Market Research: Foundations to Practice · Survey Data Quality and Fraud Prevention
Building a Data Quality Culture
Individual detection techniques matter less than the organizational habits that determine whether those techniques get applied consistently. This chapter covers what that looks like in practice.
Given the current fraud landscape covered in this module, treating data quality as a single checkpoint performed once at the end of data collection is no longer adequate; the current recommended approach monitors quality signals continuously throughout a study's fielding period, catching a developing fraud pattern, such as a sudden, suspicious spike in unusually fast completions, while a survey is still actively in the field and can still be corrected.
Key Takeaways
- Quality monitoring should be continuous throughout fielding, not a single end-of-collection checkpoint, to catch developing fraud patterns while still correctable.
- Sample quality is a decision-risk issue, not just an operations inconvenience, since contaminated data can make a poor idea look promising or distort willingness to pay.
- Specific, measurable quality requirements should be built into vendor contracts upfront, not discovered only after a problematic study has been delivered.
- Researchers should help stakeholders understand why rigorous quality practices sometimes cost more or take longer, so stakeholders don't push for undermining shortcuts.