Finding the respondents you want in your research

Most data quality systems are designed to identify problems. Fraud signals, suspicious devices, speeding, poor open ends and other checks help researchers decide who may need to be removed from a study. But growing respondent histories can also tell us something about the people researchers may want to keep.

In a new article for Research Live, our CEO Bob Fawson looks at the other side of respondent quality: identifying people who have consistently provided good data across multiple studies.

As Bob writes, “Respondent history gives us a chance to learn something about the people who consistently do a good job, too.”

Gold Standard respondents account for 29.5% of starts among known respondents

As DQC’s respondent history has grown, we have been able to identify people with an established record of good participation across client surveys. We call these Gold Standard respondents.

Among known respondents with sufficient history, 22% currently meet the Gold Standard, and they account for 29.5% of survey starts among that group.

Those numbers don’t mean frequent survey participation creates better respondents. They do show that a meaningful share of survey activity is coming from people who have already demonstrated their ability to provide good data.

As Bob explains, “Frequency by itself doesn't tell us much about quality. Someone may take a lot of surveys and consistently provide thoughtful, reliable data, while someone who participates less frequently may repeatedly exhibit quality problems.”

That distinction becomes much easier to see when researchers have access to quality information from previous studies.

Positive history adds context to current quality signals

Respondent history can also help researchers interpret what happens within an individual survey.

Bob gives the example of a participant who triggers a speeding flag. That signal still matters, but its meaning may be different when researchers can see that the same person has performed well across a number of unrelated studies.

“If the same person has participated in a number of unrelated studies and consistently performed well, the speeding flag still matters, but there is more evidence available to evaluate it.”

Historical performance gives researchers additional context when evaluating current-study behavior and technical quality signals. It can help determine whether a flag warrants further review, particularly when it appears inconsistent with a respondent's established record.

Good respondents have value for buyers and suppliers

Identifying consistently good participants also raises practical questions about how the industry treats them.

Buyers could ask suppliers to prioritize respondents with strong participation histories or reconsider how much screening those individuals need. Suppliers could use the same information to offer better incentives and research opportunities to people who consistently provide good data.

As Bob writes, “Good respondents are the people producing successful completes and, in turn, driving revenue and profit.”

Making greater use of positive respondent history would require changes in how buyers and suppliers approach quality decisions. DQC's growing cross-study data provides a way to begin identifying these respondents and exploring how that information could improve sourcing, screening and retention.

Read Bob’s full article in Research Live: https://www.research-live.com/article/opinion/the-overlooked-side-of-data-quality/id/5153586

Want to see how respondent history and DQC’s Gold Standard work in practice? Book a demo with us.