Good quality sample – making sure you are talking to the right people – is a fundamental of good research. It's the first step in ensuring good quality research and without it, our data is meaningless, or worse, misleading. 

This is not just an issue when using a research panel – understanding and avoiding potential pitfalls also applies when running research on your own customer database. 

What does poor sample quality look like?

There are a whole range of sample issues that can arise when conducting research. 

At its most egregious examples of this involve fraud: knowingly falsifying sample, although thankfully this is rare. However, there are a range of other issues beyond this to look out for. It might be people who have signed up to a panel multiple times and complete the same survey potentially dozens of times. Others are virtually professional respondents, responding to surveys one after another. Or those who skip through surveys quickly, providing nonsense answers.

More recently, bots and AI have emerged as a threat to data quality – some of your respondents may not even be human!

How do we ensure your sample is high quality?

At Edentify, we start with our own research panel, Caféstudy, which we can ensure is not over-surveyed, and we validate panel members to ensure they are who they say they are. 

It's also important to ensure that the project specifications are realistic. This means setting achievable quotas, designing the questionnaire to be as short as possible, easy to complete, and error-free, so that respondents are encouraged to answer honestly and completely.

Once the data is collected, we then run a series of quality checks, including digital fingerprinting to identify any duplicate responses, timing checks to catch anyone who may have raced through the survey, and answer checks to highlight any abnormal open-ended responses. These quality checks are a mix of AI and human to provide a full picture and allow us to make the right call.

When we use our client databases or third party panels for research, we go through the same quality checking processes so you can be confident of the data that Edentify delivers.

Maintaining sample and data quality is an ongoing challenge for research agencies, and something that we are always working on improving at Edentify. If you have any questions or would like more information, contact us.