Improving Data Integrity in Samples Obtained From Web-Based Recruitment: Protocol for the Development of a Novel System for Assessing Participant Authenticity in a Remote Longitudinal Cohort Study of Polysubstance Use
Bibliographic record
Abstract
BACKGROUND: Remote recruitment for human participant research is increasingly popular due to its speed, cost-effectiveness, and accessibility for participants. However, in some cases, it can be particularly difficult to authenticate participants recruited remotely, which, unless adequately addressed, may pose a threat to data integrity and validity. OBJECTIVE: This protocol aims to outline authenticity concerns encountered via remote recruitment for a longitudinal cohort study of adults reporting polysubstance use. Stemming from these concerns, we describe the development of a novel system of participant authenticity checks, designed with the goal of maximizing data integrity and minimizing the introduction of additional barriers to participating in the research. Finally, we examine rates of passing each active authenticity check among participants recruited via web-based advertisements. METHODS: Participants were recruited through one of several modalities, including via electronic health records and a third-party company managing a web-based advertising campaign. All participants enrolled in the longitudinal study completed a screening survey, followed by a baseline assessment (involving a survey and an interview) before completing up to 4 weekly interviews and follow-up assessments at 4, 8, and 12 months after baseline. The authenticity check system described here was implemented for all participants recruited via web-based advertising. In addition to passive authenticity checks (ie, randomized online survey passwords), we describe a five-step active authentication protocol: (1) reviewing interest forms for duplication (interest form duplication review), (2) an attention check at screening (attention check), (3) reviewing personal information after completion of the screening survey for duplicates or inconsistencies (personal information verification), (4) a verbal identity confirmation at baseline (verbal identity confirmation), and (5) a review of participant responses for inconsistent reporting at baseline (consistent reporting review). RESULTS: In total, 178 (6.85%) of the 2598 active authenticity checks administered were failed, leading to the exclusion of 119 unique potential participants due to fraudulent, inconsistent, or ineligible submissions. The 119 unique exclusions represented 11.13% (119/1069) of the potential participants identified via web-based advertising. Reviewing personal information provided at screening for inconsistencies (personal information verification) accounted for the largest number of failed checks (100/178, 56.2%), whereas reviewing interest form entries for duplicate personal information (interest form duplication review) yielded the fewest failures (7/178, 3.9%). CONCLUSIONS: The system presented provides an example of how researchers may increase confidence in the authenticity of participants recruited remotely, while avoiding the introduction of potential barriers to participating in research, such as requiring photo ID, online video call verification, or in-person verification. Such additional requirements for participants may systematically bias samples, especially when conducting research with populations that have been historically marginalized or those with stigmatized health conditions or behaviors. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR1-10.2196/69956.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.101 | 0.099 |
| Meta-epidemiology (narrow) | 0.003 | 0.003 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.007 | 0.004 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.003 | 0.004 |
| Research integrity | 0.005 | 0.007 |
| Insufficient payload (model declined to judge) | 0.047 | 0.015 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".