Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 May 7;34(3):e70048. doi: 10.1002/jgc4.70048

Preventing fraudulent research participation: Methodological strategies and ethical impacts

Cassie Mayer 1, Rebecca Tryon 2,3, Sarah Ricks 4,5, Kara Lane 3, Emma Van Hook 1,6, Kimberly Zayhowski 3,7,✉
PMCID: PMC12057331  PMID: 40331685

Abstract

This article addresses a growing issue in genetic counseling research: the participation of individuals who falsify demographic information or experiences to qualify for studies, often motivated by remuneration. This issue is particularly pressing in studies using social media recruitment, where detecting fraudulent participants has become increasingly difficult. The prevalence of inaccurate data raises serious questions about the integrity, reliability, and validity of research findings. We explore potential sources of participant fraud and inconsistency informed by prior, direct experience with fraudulent participants, discussing various strategies to strengthen participant screening, data validation, and interview protocols. We also examine the challenges screening methods pose for marginalized communities, who may already harbor distrust in research or face privacy concerns. We emphasize the need for transparent, responsible approaches to participant verification and advocate for developing standardized guidelines to protect both research quality and participant rights. By promoting trust, inclusivity, and fairness in research practices, we aim to preserve the dignity of participants and ensure that genetic counseling research remains rigorous. Ultimately, we call for collective action to reinforce the integrity of research and enhance its impact on clinical practice, ensuring that evidence‐based insights guide patient‐centered care.

Keywords: data accuracy, fraudulent participants, genetic counseling, online recruitment, qualitative research, quantitative research, research ethics

What is known about the topic

The genetic counseling community widely recognizes the importance of research integrity, given that the reliability of data depends on the genuine contributions of participants. Authenticity is crucial not only for preserving the integrity of the research process but also for effectively informing clinical practices in genetic counseling. Ensuring that insights and evidence from studies accurately reflect the experiences of participants is essential for optimizing patient care and counseling strategies.

What this paper adds to the topic

This article addresses the pressing issue of fraudulent participation in genetic counseling research and advocates for improved participant screening protocols. It highlights the necessity for clear guidance and research standards to uphold data integrity while minimizing potential harm to marginalized communities. By contributing to the ongoing discourse in the field, we aim to promote ethical research practices that ensure the representation and inclusion of a diversity of voices in genetic counseling research.

1. INTRODUCTION

The integrity of genetic counseling research depends on our ability to recruit genuine participants and be entrusted with their authentic experiences. Through the rise of online recruitment methods, genetic counseling researchers have an increasing ability to engage with communities that can be otherwise difficult to reach due to small population size or the significantly higher demands of in‐person research. The same online recruitment methods that have allowed for broader reach and increased participation have, unfortunately, also opened the door to exploitation, and researchers are increasingly confronting falsified participation for the presumed purpose of remuneration.

The authors of this article are all genetic counselors with direct experience addressing a range of fraudulent tactics that emerged during quantitative, qualitative, and mixed‐methods research projects. Our projects were variously impacted by falsified survey data, imposter interview participants, and spam bots, and the challenges we faced underscored the need for robust safeguards to protect the integrity of our research data. While these phenomena and suggested precautions have been documented in other disciplines (Bybee et al., 2022; Goodrich et al., 2023; Jones et al., 2021; Pellicano et al., 2024; Roehl & Harland, 2022; Schneider et al., 2024), only a handful of articles published in the Journal of Genetic Counseling have explicitly discussed fraudulent participation (Chang et al., 2024; Gallion et al., 2023; Jagannathan et al., 2024; Pasca et al., 2021; Pederson et al., 2024; Sanchez et al., 2024). We suspect the lack of discussion could reflect both a recent increase in fraudulent participation in genetic counseling research and a potential reticence toward publishing data from affected projects. The growing issue of fraudulent participation threatens the reliability of research data and, consequently, the credibility of findings meant to inform clinical practice and policy decisions in genetic counseling.

Through our own research, we each developed response protocols, which largely included manual data cleaning and enhanced screening of participants. In developing our protocols, we recognized that many screening methods may add additional burden or inadvertently screen out participants from already underrepresented and marginalized communities, which called into question the equity of such measures. Mitigating the impact of falsified participants added a tremendous logistical and emotional burden to each project, especially for those of us who were genetic counseling students embarking on our first research projects. As emerging professionals, several of us on the author team began our research projects with enthusiasm and a commitment to contributing meaningful insights to the field. We were discouraged and distressed when we encountered issues related to fraudulent research participation, which undermined our confidence in our ability to conduct research effectively.

Drawing on both our experiences and existing literature, we aim to shed light on the growing issue of false participation in genetic counseling research. We offer practical, experience‐based strategies for improving recruitment practices and safeguarding data quality, while maintaining ethical standards. As the challenge of fraudulent participants continues to evolve and we grapple with the complexities it presents, we recognize that firm solutions remain elusive. We hope to contribute to the dialogue surrounding practices and innovative approaches to safeguard the authenticity of our research participants. We have included checklists summarizing key points from this article for research groups to consider as strategies to mitigate fraudulent participants, available in Data S1.

2. RISKS OF FURTHER MARGINALIZATION FOR ALREADY MINORITIZED COMMUNITIES

Efforts to address fraudulent participation in genetic counseling research, while essential for ensuring data integrity, can perpetuate harm to already marginalized communities. Although enhanced participant screening aims to filter out bots and fraudulent respondents, the screening methods we use can create barriers that disproportionately affect marginalized groups, leading to further underrepresentation. Additionally, individuals from communities with historical experiences of eugenics and genetic discrimination may already approach genetic research with caution (Angelo et al., 2022; Scanlon et al., 2021). The introduction of screening measures that feel invasive or exclusionary could deepen distrust and further deter participation in future studies.

In the context of genetic counseling, the consequences of false participation in studies on rare genetic conditions extend beyond academic integrity. Fraudulent data can distort our understanding of genetic conditions, resulting in misguided clinical practices that negatively impact diagnosis, treatment, and care. For individuals living with rare genetic conditions who already have limited representation in research, the inclusion of false data can obscure the nuances of their experiences and further marginalize the voices that research is meant to amplify.

Researchers must develop strategies that improve participant screening accuracy while ensuring that real participants—especially those from minoritized backgrounds—are not unfairly excluded. Addressing the challenge of fraudulent participants demands that researchers create standards and ethical guidelines that balance data integrity with inclusivity, allowing genetic counseling research to continue serving both scientific progress and social equity. When researchers are compelled to question the validity of participants' responses, trust between researchers and communities can erode. Participant skepticism not only jeopardizes the reliability of the data but also risks alienating genuine participants who may feel unfairly scrutinized or misrepresented. Distrust could further diminish engagement in future research, particularly among marginalized communities already wary of academic inquiry (Drysdale et al., 2023).

Throughout this article, we propose steps to mitigate fraudulent participation in research, but we also explore the potential pitfalls of the proposed methods. Nearly all approaches could inadvertently marginalize already underrepresented communities, highlighting the delicate balance required to safeguard both data integrity and inclusivity.

3. RECRUITMENT AND COMPENSATION

Steps to mitigate fraudulent participation begin with strategic recruitment protocols that balance risk reduction with accessibility to the target population. Researchers who recruit via online channels may variably utilize websites, registries, listservs, organization newsletters, and social media platforms (Bybee et al., 2022; Jagannathan et al., 2024; Sanchez et al., 2024). The reach of social media recruitment is predictably variable due to the diversity of social media itself; recruitment materials posted to closed or moderated social media groups may have more limited initial reach than materials posted on publicly available forums (Bybee et al., 2022; Goodrich et al., 2023; Jagannathan et al., 2024; Schneider et al., 2024). While the expanded reach facilitated by public social media platforms may be attractive to research teams aiming to increase the size and diversity of their participant cohorts, uncontrolled public distribution also increases the risk of fraudulent participation (Goodrich et al., 2023; Schneider et al., 2024). Several authors on our team identified increases in suspicious survey activity after their recruitment materials were shared—purposely or by a well‐meaning third party—to publicly accessible Facebook, LinkedIn, or Twitter pages. Closed groups, such as email lists or invite‐only Facebook pages relevant to the target population, are generally less susceptible to fraudulent participation (Goodrich et al., 2023; Pratt‐Chapman et al., 2021; Schneider et al., 2024). Similarly, “snowball sampling,” wherein partner organizations and known authentic participants distribute recruitment materials to other known individuals, is also a method of avoiding fraudulent participation (Salinas, 2023).

Like the other methodologies discussed in this article, targeting the distribution of recruitment materials is insufficient as an isolated strategy to prevent fraudulent research participation. As an additional safeguard, research teams could generate a unique survey link for each recruitment avenue, thereby allowing quick identification and deactivation of compromised links (Goodrich et al., 2023; Schneider et al., 2024). One member of our authorship team had success in activating a Qualtrics setting to require a password to access the survey instrument. As noted in the literature, a password may only provide protection as long as it is not shared on social media (Goodrich et al., 2023). Our author shared the password by including it in a non‐clickable image used for recruitment advertisement, so participants were required to read and type, which may have been a more difficult task for bots that are designed only to identify text‐based links. While not feasible for all study designs, unique survey links for each participant, distributed only after eligibility requirements have been verified, could provide an alternate avenue of protection (Jones et al., 2021; Schneider et al., 2024). Conducting screening phone calls with participants, corresponding several times through email, or requiring documentation of key criteria prior to distributing the survey link or issuing an interview invitation gives investigators the ability to control access to their data collection tools and limit fraudulent participation in qualitative research approaches (Pellicano et al., 2024; Roehl & Harland, 2022; Schneider et al., 2024). Similarly, providing only general inclusion criteria on study advertisements—for example, people with condition X, rather than women with condition X between the ages of 18–30 who have had genetic testing—may prevent fraudulent participants from easily mimicking criteria and bypassing screening and eligibility questions (Salinas, 2023). Although some literature recommends screening participants through social media accounts or internet searches to verify identity, we do not recommend this approach as it could introduce privacy concerns and may inadvertently eliminate potential participants who do not have a robust online presence (Dike et al., 2019).

Much like online recruitment protocols, one intention behind providing financial compensation to research participants was to ease the burden of research participation on vulnerable communities (Gelinas et al., 2020). Unfortunately, it is likely that financial incentives simultaneously increase the risk of research projects being targeted by bots, imposter participants, and duplicate responders (Griffin et al., 2022; Jones et al., 2021). Some literature suggests that information about incentives and their value should be avoided within recruitment materials, which as a result would avoid triggering bots that are designed to detect specific terms like ‘gift card’ (Bybee et al., 2022; Griffin et al., 2022; Pellicano et al., 2024; Schneider et al., 2024). Providing indirect incentives to participants, such as charitable donations, may increase interest in participation without providing direct financial compensation to research participants (Bybee et al., 2022; Goodrich et al., 2023; Griffin et al., 2022; Pellicano et al., 2024). Other literature suggests offering financial compensation to a subset of participants via random lottery (Goodrich et al., 2023; Griffin et al., 2022; Teitcher et al., 2015), although it should be noted that some members of our author team used this methodology and their projects nevertheless experienced fraudulent participation. Alternatively, researchers could require a physical mailing address rather than distributing compensation via email, which may not eliminate fraudulent participation, but could avoid compensation of fraudulent participants (Bybee et al., 2022; Pellicano et al., 2024; Schneider et al., 2024). However, participants who prioritize compensation in a specific form may do so out of necessity, not deception. The focus on compensation could make a research team unfairly suspicious toward participants relying on study income, such as individuals from lower socioeconomic backgrounds (Ridge et al., 2023).

Before deploying any of the above tactics to reduce the risk of fraudulent participation, researchers must evaluate the inadvertent potential harms against authentic participants. For example, while requiring a genetic test result as documentation of a diagnosis may decrease the likelihood of fraudulent participation, requiring this sensitive information would also increase possible risks to participant privacy and confidentiality, if this information is not needed for other study purposes. Participants whose diagnosis is clinically known, but not genetically confirmed, would also be excluded. Not only could some of the above methodologies for protected advertising and participant verification introduce recruitment bias, legitimate participants—especially those already distrustful of academic research—could experience increased barriers to participation, and therefore study attrition may increase (Schneider et al., 2024). Likewise, while we as researchers do not want our limited research budgets to go to fraudulent participants, we must recognize the time and effort given to us by those who participate in research, and financial compensation is effective in doing so (Jones et al., 2021; Pellicano et al., 2024). While moving toward alternate compensation structures or requiring mailing addresses for compensation could protect our research budgets, such strategies could disproportionately discourage people with less economic means or people who are unhoused from research participation (Pellicano et al., 2024).

Researchers must carefully consider recruitment protocols and transparency throughout the research process to foster the inclusion of diverse participants. When designing research protocols and consent documents, researchers should include straightforward information about what will be collected, how it will be used, what steps researchers will take to detect fraudulent participation, and under which conditions compensation may be withheld (Bybee et al., 2022; Pellicano et al., 2024; Roehl & Harland, 2022; Schneider et al., 2024). Research teams should collaborate with relevant ethics and institutional review boards to discuss proposed methods to mitigate or minimize fraudulent participation, along with the associated benefits and limitations of these strategies (Schneider et al., 2024).

4. SURVEY TECHNOLOGIES

Researchers have access to a variety of software tools that are intended to facilitate easier detection of bots and fraudulent participants in survey data. CAPTCHA verification, for example, is intended as a screening tool that should be easily solvable by humans and difficult for bots to overcome (Rodriguez & Oppenheimer, 2024; Searles et al., 2023; Sivakorn et al., 2016). Several authors on our team opted to use CAPTCHA v2, the version available in Qualtrics at the time of their studies, at the beginning of their surveys and nevertheless found their studies overcome with fraudulent responses, which is an experience that has been reported elsewhere in the literature (Bybee et al., 2022; Pederson et al., 2024; Pratt‐Chapman et al., 2021). Other research teams only deployed CAPTCHA verification in a second round of recruitment after their first was affected by fraudulent participation, with varying success at preventing further fraudulent responses (Bybee et al., 2022; Levi et al., 2022; Salinas, 2023).

Searles et al. offer a potential explanation: advances in computer vision and machine learning, coupled with the rise of ‘CAPTCHA farms,’ where people are paid to solve CAPTCHA tests, have rendered obsolete much of the technology used in early versions of CAPTCHA technology (Searles et al., 2023). Similarly, more recent versions of CAPTCHA that analyze data such as browser information, mouse and keyboard actions and execution time, can be fooled by bots that are able to mimic such information (Sivakorn et al., 2016). As a result, many researchers consider CAPTCHA verification to be an insufficient tool for screening out fraudulent participants and recommend that it should be used in conjunction with other technologies and manual checks (Goodrich et al., 2023; Pratt‐Chapman et al., 2021; Schneider et al., 2024).

In addition to CAPTCHA verification, which is widely available, some survey software has options to deploy settings such as preventing multiple submissions from the same IP address (sometimes known as “ballot box stuffing”), requiring passwords to access the survey, specifying a web address from which all participants must have been directed in order to access the survey, and implementing proprietary bot detection scoring. Several of our authors used a variety of these methods, which were met with varying success. Unfortunately, much like CAPTCHA verification, IP addresses can be easily spoofed by bots to appear as if each response has a unique origin (Bowen et al., 2008). The other technologies may only make it more difficult, not impossible, for bots and fraudulent actors to complete research surveys (Teitcher et al., 2015).

As researchers consider which technologies to use in their efforts to protect their projects from fraudulent participation, we must equally consider the impact of these technologies on authentic participants. For example, the visual nature of some CAPTCHA verifications could be prohibitive to participants with vision impairments (Gaggi, 2022; Moreno et al., 2014). Preventing multiple submissions from the same IP address could prevent people who share living or working spaces, or people who access the internet in public spaces like libraries from participating (Teitcher et al., 2015). Additionally, people with a lower ability to navigate technology or cognitive disabilities may similarly be dissuaded from completing surveys if faced with increased technological barriers (Pratt‐Chapman et al., 2021).

5. SURVEY DESIGN

Anti‐fraud recruitment protocols and survey technologies aim to prevent fraudulent participants from accessing research surveys, but they remain imperfect. Once fraudulent participants access a survey, thoughtfully designed survey instruments can aid in highlighting differences between authentic and fraudulent respondents. Common suggestions include evaluating responses to open‐ended questions, including paired questions, inserting attention check questions, and developing specific questions that demonstrate institutional knowledge. The richer data from open‐ended responses provides an opportunity to screen out responses that are nonsensical, irrelevant, or verbatim to other responses (Goodrich et al., 2023; Pozzar et al., 2020; Pratt‐Chapman et al., 2021; Schneider et al., 2024). Including a requirement for responses to all questions, along with a minimum character count for free‐text responses, can help deter bots and fraudulent participants, though it may increase the burden on participants (Bybee et al., 2022; Pellicano et al., 2024). Paired questions, or questions that ask for the same information in different formats, allow researchers to check for consistency within responses (Burnette et al., 2022; Goodrich et al., 2023; Griffin et al., 2022; Levi et al., 2022; Pratt‐Chapman et al., 2021; Salinas, 2023). For example, researchers could reasonably expect concordance if participants are asked both their age and their year of birth. Attention‐check questions serve a similar purpose: they give the participants instructions such as ‘select the fourth option,’ enabling researchers to eliminate responses that do not follow instructions (Burnette et al., 2022; Bybee et al., 2022; Gallion et al., 2023; Pasca et al., 2021; Pratt‐Chapman et al., 2021). Pederson et al. (2024) combined the idea of paired questions with attention check questions and at different points in their survey asked participants, “Which of the following fruits do you like most?” followed by “What is the color of the fruit you previously selected?” (Pederson et al., 2024). Finally, institutional knowledge questions query information that is easily answered by an authentic member of the target population, but would be difficult for a fraudulent participant (Goodrich et al., 2023). An example in a survey intended for genetic counselors could be, ‘Describe what you include when contracting with a patient.’ Institutional knowledge questions can alternatively employ low probability multiple choice questions, such as ‘Where did you hear of this survey?,’ with several false answers included as options, such that a randomly selected answer would have a low probability of being correct (Goodrich et al., 2023; Pratt‐Chapman et al., 2021; Schneider et al., 2024).

Using one or more of the above question styles may make fraudulent involvement in surveys easier to exclude from data analysis, yet they are all accompanied by pitfalls. For example, if researchers flag responses with grammatical errors or unexpected information as fraudulent, they inadvertently screen out non‐native English speakers, people with lower literacy levels, or people unfamiliar with digital technology and fail to detect bots that use more sophisticated language models (Schneider et al., 2024). Additionally, we risk increasing respondent fatigue with each additional question we add for screening purposes (Goodrich et al., 2023), although piloting surveys before deployment may provide insight into the appropriate balance of screening versus content questions (Salinas, 2023).

6. SURVEY DATA CLEANING

Reviewing data on the backend to assess which responses are valid for inclusion is an important and complex process. Unfortunately, it is challenging to introduce automation into backend data cleaning, given the complexity of research data and parameters that are available to evaluate, and therefore data cleaning is inherently manual. Our authors developed processes to remove as many fraudulent responses as possible without excluding authentic responses. To limit the need for judgment calls, especially considering that authentic participants can also make mistakes, the literature suggests determining a threshold of suspicion with multiple factors considered, after which a respondent will be excluded (Gallion et al., 2023; Goodrich et al., 2023).

In addition to integrating screening questions into the survey, research teams can consider survey metadata in aggregate and in a systematic order to best identify and exclude fraudulent responses (Storozuk et al., 2020). Survey platforms often collect the time taken to complete the survey, and some research teams choose to exclude surveys that are completed unusually quickly (Arevalo et al., 2022; Pasca et al., 2021; Pederson et al., 2024; Pozzar et al., 2020). The ability to use time information may vary. While one author found time metrics to be unreliable within her data set, another author was able to exclude responses that took 50% less time to complete than her quickest pilot survey. Importantly, eliminating responses that took significantly longer than the pilot time is an unreliable metric, as it could exclude participants with slower reading abilities or those who left their browser window open while completing the survey.

Survey software may also collect IP addresses, and some research teams create protocols for excluding responses based on repeat or consecutive IP addresses, information about a respondent's location, or their internet service provider (Schneider et al., 2024; Teitcher et al., 2015). This type of information must be used thoughtfully to prevent excluding authentic participants who may have unusual IP address data due to travel outside their typical area of residence, use of a secure virtual personal network, or other legitimate reasons. For example, one author received several responses from the same IP address that were later identified to be several unique clinicians sharing the same workstation.

Researchers can use patterns within the complete data to evaluate the validity of individual responses. While there is sometimes a natural spike in the response rate directly following survey launch, it would be unusual to see additional waves of responses that were not subsequent to additional dissemination or recruitment events. It would also be suspicious for a participant to select the same response for every question, especially across variables that seem unlikely to truly align (Pozzar et al., 2020). This response pattern, called ‘straight lining,’ can be particularly evident across Likert scale questions that occasionally flip the pattern of phrasing. Moreover, manually inspecting responses for identical open‐ended responses, including content that may have been copied from online resources or created by generative artificial intelligence, can help identify potential fraud (Pozzar et al., 2020). Several members of our author group noticed unusual patterns in respondent's email addresses including a high number of identically configured addresses from the same email provider, uncommon account providers, and non‐trivial mismatches between the provided name and email address, the majority of which eventually screened out as fraudulent based on other factors (Chang et al., 2024; Pellicano et al., 2024; Pratt‐Chapman et al., 2021).

Stricter screening protocols, such as enhanced bot detection methods or more complex question patterns, often target non‐standard behaviors or inconsistencies in responses. However, strict measures can exclude participants who are authentic but may struggle with certain aspects of the research process. Depending on the target population, additional factors could be considered to provide a more holistic view of fraudulent versus authentic responses. For example, some of our authors surveying rare disease patient populations found their high response rates unusual given the small number of eligible individuals in the target community. Our authors also noted that a disproportionate number of their fraudulent responses selected underrepresented racial identities, likely because fraudulent participants were selecting the first of the alphabetically‐listed options (such as “African American and/or Black”). However, the use of demographic information as a screen can be fraught and cast undue suspicion on authentic participants with marginalized identities less represented in research, versus a fraudulent participant who claims to be white, which is an overrepresented demographic in research (Scanlon et al., 2021). Without careful consideration, the criteria used for exclusion could not only undermine efforts to promote inclusive research but also perpetuate existing disparities in the representation of diverse populations in genetic counseling studies. For communities already underrepresented in genetic research—due to systemic barriers, historical distrust, or socioeconomic constraints—such exclusion exacerbates their marginalization.

7. INTERVIEW DATA COLLECTION AND ANALYSIS

The format and depth of interview data require additional considerations for addressing fraudulent participation. Researchers should carefully consider methodologies intended to protect both quantitative and qualitative data within the context of each research study and the potential disadvantages for both researchers and legitimate participants.

Active listening during interviews is perhaps the most crucial strategy for identifying fraudulent qualitative research participants. Interviewers should be attentive to contradictions in participants' responses, ambiguous or generic responses, unusually similar responses and mannerisms across interviews, and reported recruitment sources discordant with the strategies chosen by the research team, all of which may signal fraudulent participation (Chang et al., 2024; Pellicano et al., 2024; Ridge et al., 2023; Roehl & Harland, 2022). Some research teams choose to begin interviews by re‐asking demographic or inclusion criteria questions from earlier eligibility questionnaires with the goal of identifying such inconsistencies (Roehl & Harland, 2022). Fraudulent participants are less likely to recall fabricated details, which can lead them to providing information that could be flagged by the interviewer (Roehl & Harland, 2022). An alternative approach could involve re‐asking eligibility questions during the interview, but concealing the participant's responses on the eligibility questionnaire from the interviewer. This approach maintains the opportunity for post‐interview exclusion based on inconsistencies but limits the bias that overt suspicion could exert on an interview.

Active listening can help researchers note subtle patterns and prompt further inquiry to clarify vague responses, uncertainties, and inconsistencies through the use of open‐ended questions, pointed probes, or challenge statements (Roehl & Harland, 2022). While an experienced interviewer might feel comfortable directly confronting suspicious interview subjects, novice researchers might find themselves uncomfortable or under‐equipped to do so. For researchers who may be inexperienced in interviewing, including a second interviewer when feasible could provide additional support. Research teams could also pre‐define and create language regarding circumstances when they might choose to challenge suspected fraudulent participants, when they might end interviews due to suspected fraud, and whether suspected fraudulent participants will be compensated. Concerns regarding fraudulent participation and data integrity should be discussed at regular research team meetings (Pellicano et al., 2024; Ridge et al., 2023).

To aid in detecting suspicious patterns across interviews, researchers can code their data to highlight responses that deviate from expected patterns, which facilitates assessing the suspicious data as a collective (Roehl & Harland, 2022). To ensure thorough documentation, interviewers should maintain a reflexive journal throughout the data collection process where they can record their observations and reflections on the data, including their perceptions of participants' trustworthiness (Pellicano et al., 2024; Roehl & Harland, 2022). Reflexive journaling can thereby help researchers recognize patterns that may be indicative of fraudulent respondents, while also clarifying their criteria for assessing participant trustworthiness and elucidating biases that could negatively impact these efforts.

In addition to evaluating the interview content, some research teams encourage or require participants to have their cameras active during part or all of the interview process if conducted via video platforms (Pellicano et al., 2024; Roehl & Harland, 2022; Schneider et al., 2024). Even requiring participants to turn on their camera momentarily before interview recording begins can promote accountability, enhance the investigator's ability to detect participants attempting to interview multiple times, and allow observation of non‐verbal cues, facilitating more genuine interactions and enhanced rapport between interviewers and participants (Pellicano et al., 2024; Roehl & Harland, 2022; Schneider et al., 2024).

In creating their fraud mitigation protocols for qualitative research, interviewers should seek to maintain trust with authentic participants. The methods we discuss in this article risk overlooking genuine responses if the criteria for exclusion are overly rigid or not inclusive enough of diverse participant perspectives. The significant subjectivity and intuitive nature of some methods risk allowing the researchers' own biases to exclude legitimate participants with marginalized identities more readily than a universally applied process described in other sections. Preconceived beliefs and expectations, as well as inflexible systems for fraud mitigation, may further marginalize voices that do not fit conventional patterns, undermining the goal of exploration and inclusivity in research. Asking repetitive questions or directly challenging inconsistencies risks feeling redundant or intrusive to authentic participants, potentially affecting rapport and openness during interviews, especially for individuals who already feel scrutinized or distrustful of research. There may also be legitimate reasons for gaps in a participant's knowledge. For example, an individual who uses “Hurler syndrome” generically without realizing it is a phenotypically specific term may describe a more mild phenotype of mucopolysaccharidosis type I and could be erroneously disqualified for describing the ‘wrong’ symptoms. However, researchers could perhaps reasonably exclude participants who are only able to describe mucopolysaccharidosis type I as a condition that makes someone “tired.” A strength of qualitative interviews compared to quantitative surveys is the ability of researchers to clarify ambiguities with participants in real time, which can provide significant insight into the differences between a legitimate participant with knowledge gaps and a fraudulent participant without relevant life experiences.

Similarly, while requiring camera activity could deter fraudulent participation, it would also reduce privacy and create discomfort or anxiety for authentic participants and may require justification to institutional review boards (Pellicano et al., 2024; Roehl & Harland, 2022). Discomfort prompted by video requirements could discourage interview participation, particularly for individuals who are concerned about anonymity in research, feel self‐conscious about their appearance or home environment, or lack access to adequate technology. While researchers could build trust with research participants by transparently explaining that video requirements aim to limit fraudulent participation, such transparency also risks alienating participants by implying a sense of distrust.

Researchers must remember that similar phrasing, mannerisms, and responses can indicate shared cultural backgrounds or socioeconomic experiences among legitimate participants, and that misinterpreting patterns in responses may lead to unfair scrutiny of participant narratives. It is also possible that participants concerned about confidentiality might intentionally misrepresent their identities to protect themselves, while still providing authentic interview responses. For instance, a transgender person might alter certain demographic details to avoid exposure in states with anti‐transgender policies. Overall, an overzealous focus on identifying inconsistencies may lead interviewers to adopt a more adversarial stance while facilitating interviews. If our default perspective as researchers is distrustful of participants, that could in itself discourage authentic responses, particularly from individuals who have historical reasons for being wary of research.

8. CONCLUSIONS

The challenge of fraudulent participation in genetic counseling research poses a significant threat to the integrity of our findings and overall trust in the research process. Frequently driven by financial incentives, fraudulent participants are increasingly engaging in studies, particularly through online recruitment methods. While implementing robust strategies—such as targeted recruitment through professional organizations and meticulous vetting of participants—is critical, we must remain vigilant to ensure that efforts to counteract fraud do not introduce bias and discrimination that further marginalize communities underrepresented in research. Trust is fundamental in genetic counseling, especially for individuals affected by rare genetic conditions or communities with an earned distrust of research and practice. To address these complexities, we advocate for collaborative initiatives involving researchers, ethicists, and community members to develop clear guidelines and ethical standards that balance stringent participant screening with the necessity of including a diversity of voices in our research. By employing thoughtful strategies—such as ethical recruitment practices, transparent communication regarding data collection, and multifaceted data cleaning—we can safeguard participants' rights and dignity while enhancing the quality and reliability of our findings. Ultimately, promoting a research environment that harmonizes data integrity with ethical recruitment practices is essential for fortifying the foundations of genetic counseling research, enriching our understanding of the lived experiences of patients, and improving clinical practices for all individuals and families involved.

AUTHOR CONTRIBUTIONS

Cassie Mayer: Conceptualization; writing – original draft; writing – review and editing. Rebecca Tryon: Conceptualization; supervision; writing – original draft; writing – review and editing. Emma Van Hook: Conceptualization; writing – original draft; writing – review and editing. Kara Lane: Conceptualization; writing – original draft; writing – review and editing. Sarah Ricks: Conceptualization; writing – original draft; writing – review and editing. Kimberly Zayhowski: Conceptualization; supervision; writing – original draft; writing – review and editing.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Supporting information

Data S1.

JGC4-34-0-s001.pdf (96.7KB, pdf)

Mayer, C. , Tryon, R. , Ricks, S. , Lane, K. , Van Hook, E. , & Zayhowski, K. (2025). Preventing fraudulent research participation: Methodological strategies and ethical impacts. Journal of Genetic Counseling, 34, e70048. 10.1002/jgc4.70048

REFERENCES

  1. Angelo, F. , Veenstra, D. , Knerr, S. , & Devine, B. (2022). Prevalence and prediction of medical distrust in a diverse medical genomic research sample. Genetics in Medicine, 24(7), 1459–1467. 10.1016/j.gim.2022.03.007 [DOI] [PubMed] [Google Scholar]
  2. Arevalo, M. , Brownstein, N. C. , Whiting, J. , Meade, C. D. , Gwede, C. K. , Vadaparampil, S. T. , Tillery, K. J. , Islam, J. Y. , Giuliano, A. R. , & Christy, S. M. (2022). Strategies and lessons learned during cleaning of data from research panel participants: Cross‐sectional web‐based health behavior survey study. JMIR Formative Research, 6(6), e35797. 10.2196/35797 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bowen, A. M. , Daniel, C. M. , Williams, M. L. , & Baird, G. L. (2008). Identifying multiple submissions in internet research: Preserving data integrity. AIDS and Behavior, 12(6), 964–973. 10.1007/s10461-007-9352-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Burnette, C. B. , Luzier, J. , Bennett, B. L. , Weisenmuller, C. , Kerr, P. , Martin, S. , Keener, J. , & Calderwood, L. (2022). Concerns and recommendations for using Amazon MTurk for eating disorder research. The International Journal of Eating Disorders, 55(2), 263–272. 10.1002/eat.23614 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bybee, S. , Cloyes, K. , Ellington, L. , Baucom, B. , Supiano, K. , & Mooney, K. (2022). Bots and nots: Safeguarding online survey research with underrepresented and diverse populations. Psychology and Sexuality, 13(4), 901–911. 10.1080/19419899.2021.1936617 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chang, A. , Huang, S. D. , Benjamin, D. J. , Schmidt, J. L. , Palmer, C. G. S. , & Garrison, N. A. (2024). Exploring the role of digital tools in rare disease management: An interview‐based study. Journal of Genetic Counseling, 34, e1908. 10.1002/jgc4.1908 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dike, C. C. , Candilis, P. , Kocsis, B. , Sidhu, N. , & Recupero, P. (2019). Ethical considerations regarding internet searches for patient information. Psychiatric Services, 70(4), 324–328. 10.1176/appi.ps.201800495 [DOI] [PubMed] [Google Scholar]
  8. Drysdale, K. , Wells, N. , Smith, A. K. J. , Gunatillaka, N. , Sturgiss, E. A. , & Wark, T. (2023). Beyond the challenge to research integrity: Imposter participation in incentivised qualitative research and its impact on community engagement. Health Sociology Review, 32(3), 372–380. 10.1080/14461242.2023.2261433 [DOI] [PubMed] [Google Scholar]
  9. Gaggi, O. (2022). A study on accessibility of Google ReCAPTCHA systems . Open Challenges in Online Social Networks, 25–30. 10.1145/3524010.3539498 [DOI]
  10. Gallion, T. , Williams, Z. J. , Niarchou, M. , Duncan, L. , Hooker, G. , & Taylor, K. A. (2023). Attitudes of autistic adults toward genetic testing for autism. Journal of Genetic Counseling, 34, e1918. 10.1002/jgc4.1918 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gelinas, L. , White, S. A. , & Bierer, B. E. (2020). Economic vulnerability and payment for research participation. Clinical Trials (London, England), 17(3), 264–272. 10.1177/1740774520905596 [DOI] [PubMed] [Google Scholar]
  12. Goodrich, B. , Fenton, M. , Penn, J. , Bovay, J. , & Mountain, T. (2023). Battling bots: Experiences and strategies to mitigate fraudulent responses in online surveys. Applied Economic Perspectives and Policy, 45(2), 762–784. 10.1002/aepp.13353 [DOI] [Google Scholar]
  13. Griffin, M. , Martino, R. J. , LoSchiavo, C. , Comer‐Carruthers, C. , Krause, K. D. , Stults, C. B. , & Halkitis, P. N. (2022). Ensuring survey research data integrity in the era of internet bots. Quality and Quantity, 56(4), 2841–2852. 10.1007/s11135-021-01252-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Jagannathan, D. , MacFarlane, I. M. , & Zierhut, H. (2024). Exploration of support for black, indigenous, and people of color students in genetic counseling programs. Journal of Genetic Counseling, 33(1), 54–70. 10.1002/jgc4.1849 [DOI] [PubMed] [Google Scholar]
  15. Jones, A. , Caes, L. , Rugg, T. , Noel, M. , Bateman, S. , & Jordan, A. (2021). Challenging issues of integrity and identity of participants in non‐synchronous online qualitative methods. Methods in Psychology, 5, 100072. 10.1016/j.metip.2021.100072 [DOI] [Google Scholar]
  16. Levi, R. , Ridberg, R. , Akers, M. , & Seligman, H. (2022). Survey fraud and the integrity of web‐based survey research. American Journal of Health Promotion, 36(1), 18–20. 10.1177/08901171211037531 [DOI] [PubMed] [Google Scholar]
  17. Moreno, L. , González, M. , & Martínez, P. (2014). CAPTCHA and accessibility—Is this the best we can do? In: Proceedings of the 10th International Conference on Web Information Systems and Technologies, 115–122 10.5220/0004858901150122 [DOI]
  18. Pasca, C. , Carroll, R. , Cragun, R. T. , & Cragun, D. L. (2021). Knowledge and perceptions of the genetic counseling profession among a national cross‐sectional sample of U.S. adults. Journal of Genetic Counseling, 31(1), 206–217. 10.1002/jgc4.1473 [DOI] [PubMed] [Google Scholar]
  19. Pederson, V. , Rietzler, J. , Freeman, A. , & Petty, E. M. (2024). Picture this: Evaluating the efficacy of genetic counseling visual aids. Journal of Genetic Counseling, 33, 1365–1374. 10.1002/jgc4.1862 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Pellicano, E. , Adams, D. , Crane, L. , Hollingue, C. , Allen, C. , Almendinger, K. , Botha, M. , Haar, T. , Kapp, S. K. , & Wheeley, E. (2024). Letter to the Editor: A possible threat to data integrity for online qualitative autism research. Autism, 28(3), 786–792. 10.1177/13623613231174543 [DOI] [PubMed] [Google Scholar]
  21. Pozzar, R. , Hammer, M. J. , Underhill‐Blazey, M. , Wright, A. A. , Tulsky, J. A. , Hong, F. , Gundersen, D. A. , & Berry, D. L. (2020). Threats of bots and other bad actors to data quality following research participant recruitment through social media: Cross‐sectional questionnaire. Journal of Medical Internet Research, 22(10), e23021. 10.2196/23021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Pratt‐Chapman, M. , Moses, J. , & Arem, H. (2021). Strategies for the identification and prevention of survey fraud: Data analysis of a web‐based survey. JMIR Cancer, 7(3), e30730. 10.2196/30730 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Ridge, D. , Bullock, L. , Causer, H. , Fisher, T. , Hider, S. , Kingstone, T. , Gray, L. , Riley, R. , Smyth, N. , Silverwood, V. , Spiers, J. , & Southam, J. (2023). ‘Imposter participants’ in online qualitative research, a new and increasing threat to data integrity? Health Expectations, 26, 941–944. 10.1111/hex.13724 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Rodriguez, C. , & Oppenheimer, D. M. (2024). Creating a bot‐tleneck for malicious AI: Psychological methods for bot detection. Behavior Research Methods, 56(6), 6258–6275. 10.3758/s13428-024-02357-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Roehl, J. M. , & Harland, D. J. (2022). Imposter participants: Overcoming methodological challenges related to balancing participant privacy with data quality when using online recruitment and data collection. The Qualitative Report, 27(11), 2469–2485. 10.46743/2160-3715/2022.5475 [DOI] [Google Scholar]
  26. Salinas, M. R. (2023). Are your participants real? Dealing with fraud in recruiting older adults online. Western Journal of Nursing Research, 45(1), 93–99. 10.1177/01939459221098468 [DOI] [PubMed] [Google Scholar]
  27. Sanchez, G. , Hashmi, S. S. , Bednar, E. , Horvath, S. , Kumar, B. , Sagaser, K. , Singletary, C. N. , & Ramdaney, A. (2024). Status of abortion curriculum in genetic counseling: Survey of graduate programs and recent graduates in the United States. Journal of Genetic Counseling, 34, e1875. 10.1002/jgc4.1875 [DOI] [PubMed] [Google Scholar]
  28. Scanlon, J. K. , Wofford, L. , Fair, A. , & Philippi, D. (2021). Predictors of participation in clinical research. Nursing Research, 70(4), 289–297. 10.1097/NNR.0000000000000513 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Schneider, J. , Ahuja, L. , Dietch, J. R. , Folan, A. M. , Coleman, J. , & Bogart, K. (2024). Addressing fraudulent responses in quantitative and qualitative internet research: case studies from body image and appearance research. Ethics & Behavior, 1–13. 10.1080/10508422.2024.2411400 [DOI] [Google Scholar]
  30. Searles, A. , Nakatsuka, Y. , Ozturk, E. , Paverd, A. , Tsudik, G. , & Enkoji, A. (2023). An Empirical Study & Evaluation of modern CAPTCHAs. arXiv. No. arXiv:2307.12108. http://arxiv.org/abs/2307.12108
  31. Sivakorn, S. , Polakis, J. , & Keromytis, A. D. (2016). I'm not a human: Breaking the Google reCAPTCHA . Black Hat ASIA. https://www.blackhat.com/docs/asia‐16/materials/asia‐16‐Sivakorn‐Im‐Not‐a‐Human‐Breaking‐the‐Google‐reCAPTCHA‐wp.pdf
  32. Storozuk, A. , Ashley, M. , Delage, V. , & Maloney, E. (2020). Got bots? Practical recommendations to protect online survey data from bot attacks. The Quantitative Methods for Psychology, 16, 472–481. 10.20982/tqmp.16.5.p472 [DOI] [Google Scholar]
  33. Teitcher, J. E. F. , Bockting, W. O. , Bauermeister, J. A. , Hoefer, C. J. , Miner, M. H. , & Klitzman, R. L. (2015). Detecting, preventing, and responding to “fraudsters” in internet research: Ethics and tradeoffs. The Journal of Law, Medicine and Ethics, 43(1), 116–133. 10.1111/jlme.12200 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1.

JGC4-34-0-s001.pdf (96.7KB, pdf)

Articles from Journal of Genetic Counseling are provided here courtesy of Wiley

RESOURCES