Survey research is an important method of collecting data, particularly in the social sciences. Fowler (1993) gives three important usages for surveys: “…producing quantitative or numerical descriptions of some aspects of the study population…,” collecting information by asking people questions and analyzing their responses, and collecting information from a “…fraction of the population…” rather than the entire population (p. 1). There are different methods of distributing surveys: mail, telephone, Internet, or in-person interviews. All of these approaches can be used for survey research, and as with all types of research, there are ethical concerns that can surface. Researchers can play a role in minimizing ethical issues by adhering to the different standards that are set by the American Psychological Association (APA) and university ethics committees.
The purpose of this paper is to discuss three ethical principles established by the APA that can be applied to survey research: beneficence, respect, and justice (Knapp & VandeCreek, 2003). For the purpose of this paper, survey research will include research that is conducted using only surveys (e.g., questionnaires, checklists) as well as research that uses surveys as a component of the research.
Beneficence is defined as “maximizing good outcomes for science, humanity, and the individual research participants while avoiding or minimizing unnecessary risk, harm, or wrong” (Sieber, 1992, p. 18). There are survey research topics that deal with sensitive areas that can induce either participant distress or a positive emotional reaction (Daugherty, 1996). Helweg-Larsen and Bøving-Larsen (2003) examined the ethical issues that should be considered when using sensitive questions in youth surveys that deal with child sexual abuse (CSA). The results showed that women were extremely receptive to the survey and had a positive experience. Overall, less than 2% of the 1,000 participants reported some distress from taking the survey. Again, while 2% may seem like a small number of participants, 20 individuals reported some distress and that number should not be overlooked. The researchers concluded that ethics committees should assess the topics. In addition, mental health resources (e.g. clinician, educational pamphlets, etc.) should be mandatory and incorporated into the research budget for those participants who may need them if they were adversely affected by the sensitive topics in the survey, especially for the youth.
Black, Kresnow, Simon, Arias, and Shelley (2006) examined the appropriateness of asking participants sensitive questions that dealt with interpersonal violence using telephone surveys. The two surveys used in the study were the Second Injury Control and Risk Survey (ICARIS-2) and a pilot survey of sexual violence and intimate partner violence modules (SIPV) for the Behavioral Risk Factor Surveillance (BRFSS). Although both surveys contained questions about the participants’ own experiences concerning interpersonal violence, the SIPV was used to assess problems (i.e., participant distress) that can occur from the use of a telephone survey, especially when the surveyer asks sensitive questions.
In each of the surveys, after the participants were asked a question related to interpersonal violence, they were asked identical questions about their reactions to the sensitive questions (Black et al., 2006). The participants felt that questions about interpersonal violence were of some importance and did not find the survey too upsetting and also did not experience a high amount of fear from participating in the study. A large number of participants felt comfortable answering the sensitive questions; however, 11% (of the 1,690 participants) felt discomfort when answering the questions. This study demonstrates that researchers should prepare themselves for the different reactions they could receive from the participants and plan accordingly.
When questions that could cause distress cannot be avoided, the researchers have an obligation to assess for levels of distress and provide a means of reducing the distress. McKeown and Weed (2004) suggest having a mental health clinician present for immediate assistance, especially when using surveys that could invoke a strong response from participants who might cause harm to themselves or others. The feasibility of having a mental health clinician present during a research study may be difficult and is important to consider when designing a study.
For studies involving a lower risk of harm, providing a referral source may be appropriate. For example, Phillips (1994) explored asking sensitive questions of teens in surveys dealing with teen sex. In the article, Phillips discussed a study she conducted and some of the ethical problems she encountered during the study. Phillips was interested in teen attitudes and behaviors related to HIV/ AIDS prevention or, in other words, teen attitudes towards the use of condoms. Some of the students who participated voluntarily reported being victims of rape or incest during early childhood. This was not anticipated by the researchers because this information was not asked on the survey. Phillips did note that some of the wording on the survey could have invoked some emotional uneasiness in the participants and even the research staff who dealt with questions. Pamphlets published by the American Red Cross that had phone numbers of counseling services were handed out to the participants.
Langhinrichsen-Rohling, Arata, O’Brien, Bowers, and Klibert (2006) examined distress among high school and middle school students that was associated with answering a survey with questions about drug use, suicidal behavior, physical and sexual abuse. The study found that there was a small amount of students who noted that they experienced distress “often” when completing the survey (Langhinrichsen- Rohling et al., p. 440). The authors noted that while this amount may seem small, it is “not insignificant” (Langhinrichsen-Rohling et al., p. 440). The authors suggest that the adolescents who have experienced traumatic events could have a greater chance of being distressed by taking the survey.
Both of the studies mentioned demonstrate that it is necessary to assess the risks and benefits to participants as well as determine the factors that influence participant distress. The methodological choices that are made can impact participation rates and the participants’ willingness to disclose personal information. The goal of the researcher should ideally be to conduct research that is beneficial to the participants, adds scientific knowledge to the body of science, or if applicable, is beneficial to society as a whole (Kotch, 1997).
The term respect is defined as protecting “the dignity and worth of all people, and the rights of individuals to privacy, confidentiality, and self-determination… Special safeguards may be necessary to protect the rights and welfare of persons or communities whose vulnerabilities impair autonomous decision making” (Knapp & VandeCreek, 2003, p. 27). The scientific behaviors that are associated with respect are “having a valid research design, having a competent researcher, the identification of the consequences, appropriate selection of subjects, voluntary informed consent, and compensation for injury” (Sieber, 1992, p. 19). Goodare and Smith (1995) examined a study in which the patients’ rights seemed to be overlooked. The authors stressed that patients are an integral part of the research process, especially with clinical research trials, which cannot be done without the participation of patients. An example of this can be seen in the research study at the Bristol Cancer Help Centre which offered complementary treatment to women with breast cancer. While the study was being conducted, the women heard on the television news, along with anyone else who was watching, the results of the study that “those…who had been to the center were twice as likely to die and three times as likely to relapse as women who had not been to the center” (Goodare & Smith, p. 1277). The women never received letters they had been promised regarding publication of the results and were still under the assumption that the research was not finished, because they had only completed four of the five questionnaires. The women in the study filed a complaint with the Charity Commission (the group that funded the research) and the Commission recommended the establishment of guidelines for charities funding medical research.
Goodare and Smith (1995) also discussed methods by which patients can help develop the research agenda so the research can benefit the public and the patients. The National Health Service (NHS) research and development team in Britain was trying to achieve this goal by addressing the needs of the consumer. Polling a population, such as women with breast cancer, to determine what their needs are (Goodare & Smith) would be an example of including patients in developing the research agenda. After this type of polling is done, the participants in the study should be advised on the outcomes of the study and how it can affect them. Another point is that the experimental design should be thoroughly explained to the participant (Goodare & Smith). The authors stated that participants need to fully understand the risks and benefits of the study and make sure that all of the information is stated clearly.
The importance of the protection of participant rights to privacy and confidentiality has been well established within the psychologist-client relationship as well as the researcher-participant relationship (Botkin, 2001). However, Botkin noted that the consideration of rights through family and social relationships has not been critically examined in the research ethics literature. A case that examines this situation took place at Virginia Commonwealth University (VCU). The participants in the study were contacted by the researcher to provide health information about themselves and their family members. Informed consent was obtained through the primary participant, enabling researchers to acquire such information, but consent was not obtained from family members. The case is described below:
The VCU case involved an adult woman who was being recruited to participate in a twin study. Her father read a mailed survey instrument that included questions about the health of her parents and other family members. The questionnaire reportedly asked, among many other items, whether the subject’s father suffered from depression and whether he had abnormal genitalia. The father was concerned that [his daughter] providing this information constituted a threat to [his] personal and family privacy and that informed consent should have been sought from family members (p. 207).
This case demonstrates that special considerations need to be taken in unique situations dealing with disclosing personal information about family members or social contacts. The outcome in this case resulted in the suspension of human subject research at VCU by the Office for Human Research Protections (OHRP) and the Food and Drug Administration (FDA). This case can remind researchers that it is important for them to assess these complex issues and pay careful attention in order to protect all of the research participants who are involved.
The ethical principle of justice refers to “fairness in distribution” of the risks and the benefits of research (National Institutes of Health [NIH], 1979). Also, justice ensures “reasonable, nonexploitative, and carefully considered procedures and their fair administration” (Sieber, 1992, p. 18). According to the Belmont Report, there are three populations that researchers should take into account with this principle: ethnic minorities, women, and children (NIH, 1979). For the purpose of this paper, children under the age of 18 years will be the focus. Fisher (2003) examined adolescent and parent perspectives on ethical issues in a youth drug use and suicide survey. Fisher stated that “engaging teenagers and parents in a dialogue about the ethical challenges of survey research can also enhance their support and participation, ensuring that research procedures reflect participants’ values and merit their trust” (p. 304). Also, by understanding adolescent and parent perspectives, these four issues can be enhanced: (a) risks and benefits, (b) obtaining parental consent and adolescent assent, (c) confidentiality policies, and (d) the use of cash incentives.
The use of cash incentives can create a dilemma when conducting a survey research study on drug use. Ideally, offering cash incentives “should secure a balanced representation of individuals from all economic levels” (Fisher, 2003, p. 307). Depending on one’s point of view, the use of cash incentives can be seen as either coercive or fair. But, when used in a study such as this (i.e., assessing drug use), it is unknown if the adolescent is going to purchase illegal drugs with the money received from the study. Another setback is if adolescents have the notion that if they participate in studies such as this, they will receive some type of compensation that could reinforce negative behaviors (if the adolescent is engaging in drug use; Fisher). Receiving the opinions of adolescents and parents regarding the nature of cash payments can aid in determining an ethical use of incentives in survey research.
By allowing parents to partner with their teens to make decisions about issues concerning risks and benefits, confidentiality, consent, and participant recruitment, trust can develop between the researcher and the families participating in the studies (Fisher, 2003). In terms of risks and benefits, both the parents and adolescents agreed that these types of surveys could “improve public policies aimed at drug prevention” and also assist in suicide prevention (Fisher, p. 325). In terms of confidentiality, both parents and adolescents agreed that the researcher has an obligation to help the youth who are contemplating suicide. On the other hand, the parents and teens thought that a disclosure policy would inhibit individuals from completing honest questionnaires. While there is some ambiguity in this area, both parents and teens believed that it would be important for researchers to ask for permission from the individuals before referring them to mental health services.
In terms of consent, the parents and teens stated that teens would provide honest answers if their parents were not aware that they were participating in a study (Fisher, 2003). This creates an ethical dilemma because researchers have an obligation to obtain parental consent from participants who are under 18 years of age. It is important that parents and teens consult with each other before consenting to a study so both parties have a role in the decision-making process (enhancing the parent-child relationship). In terms of participant incentives, both parents and teens thought the use of a cash incentive was not coercive and was “… fair reimbursement for survey participation…” (Fisher, p. 328). Socioeconomic status was considered and did not affect parent and teen perceptions towards the use of cash incentives in survey research. In short, the perspectives of adolescents and parents can play a valuable role in creating ethical surveys for a special population (i.e., youth) that concern drug use and suicidal behavior.
Rather than generalizing the results of studies assessing drug use and suicidal behaviors with adult populations to adolescents, it is important to conduct research in this population so that adolescents are able to benefi t from the research as well (the justice principle). But when dealing with adolescents, precautions need to be taken in order to ensure that their well-being is protected. Incorporating adolescents and parents into the ethical planning of surveys will not only enhance the value of survey research, it will enhance the value for youth, behavioral science, and public policy (Fisher, 2003).
The scope of this paper focused on three ethical principles established by the American Psychological Association that deal with survey research: beneficence, respect, and justice.
The articles used in this literature review describe each principle in order to build a general understanding and to provide a framework to guide the resolutions for ethical problems. The descriptions will hopefully help graduate students and psychologists decipher and interpret situations that they may encounter while conducting survey research. By drawing attention to ethical issues that can arise, the hope is that this will encourage discussion among colleagues to assess troubling ethical questions as well as assure that survey research will be carried out in an ethical manner. If psychology researchers and practitioners alike engage in this process, the resulting discussions will contribute to the production of research that has substantial social benefits.