Fixing the neighborhood: what's actually broken in data quality with Karine Pepin
September 22, 2026
Data quality isn’t just about better fraud detection or securing individual surveys. In this episode of The Curiosity Current, Stephanie Vance and Jonathan sit down with Karine Pepin to explore what’s actually broken in the research ecosystem and why fixing one survey at a time can only go so far. Karine shares how researchers can establish their own “ground truth,” interpret conflicting fraud signals, question unrealistic sample claims, and rethink what’s possible when researching hard-to-reach audiences.
In this episode of The Curiosity Current, Stephanie Vance and Jonathan sit down with Karine Pepin to unpack what’s actually broken in data quality and why the problem is bigger than any individual survey.
Karine explains why researchers need to look beyond fraud detection tools and think about the wider ecosystem that produces their data. She shares how creating a “ground truth” can help researchers evaluate the quality of their sample, why fraud signals need to be interpreted in context, and what incidence rates can reveal before a project is complete.
The conversation also challenges some of the assumptions researchers make about online sampling, from impressive panel-size claims to the idea that every audience can be reached online.
Ultimately, the episode comes back to one bigger question: how do we move from securing our own “house” to fixing the “neighborhood” around it? For Karine, that means changing an ecosystem where volume is often rewarded more than quality.
What You’ll Learn:
- Why data quality is an ecosystem problem, not just a survey problem
- How to create a “ground truth” for evaluating your research data
- Why fraud detection tools alone can’t tell you whether a respondent is trustworthy
- How to interpret false positives and false negatives in context
- What incidence rates can reveal about the quality of your sample
- Why panel size doesn’t necessarily tell you anything about sample quality
- Why researchers need more transparency around where their sample comes from
- When low-incidence and niche audiences may not be realistic targets for online research
- How incentives around volume can contribute to ongoing data quality problems
- What researchers can do to improve the quality of their own work while the wider ecosystem changes
About the Guest:
Karine Pepin is a co-founder of The Research Heads and an advocate for better data quality and participant experiences in market research. With more than two decades of experience in research, Karine has worked across the industry with a particular focus on data quality, sampling, fraud, and the systems that shape the quality of research data.
Karine Pepin is a co-founder of The Research Heads and an advocate for better data quality and participant experiences in market research. With more than two decades of experience in research, Karine has worked across the industry with a particular focus on data quality, sampling, fraud, and the systems that shape the quality of research data.
Her work looks beyond individual tools and processes to examine the broader ecosystem around online research and how the industry can create better incentives for quality, transparency, and more reliable data.
Episode Resources:
- Karine Pepin on LinkedIn
- The Research Heads Website
- Stephanie Vance on LinkedIn
- Jonathan on LinkedIn
- The Curiosity Current: A Market Research Podcast on Apple Podcasts
- The Curiosity Current: A Market Research Podcast on Spotify
- The Curiosity Current: A Market Research Podcast on YouTube
The Curiosity Current: A Market Research Podcast is handcrafted by our friends over at: fame.so