GMI Honesty Detector


BE CONFIDENT YOUR RESPONDENTS ARE TELLING THE TRUTH

Good research requires good respondents. Since the quality of respondents varies by market, source and supplier, we believe that research requires enhanced validation that ensures participants are telling the truth. For B2B/IT studies, low incidence rate studies and research in countries with high dishonesty rates, honesty certification matters even more. For all these applications, GMI’s Honesty Detector offers a groundbreaking solution to give you confidence that your respondents are answering honestly.

PATENT-PENDING, PROPRIETARY HONESTY CERTIFICATION

GMI has developed an online, statistical approach to remove over-reporters from your studies. Based on GMI’s pioneering research, Honesty Detector analyzes panelists’ responses to a unique combination of high and low probability statements as well as a benchmark question to identify dishonest respondents. Any respondent that is deemed “dishonest” is blocked from entering surveys, and as a result clients can be confident that the data they are receiving is trustworthy.

With GMI Honesty Detector, data quality improvements include better estimates of market structure, brand awareness, low incidence behaviors and ownership. Our patent-pending approach is simple to use and built into our core research platform; it is complimentary when you use GMI to manage sample for your study. GMI Honesty Detector is available across more than 20 markets for B2B, IT, and consumer audiences, and can be requested for medical professional studies.

INDEPENDENT VALIDITY CHECK

Unlike many existing quality checks, GMI Honesty Detector does not rely on information from consumer credit and reporting databases, which is available only for certain segments of the population and can exclude important demographic groups. GMI Honesty Detector works independently, as a self-contained, transparent approach. It has been proven effective regardless of demographic group or topic of study.

A NECESSITY FOR LOW INCIDENCE STUDIES

Completes = Dishonest Respondents + Honest Completes.

Low incidence studies are more likely to be affected by over-reporters, as dishonest respondents artificially increase incidence rates by claiming to be part of the true target population. Also, GMI has found that studies measuring perceived high prestige activities, such as owning electronics, high-end shopping and luxury travel are at risk of over-reporting by respondents. In these types of studies, removing over-claimers from the sample using GMI Honesty Detector has improved the accuracy of survey results dramatically.

STABILIZE EMERGING MARKET SAMPLE

GMI’s multi-country research studies revealed that certain countries have higher percentages of untruthful online respondents than other countries. In GMI’s experiments, lying varied between 3 percent and 30 percent from country to country, and in every case, posed a major risk factor to accurate data. GMI Honesty Detector effectively regulated the sample from country to country, allowing for accurate comparison of data across international studies.

GMI HONESTY DETECTOR DOES NOT:

  • Pose a risk of signaling respondents
  • Depend upon any other data source
  • Introduce sample bias

HONESTY DETECTOR IMPACT

  • Less back end data cleaning
  • Reduced need to replace “bad completes”
  • More realistic frequency distributions that match market share data
  • Improved verbatim quality
  • Increased confidence in research findings
We recently had a rush project to decide on a title for an iPad app. Within two weeks, GMI wrote the program, and worked with our team to revise, field and return the dataset. The survey interface was very user-friendly and the GMI team was helpful, professional, and met our very tight budget guidelines. We look forward to our next project with GMI.
JAMES WILLIAMS-NESS, SESAME WORKSHOP