types of research that involve what Hunter S. Thompson referred to as edgework, defined as "voluntary risk-taking" (Lyng 1990). Edge ethnography is a term used to describe researchers utilizing covert or full participation methods to conduct research (Tewksbury 2009). Regardless of the research method employed, many social researchers may not fully comprehend risk until they are in the process of research. However, there is a dearth of academic research pertaining to risk. This study addresses that gap by collecting information on the experience of risk from researchers employing various types of research methods to study social deviance and criminal behavior.
Inexperienced researchers are at greater risk because they are not able to anticipate the risk that awaits them (Paterson et al. 1999). Students sometimes take part in research (Morris and Marquart 2010;Pogrebin 2010). This may be during a methods course or as part of a thesis or dissertation project. "Novice researchers may be especially at risk, as they are often more concerned with their methodology and response rate than ensuring their own safety" (Sharp and Kremer 2006, p. 321). According to Lee (1995), a researcher's first fieldwork endeavor is a type of rite of passage, whereby the young researcher must survive on his/her own. Vanderstaay (2005) provides an example of a researcher's first fieldwork experience that did not go well. He described an experience with fieldwork that left him with feelings of guilt and suffering secondary trauma. Vanderstaay began dissertation research on the topic of the relationships between schools and the juvenile justice system. His study of one juvenile led him to face ethical dilemmas regarding his own level of intervention in the life of the subject. He was left feeling that his involvement contributed to the situation in which the juvenile subject killed a family friend. Vanderstaay entered the field unprepared for such ethical situations and with no emotional support in place to assist him in dealing with the stress of the experience. He did not finish this dissertation work and would not write about it for another ten years. This example demonstrates how researchers with a lower level of experience may enter research unprepared for the risks that await them.
It is suggested that gender plays a role in the experience of difficulties during research (Arendell 1997;Gurney 1985;Easterday et al. 1977). Liebling (1999) reports that the interactions were different for male and female members of her research team conducting interviews with prisoners. Perrone (2010) reports that during her research on drug use, subjects requested sex or nudity as a type of payment for their participation. For Sampson and Thomas (2003), being female most likely contributed to their experience of sexual harassment while conducting research on seafarers. Westmarland (2000) offers a slightly different view. In conducting research with police, she noticed that being a female seemed to bring about a protective nature in the male officers. This would indicate that in some research, being female may provide a protective factor. According to Paterson et al. (1999), the experience of risk to researchers is not strictly a gender issue; "researchers of both genders have experienced threats to their safety by persons of either sex" (p. 261). Gender does not produce risks, but rather it may amplify existing risks in certain settings (Bloor et al. 2007). "In some cases the characteristics of the researcher with respect to participants may create the conditions for harassment or violence" (Sharp and Kremer 2006, p. 318). Inciardi (1993) suggests that women should not enter the crack house research environment due to the potential for rape. It is inferred that risk is associated with the environment, as well as, participants and the topic under study (Paterson et al. 1999). Worley et al. (2016) recently conducted phone interviews with eight social scientists who had utilized fieldwork methods to study crime, deviance, or social control. They discovered that all of the researchers experienced some sort of "significant trauma" of a professional, legal, physical, or emotional nature (p. 204). Research is conducted on a variety of topics involving an array of individuals of different ages and genders living in different geographical settings. Sociologists and criminologists conduct research on topics such as drug addicts (Perrone 2010;Inciardi 1993), strippers (Price-Glynn 2010Israel 2006), burglars (Wright et al. 1992), law enforcement (Westmarland 2000;Van Maanen 1988;Marks 2003;Holdaway 1983;Goldsmith 2003), gangs (Hopper and Moore 1990;Venkatesh 2008), phone sex workers (Mattley 1998), and sex in public places (Humphreys 1968). However, a breakdown of risk across the range of research topics or locations does not exist. What was more readily identifiable were types of risks experienced. For the purpose of this study, these risks were logically placed into risk categories of (1) physical/health, (2) legal, (3) emotional, and (4) personal/professional.

Physical/Health Risks
Physical dangers exist in various forms when conducting research. Respondents in a study by Kenyon and Hawker (1999) reported issues of a serious nature including physical assault, a shooting, sexual assault, and a near rape. Inciardi (1993) reports that he was present in a crack house once when a gun was fired in another room of the house. Fortunately, as everyone took shelter, no one was hurt. Another example is provided by Hopper and Moore (1990) who had guns pulled on them one morning when they entered a bikers' camp while conducting research on motorcycle gangs. The bikers, sleepy and hung over from the previous night's party, thought Hopper and Moore were competing gang members. They very well may have been killed or experienced serious injury if one biker had not recognized them and interceded on their behalf. Marks (2003) also encountered risk from guns on several occasions as she embarked on patrols with police officers, often entering into situations knowing that she may not survive the day. She reports that it was typical to be shot at while on routine patrol. Westmarland (2000) wore a bulletproof vest or anti-stab vest while observing police in England. On one occasion she was forced to hide behind a tree to avoid being shot by an offender. Venkatesh (2008) also describes a gang shootout in which he was forced to take cover while watching others get shot. Martin Sanchez-Jankowski was stabbed once and shot twice during his time researching gangs (Worley et al. 2016). Calvey (2000) reports being in a threatening situation involving armed men while he was conducted ethnographic research on door supervisors, often called bouncers, in Manchester, England. These anecdotal examples illustrate the very real danger of physical risk from firearms/guns that exists for researchers.
In 2009, Annie Le, a Yale University pharmacological graduate student was murdered in her research laboratory where she was conducting biological research (Vitagliano and Solomon 2011). Her body was later found inside a laboratory wall. On 17 March 2011, Raymond Clark, III, a university lab technician, plead guilty to both attempted rape and murder of Annie Le. While this example is not taken from the discipline of sociology or criminology, it is noteworthy that such violence could occur in non-fieldwork-based research. An example within the field of criminology is the death of Caribbean criminologist, Ken Pryce. He mysteriously disappeared while studying criminality in Jamaica. "His body was later found washed-up on a beach" (Bloor et al. 2007, p. 18). Fortunately, there is not an abundance of examples of death during research.
Some researchers have reported physical assault, sexual assault, and a near rape (Kenyon and Hawker 1999). During his fieldwork on homeless families, Blackman (2007) was physically attacked by a man who thought he was someone else. While interviewing a prisoner, Liebling (1999) had a pen plucked from her hand. The prisoner pointed out that if he had wanted to he could kill her with the pen. James Marquart was assaulted by an inmate while studying correctional officers (Worley et al. 2016). As graduate students, Sharp and Kremer (2006) experienced sexual harassment and intimidation during research projects. Sampson and Thomas (2003) also report experiencing sexual harassment while conducting research on seafarers.
One might consider that females researching deviant populations may find themselves in locations and, at times, that makes them especially vulnerable to sexual harassment or assault (Lee 1995). However, it is not only the criminal or deviant who pose a threat to female researchers. Stanko (in Lee 1995) surveyed female members of the American Society of Criminology about sexual harassment. He found that one in three of the respondents had been on the receiving end of sexual comments or sexual harassing behavior. This behavior was perpetrated by police, prison officers and court officials.
Researchers are also sometimes placed in harm's way due to the political climate of the country in which they are conducting research. These researchers are sometimes viewed as spies working with or helping the enemy. Thirteen percent of respondents in Howell (1990) study of anthropologists reported being suspected of spying while conducting fieldwork (p. 97). Five of the 204 respondents noted being involved in a hostage-taking event while in the field (p. 99). Perhaps most of the sociologists and criminologists studying social deviance and criminal behavior are conducting research in countries of a more stable political climate. However, this does not rule out the possibility of them being suspected of spying or being held hostage. These types of risks are well illustrated by Venkatesh (2008). In the beginning stages of his research on gangs, Venkatesh was held hostage in a stairwell. Gang members thought he was a member of a rival gang and that he was there to spy on them. On that occasion, gang members attempted to intimidate him by flashing guns and knives in a threatening manner.

Legal Risks
Legal problems may also arise as a result of conducting research. Two main areas of research that seem to capture the interest of police and prosecutors have been illegal drug culture and sexology. Researchers have had information confiscated by authorities that wish to prosecute participants in the study (Sonenschein 2001). Sometimes that data are destroyed or simply not returned to the researcher for an extended period of time. There have been times when researchers found themselves being pressured to testify against their subjects. Refusal to do so can result in imprisonment. For example, Scarce (1994Scarce ( , 2001 spent 159 days in jail. Police agencies believed that he possessed valuable information and held him in contempt of court for not disclosing such information. Martin Sanchez-Jankowski was also jailed for refusal to disclose information discovered during his research (Worley et al. 2016).
According to Polsky (1985), legal danger for researchers comes primarily from law enforcement professionals. This is due to the type of information researchers uncover during the course of fieldwork that law enforcement would like to access. An example of this issue is seen in the Boston College Belfast Oral Project where researchers had gathered oral accounts of individuals who had been involved with the Irish Republican Army (IRA) and loyalist groups. Participants had been ensured that their information would not be released until their death. However, through the use of an international subpoena, officials in the United Kingdom were able to obtain a portion of the files (Borgersen 2017).
Inciardi (1993) states that while there are dangers from grand juries, prosecutors, and police, these are not typical issues for researchers. Law enforcement agencies likely have better information that can be used for prosecution. However, researchers do run the risk of being arrested and imprisoned for participation in activities while conducting research (Sonenschein 2001). This is often the case when the researcher is simply in the wrong place at the wrong time.
An example is provided by Inciardi (1993) who was in a car with three drug users who were giving him a tour of the drug scene when they stopped at a convenience store for one of the men to pick up some cigarettes. The man quickly ran back to the car with cash and a gun after robbing the store. As they fled the area, Inciardi convinced them to let him out of the car. The three men were shortly arrested. Certainly, in this case, if he had remained in the car he would also have been arrested. Inciardi was not so lucky during a drug bust on a crack house while he was present. On this occasion, he was handcuffed and put in jail. He was later released with no drug charge. Crimes sometimes do happen when a researcher is present and, therefore, the researcher is at risk of arrest.
Another example is illustrated by Humphreys (1968) who was also in the wrong place at the wrong time. He had been standing outside a tearoom (public place for sexual encounters) talking to another man when police approached and asked for his identification. When he refused to provide any information, the police harassed him for not cooperating. Humphreys was subsequently arrested for loitering.
Being arrested is also a possibility when a researcher becomes an active participant in the research. Jeff Ferrell was arrested for destruction of property while he was conducting a study on graffiti artists (Worley et al. 2016). He was convicted and placed on probation. Wright et al. (1992) report their most dangerous experience occurred while driving with a subject in the car. The subject saw someone on the street and wanted the car stopped so that he could get out of the car and kill the individual. They managed to calm the subject after refusing to stop the car and driving away from the area. The management of a volatile subject could weigh heavily on the researcher. Certainly, to be present when the subject acts in a violent manner would have some impact on the researcher. Yet, researchers are expected to be objective in their report of research. Perhaps for this reason, the emotions experienced by the researcher have escaped many publications. Due to the stressful situations researchers face, mental illness and emotional stress are a concern.

Emotional Risks
According to Liebling (1999), as her research team's time in the field went on, their interviews with prisoners became "harrowing" to the point of being "traumatic encounters" (p. 150). Team members would often emerge from interviews exhausted and upset. Liebling reports that the team members would engage in risky behaviors such as alcohol abuse and driving in an unsafe manor to relieve stress.
Respondents in a study by Kenyon and Hawker (1999) reported feelings of "isolation, vulnerability and/or fear" while engaged in fieldwork (p. 317). An example of this type of emotional stress is provided by Miller (1986) who conducted research on street women, female hustlers, and prostitutes. She interviewed women in various settings such as halfway houses and prison. While conducting interviews in a prison, she was sometimes interrupted by male inmates who caused her fear. On one occasion, she was alone in a prison chapel, waiting on an interviewee, when a male inmate entered. During her conversation with the man she feared that he was thinking about sexually assaulting her. The person Miller was to interview entered the room and ran him off. On another occasion, while at a home for women, she was watching TV with some of the women when a man on a motorcycle arrived. Miller reports that he gave her the message that she was not welcome there. She reported feeling physically threatened on that occasion. Lecocq (2002) raises the point that many researchers do not express or discuss their feelings, especially feelings of fear because of the emphasis that one must remain neutral while conducting research. The researcher is expected to remain objective and detached; thereby leaving any personal reflection and what might be viewed as contamination out of the analysis report on the research endeavor (Punch 1986).
The element of fear has received little scholarly attention, especially within the discipline of criminology. This is most likely because many researchers manage difficult situations without experiencing major harm (Goldsmith 2003). For example, Price-Glynn (2010) reports that although she did not directly experience any major issues, she did fear possible aggression from men while conducting research in a strip club (p. 207). Jamieson (2000) recounts that on many occasions when interviewing youth of varying levels of deviance or criminal behavior, she feared for her physical well-being. Jamieson points out that entering an unfamiliar environment can lead to feelings of stress and apprehension. She recalls several occasions where she went to meet with one youth and found herself surrounded by a group of males. Austin (2003) reports that while conducting research in Mindanao he felt the need to be on guard. When talking about working with the police, Van Maanen (1988) also expresses the experience of fear. Experience of this type of fear or feeling the need to be on constant alert could cause the researcher immense stress. Calvey (2000) reports this type of stress in his account of research as a doorman. He expresses the experience of tremendous fear while walking home one evening after a particularly threatening encounter. He likened the research encounter with a "paranoid nightmare" (p. 51). Miller (1986) experienced other feelings during the course of her research on street hustlers. She felt "angry, generally upset, and depressed" (p. 189). She was angry that children grew up in the types of situations she was hearing about and that people could be so brutal to one another. She was depressed over the lack of options for the women and upset that society bred the kind of hatred, discrimination, and inequality she was seeing. Miller did make mention that as paralyzing as her emotions could be, they provided motivation for the continuation of research. Dickson-Swift et al. (2007) point out that researchers conducting secondary analysis also experienced emotional issues including "sleep disorders, emotional changes and a need for social support" (p. 328). Kinard (1996) found that researchers reviewing case records of abused children experienced sadness, anger, frustration and a sense of powerlessness. There was a similar discovery for researchers reviewing case records of rape victims. The researchers reported the experience of emotions similar to those reported by victims of rape (Alexander et al. 1989). They reported feelings of anger, anxiety, fear, and sadness. They also experienced insomnia, nightmares, nausea and generalized pain. Worley et al. (2016) found that some of the participants in their study experienced what they referred to as compassion fatigue. In clinical psychology, the term is used to describe "a condition where care providers, or in this case field researchers, experience stress as a result of over empathizing with their clients" (Worley et al. 2016, p. 297). Sampson and Thomas (2003) report that long hours of study, combined with the isolation of being on a ship during their research, created emotional stress for them. They report one incident in which a researcher was confronted with great hostility by the captain of the ship. He isolated the researcher from the crew while they were out to sea for 16 days. Other researchers who report experiencing hostility include Marks (2003) and Wright et al. (1992). Marks (2003) experienced hostility from both sides of the law while conducting a study on police in South Africa. Wright et al. (1992), who were studying active residential burglars, encountered hostility from subjects who feared that they were being set up.
When research experiences become stressful or dangerous, emotions are sure to surface. If not dealt with appropriately, these emotional issues can lead to issues of substance abuse and reckless behavior. Individuals could be left emotionally scarred by events encountered during the course of research. The research reveals emotional issues in accounts of both fieldwork and non-fieldwork-based research. These emotions include fear, depression, isolation, anger, vulnerability, stress, and apprehension.

Personal and Professional Risks
Researchers may also experience personal and professional difficulties due to their research. Researchers might experience stigma, a type of social disgrace, due to their work. Stigma can affect a researcher on both a personal and professional level. Some researchers are subject to what Erving Goffman called courtesy stigma (Kirby and Corzine 1981;Mattley 1998;Miller and Tewksbury 2001). This is when the stigma of the topic is assigned to the researcher and they experience the labeling effect of guilt by association (Kirby and Corzine 1981;Mattley 1998;Miller and Tewksbury 2001). Mattley (1998) more accurately refers to this as a (Dis)courtesy stigma. She feels that having a stigma normally associated with deviance transferred to her serves as a discourtesy.
During their research on homosexuals, Kirby and Corzine (1981) found that nonacademic individuals rather than academics more quickly labeled a researcher. Conversely, Mattley (1998), while conducting a study on fantasy phone sex workers, found that stigma was more of an issue with academics than with nonacademic individuals. Her interactions with colleagues changed as she began to realize that other professionals viewed her differently. Some male colleagues began to make sexual comments about the study. Many seemed incapable of focusing on the topic as viable and valuable research. They would often ask inappropriate questions about the fantasy phone sex conversations. Many times, Mattley heard from other people that they could not do that type of research. They were clearly not comfortable with the topic. Rik Scarce and Carol Rambo attribute difficulty with gaining employment in academia to stigma associated with their research (Worley et al. 2016). Israel (2006) also had experiences with stigma during her research on strippers. However, she received much support from her academic affiliations and her family, while it was her friends (especially males) who seemed to associate her with her work.
The experience of an ethical dilemma is another area of concern on both a personal and professional level. Marks (2003) who traveled with the police and directly observed raids, was asked on more than one occasion to take a female suspect into another room and conduct a physical search including her vaginal area. Not only did this place Marks in a potentially violent position if the suspect were to attack her, but it also raised ethical concerns for the limit of her role as a researcher. Holdaway (1983) first faced ethical concerns in deciding to conduct covert research of the police. He states that "covert research and the ethical questions it raises create conditions of stress within which the sociologist has to live with himself" (p. 9). Holdaway felt strain over balancing his ethical limits, the research, and his role as a police officer. Whyte (1981), who conducted ethnographic research on street gangs in Boston, provides another example of the stress created by ethical boundaries when he decided to vote more than once in an election. He risked his research as well as his freedom when he broke a federal law prohibiting repeated votes. He reports struggling with his conscience after the illegal act. Calvey (2000) also struggled with ethical dilemmas during his research. He faced many situations in which he witnessed illegal drug use, theft of door money, and physical assault on people. As part of the job, he also physically restrained people. Calvey openly recounts the emotional dangers he faced as he experienced vulnerability to his sense of self.
Another example is provided by Gans (1962) who conducted research in a slum district. He was not completely honest with his subjects about his role as a researcher because he feared it would limit his access. He reports a feeling of guilt over what he considers to be a misuse of his relationships to gather data.
According to Marquart ethical dilemmas are inherent in conducting research (Worley et al. 2016). He stated: There were ethical dilemmas all day long. You're seeing violence and a lot of things that go against the grain of normalcy. So, there are questions that come up. Do you blow the whistle on people or do you maintain your silence? (p. 301)

Mediating Factors
It has been inferred that risk is associated with the environment, as well as, participants and topic under study (Paterson et al. 1999), and that gender may also contribute to the level of risk experienced (Arendell 1997;Gurney 1985;Easterday et al. 1977). However, a breakdown of risk across the range of research topics or locations has not been discovered. How does the environment or population under study affect the experience of risk? Elements to be considered in relation to the experience of risk include data collection method, researcher demographics, topic of study, population under study, and research location or setting. It is important to explore these elements in relation to the experience of risk in order to formulate an accurate illustration for the occurrence of risk and how such factors may mediate risk.
Many anecdotal accounts were discovered that illustrate issues researchers have experienced while conducting research. However, there were limited in-depth research analysis discovered that took a look at the topic of risk faced by researchers conducting any type of social deviance or criminal behavior research. Many different risks were revealed. These risks can be categorized as physical/health, legal, emotional, and personal/professional. The review uncovered quite a few individuals who were willing to share their research experiences in reflective accounts of their endeavors (Inciardi 1993;Humphreys 1968;Hopper and Moore 1990;Liebling 1999;Israel 2006;Vanderstaay 2005;Marks 2003;Miller 1986;Nilan 2002;Scarce 2001;Sonenschein 2001;Westmarland 2000;Calvey 2000;Jamieson 2000;Mattley 1998;Wright et al. 1992;Kirby and Corzine 1981;Venkatesh 2008;Whyte 1981;Holdaway 1983). It is not known if their experiences are typical for research involving social deviance or criminal behavior, or if they are presenting uncommon occurrences. This research attempts to aid in the development of a better understanding of risks experienced by sociologists and criminologists conducting research on social deviance or criminal behavior. Furthermore, this research explores elements that may mediate the experience of such risk.

Methods
This cross-sectional research sought to fill the knowledge gap regarding risk to the researcher by addressing the following research questions:

1.
What types of risk do social scientists experience when conducting research on social deviance or criminal behavior? 2.
What types of risks are experienced most by social scientists studying social deviance or criminal behavior? 3.
Does gender, age, data collection method, first research experience, age of population under study, level of education, country of study, research location, or research setting, impact the occurrence of risk?
The target population for this research was social scientists, primarily criminologists and sociologists, studying social deviance or criminal behavior. This sample was not limited to qualitative fieldwork. There is possibility for both quantitative and qualitative researchers collecting data by use of fieldwork or non-fieldwork-based methods to experience risk. Therefore, the sample included social scientists who have conducted qualitative and/or quantitative fieldwork or non-fieldwork-based research on social deviance or criminal behavior. In order to establish a list of potential participants, professionals in the fields of criminology and sociology were identified from membership lists for the American Society of Criminology (ASC), the Academy of Criminal Justice Sciences (ACJS), and the Midwest Sociological Society (MSS). These organizations were chosen because they spur further inquiry that advances our understanding of societal issues and they encourage members, including students, faculty and other professionals, to conduct research that is contributory. Therefore, within these membership lists, it was likely that a sample would be captured that would lead to a greater understanding of risk to the researcher. Since students at various levels of education assist with or directly conduct research, it was only prudent to gather their experiences also.
Because the focus of this research was to discover the types of risks experienced by those studying social deviance or criminal behavior, participants were asked to indicate if their research topics fell into the categories of social deviance or criminal behavior. Those who did not meet these criteria were excluded from the project.

Instrumentation
As a large portion of the participants in this project were most likely employed in academia and accustomed to use of technology, an online survey was created for use in this endeavor. The survey queried respondents on types of risk experienced. The instrument utilized checklist and Likert scales. There were some demographic and descriptive items included to provide data that could help build a complete picture of the researcher and the research environment. These included age, gender, level of education, professional discipline, and level of research experience. Categories for level of education included bachelor's degree in progress through doctoral degree completed. Several levels were used in order to include all possible levels of education that may be involved in research. Options for sociologist and criminologist were offered for selection under professional discipline. A category of other was also offered as it was recognized that not all those who study social deviance or criminal behavior fall into the category of sociologist or criminologist. For determining level of research experience, the respondent was asked how many years they have been conducting research. Descriptive items related to the research topic, research environment, and population were also included. These include items pertaining to the respondent's first research endeavor, type of research methods used, data collection employed, research setting, research location, focus of research topic, gender of population under study, and age group of population under study. Efforts were made to ensure that all possibilities were included in the categories for each descriptive item.
Within the survey, respondents were presented with the opportunity to report the types of risk they have experienced. There were four main types of risk listed; physical/health, legal, emotional, and personal/professional. Each list had various options and allowed the respondent an opportunity to add additional risks under the category of other.

Results
Overall, 5455 individuals were invited to participate in an online survey. A total of 1065 individuals accessed the online survey. There were 27 surveys with no answers given for any item. These 27 were excluded from analysis, leaving 1038 surveys for analysis. The resulting response rate was 19%. Of those 1038 surveys, 805 respondents indicated that they have conducted research on social deviance or criminal behavior. This was a qualifying question to continue with the survey.
Of the 805 remaining respondents, 330 (41%) reported that they had experienced risk due to their involvement in research on social deviance or criminal behavior. Those who had not experienced risk were excluded from questions pertaining to research projects that involved risk. However, those individuals were included in questions on demographics. There were 718 respondents who identified their discipline as criminologists (68.8%), sociologists (20.2%), or other social scientists (11%). Among other social scientists, 16.5% of the respondents identified as sociologists and criminologists. Another 19% identified as psychologists and 12.7% identified as political scientists. Additionally, 5.1% of the respondents identified as victimologists. There were 373 (52.2%) males and 341 (47.8%) females who participated in the survey.
The first two research questions are addressed by analysis of the data from those reporting the experience of risk. Of those 330 respondents, 310 answered questions on the survey pertaining to their specific research projects that they associated with risk. Frequencies were reviewed to identify the risks experienced by researchers.

Physical/Health Related Risks
A total of 308 respondents answered questions on physical/health related risks (see Table 1). Among physical/health related risks, assault was reported by 56 (18.2%) people and this was the most reported physical/health related risk. Held hostage was reported by four (1.3%) people, rape was reported by three (1%) people, robbery was reported by 11 (3.6%) people, and shooting was reported by 14 (4.5%) people.
Interestingly, 63 (20.5%) people reported the experience of an additional physical/health related risk that was not on the survey checklist. Additionally, 17 (5.5%) of those people reported the experience of a second additional risk. These additional risks are noted in Table 1 as other physical/health related risks 1 and 2. The second most reported physical/health related risk was found within these categories. Threats were reported by 21 (6.8%) people. Additional other physical/health related risks included sexual assault, exposure to disease, contracted disease, exhaustion, extortion, gunfire, harassment, intimidation, hit by car, insects, lock down in jail, loss of appetite/digestion issues, motor vehicle accident, possible airplane crash, placed on hit list, poison-pen letter, sleep difficulty, tear gas exposure, tiredness, and verbal assault.

Legal Risks
A total of 305 respondents answered questions on legal risks (see Table 2). Among legal risks, pressure to testify was the most reported with 19 (6.2%) people. Confiscation of research materials was reported by 16 (5.2%) people, arrest was reported by 12 (3.9%) people and detainment was reported by seven (2.3%) people. Also, 25 (8.2%) people reported experiencing an additional risk not included in the survey checklist. Of those 25 people, eight (2.6%) reported a second other risk. These additional other risks are noted as other legal risks 1 and other legal risks 2 in Table 2. Additional legal risks reported by respondents included law suit or criminal prosecution, pressure to release information, telephone tap, injunction, lack of cooperation, and threats of legal action.

Emotional Risks
A total of 309 respondents answered questions on emotional risks (see Table 3). Two of the most frequently reported risks are within the category of emotional risks with emotional stress reported by 136 (44%) people and fear reported by 115 (37.2%) people. Feeling isolated was reported by 42 (13.6%) people, emotional trauma was reported by 28 (9.1%) people, and depression was reported by 26 (8.4%) people. Also, 26 (8.4%) people reported an additional emotional risk that had not been included in the survey checklist. Of those 26 people, four (1.3%) reported a second additional emotional risk. These additional risks are noted in Table 3 under other emotional risks 1 and other emotional risks 2. These additional risks included anger, anxiety, concern, frustration, disorientation, uncertainty, timidity, lack of support, and difficulty in expressing feelings.

Personal/Professional Risks
A total of 302 respondents answered questions on personal/professional risks (see Table 4). Ethical dilemma was reported by 119 (39.4%) people. Stigma was reported by 61 (20.2%) people, suspicion of spying was reported by 47 (15.6%) people, and sexual harassment was reported by 17 (5.6%) people. Also, 20 (6.6%) people reported an additional personal/professional risk that had not been included in the survey checklist. Of those 20 people, five (1.7%) reported a second additional personal/professional risk. The additional risks are noted in Table 4 under other personal/professional risks 1 and other personal/professional risks 2. These additional risks included job loss, ban from research location, harassment at workplace, harassment at home, IRB investigation, jealousy, professional ostracism, told to leave country, sabotage, TSA watch list, loss of money, loss of professional advancement, compromised objectivity, resentment, and slander.

Mediating Factors
Chi-square and logistic regression were used in analysis of question three, does gender, age, data collection method, first research experience, age of population under study, level of education, country of study, research location, or research setting, impact the occurrence of risk. For researcher's gender, 136 females and 153 males indicated that they had experienced risk, while 205 females and 220 males indicated that they had not experienced risk. For this analysis, Chi-square = 0.095, df = 1, p = 0.757, there was no statistically significant relationship found between gender and the experience of risk.
Logistic Regression was used to analyze the relationship between researcher's age and the experience of risk (see Table 5). Age ranged from 21 to 87 with a mean of 43.87. Age was not found to be a significant predictor for the experience of risk (Wald = 2.288, p > 0.05). The variable total risk was created by summing the respondents' reported risks. Respondents were asked to answer several questions about research projects (up to three separate projects) that were associated with risk. Because the first account of risk had the greatest number of respondents (N = 310), this account was utilized for analysis instead of the second (N = 99) or third (N = 25). Recall that 475 respondents indicated that they did not experience risk and were not included in questions pertaining to a research project associated with risk. Their assigned value for total risk was zero. Total risk ranged from 0 to 12. Because of issues with Skew and Kurtosis it was necessary to transform the data for analysis. The data was recoded into two categories, zero to one risk (N = 589, 73.2%) and two to 12 risks (N = 216, 26.8%).
For method of data collection, fieldwork was utilized by 190 respondents and non-fieldwork was used by 19 respondents. First research project as the primary researcher was reported by 79 respondents, while 130 respondents reported that it was not their first research project as the primary researcher. For categories of age of population under study, 28 respondents reported children (under age 18), 237 respondents reported adults (18 and older), and 31 respondents reported both. For researcher's level of education, there were 228 non-Ph.D. and 492 Ph.D. respondents. For research country, 257 respondents conducted research in the USA, while 39 reported that their research was conducted in non-USA countries. For research location, 165 respondents conducted research in an urban area, 20 respondents conducted research in a suburban area, 34 respondents conducted research in a rural area, and 72 respondents conducted research in multiple locations (combination of urban, suburban, and/or rural). For research setting, 126 respondents conducted research in an institutional or professional location, 43 respondents conducted research in a public setting, 34 respondents conducted research in an informal setting, and 96 respondents conducted research in multiple settings (institutional/professional, public, and/or informal).
Chi-square was conducted for each of these variables in relation to total risk (see Table 6). No statistically significant relationship was found between total risk and any of these possible mediating variables.

Discussion
From the data reviewed it is clear that risk does exist for social scientists studying social deviance and criminal behavior. While each of the survey options for risk were reported by at least one respondent, the risks most reported include the experience of emotional stress, fear, and ethical dilemma. The reported occurrence was low for several areas of risk (e.g., rape, held hostage, and detainment by authorities). That fact does not negate their importance. It is interesting that the greatest number of risks reported were of an emotional or personal/professional nature.
Before this study there was limited knowledge available on the types of risks experienced specifically by social scientists studying social deviance or criminal behavior. Information discovered in this research provides knowledge and awareness of the potential for physical, legal, emotional, and personal/professional issues associated with such research. While the percentages were low for occurrence of physical/health and legal risks, these results hold serious implications for researchers. For example, even though rape, shooting, and being held hostage were low in occurrence, the prospect of experiencing such risks is frightening. Untrained researchers entering into dangerous settings are potentially more vulnerable to risk because of their inexperience. Dealing with these risks is compounded when a researcher is unaware of the possibility for such risk and is therefore unprepared for handling situations that may arise. Austin (2003) reports that as a graduate student he was not told of the possibility that fieldwork would be hazardous. He was left with the impression that it would be adventurous. When Austin first began exploring Filipino justice issues, he did not realize that his fieldwork would involve risk to his life. Inexperienced researchers are at greater risk because they are not able to anticipate the risk that awaits them (Paterson et al. 1999). Hopefully the knowledge gained from this study will assist those new to the adventures of research.
The third question utilized a number of variables to assist in developing an understanding of factors that impact the occurrence of risk. An interesting finding came from looking at gender. While it is suggested that gender plays a role in the experience of difficulties during research (Arendell 1997;Gurney 1985;Easterday et al. 1977), this study found no significant relationship between gender and the experience of risk. This was unexpected. It was especially surprising to look at the report for the risk of rape. One may expect rape to be primarily a female risk; however, this research indicates otherwise. While there were three reports for the risk of rape, two of those reports were made by male respondents.

Limitations
There are some issues that must be considered with interpretation of the results. It should be noted that generalizability of results for method of data collection, age of population under study, country for research, location of research, and research setting is limited due to lack of variance or low variance that exists within the variable categories. Therefore, interpretation of these results should be taken with caution.
Also, it should be recognized that researcher age and level of education were measured at present day instead of at time of risk experience. If these variables had been measured at time of risk instead of present day, the variables may have provided a different picture. In order to accurately determine such relationships, the variables should have been assessed for a measurement at the time of the experience of risk. Based on this information it was impossible to determine that these specific variables were predictors of risk or function as mediating factors. It is suggested that future research on this topic collect data for the respondents' age and level of education at the time the research project associated with risk actually took place.
Some limits are noted on ability to generalize to the greater population as generalizability is best applied to the population included in the study. However, because it was expected that many of the recruited population conduct research on social deviance or criminal behavior, the results can be generalized to other professionals outside of the organizations represented who also conduct research on social deviance or criminal behavior. Response rate was also a concern as non-response bias contributes to limits on generalizability.

Conclusions
This research was undertaken with exploration in mind. The overarching goal was to develop a sense for the types of risks that may be encountered while researching social deviance or criminal behavior. While types of risks were established, a true and intimate understanding of the nature of risks experienced was not fully realized. This limitation may be overcome by future research endeavors of this kind that also include in-depth interviews with individuals who have experienced risks.
While much information is taken from this study, it leads to more questions for future research. It would be interesting to include a question about knowledge of potential for risks before entering into a research project. This would provide insight into the level of awareness of risk that novice researchers may or may not hold. Future research could also consider whether or not the fear of risk prevents researchers from conducting research that they would otherwise engage.
While it is typically left to the researcher to prepare for risk they may face, an ethics committee or Institutional Review Board (IRB) could assist in this preparation. Some organizations do require that researchers follow safety guidelines; however, these guidelines are generally not specific for dealing with the populations or environments of crime and deviance. It is recognized that the primary focus of an IRB is on the risk for participants, but an IRB is in a unique position to provide the novice researcher with information and caution for their own safety.
The discoveries of this study adds to the knowledge base for those new to the practice of conducting research so that they do not enter research blindly. Through education, awareness can be improved. As awareness is improved, novice researchers may become more prepared for their research experience and survive it in a healthy manner.