EBI RESEARCH PRACTICES

Quality, Integrity & Confidentiality

We are committed to delivering research you can trust. Every study we conduct follows rigorous standards to protect respondents, ensure data accuracy, and eliminate fraudulent activity. Below is an overview of the safeguards built into our methodology.

Respondent Confidentiality

• We protect all participant information and share personally identifiable information (PII) only in limited circumstances and only with the respondent’s explicit written consent.
• In studies where client follow-up is required (such as interviews, customer verifications, or incentive fulfillment), respondents opt-in to share their name, phone number, email, or address.
• Clients only receive the information respondents have expressly agreed to provide.

Trusted & Vetted Respondent Sources

• Our research uses proprietary databases built and maintained by EndeavorB2B.
• When external panelists are required, we partner with thoroughly vetted and verified sample providers.

Duplicate Prevention & Identity Controls

To maintain sample integrity, we systematically prevent duplicate entries using:

• Name, email, and IP address flagging; duplicates are never accepted.
• Unique CustomerIDs that ensure each respondent may only participate once.

Data Quality Controls

Our surveys include multiple layers of quality checks to verify participant engagement and authenticity:

Red Herring & Attention Checks

• Embedded “red herring” questions identify inattentive respondents.
• Any respondent failing attention checks is removed from the dataset.

AI-Detection & Response Validation

• Responses suspected of being generated by AI are flagged and removed.
• Open-ended answers undergo manual review to ensure authenticity.

Pattern Recognition & Straight-Lining Detection

• Surveys that contain straight-line responses across multiple grids, or identical answers within a grid, are automatically eliminated.

Speeding & Completion Time Analysis

• Completion times are monitored to detect “speeders.”
• Respondents who finish significantly faster than expected are disqualified.

Anti-Fraud Measures

Our anti-fraud practices combine automated and human review to ensure the highest levels of data quality and validity:

• Duplicate identity and IP controls
• Embedded “red herring” questions
• AI-response detection
• Speeding and response pattern reviews
• Manual verification of open-ended responses

This multi-layer approach ensures that only high-quality, authentic responses are delivered in the final dataset.