This presentation describes a quantitative study designed to learn how response errors are affected when panelists are told their data is suspect. Attendees will walk away knowing how to encourage higher quality responding, choose the lowest amount of researcher intervention required to improve data quality and be able to detect which fraudsters can be rehabilitated. The highlighted study includes five groups of active panelists who received interventions ranging from nothing to being informed that their answers were suspect and they might be deactivated. Previously deactivated panelists were also included for comparison. Straightlining, red herrings, contradictions, and other fraudulent behaviors were reviewed in terms of how often they occurred and how many times the panelist was told their data was suspect. This presentation outlines how disclosing information to panelists about data quality and their survey behaviors can affect the results of the survey.