Sample – Blended, Not Stacked

In the world of online quantitative research, it can be very common for a sample provider to “fall short” of providing the agreed-upon number of completes for a survey. Most of the time, this may not even be entirely the sample provider’s fault—study specifications that were assumed prior to launch don’t match what has occurred in field, incidence rate comes in lower than expected, the length of the interview is longer than anticipated, a quota might be unexpectedly tough to fill, or a myriad of other reasons.

Researchers are often left scrambling to get the completes they need to close their study. They throw new sample providers on as quickly as possible in hopes that they get the needed completes quickly so they don’t miss their deadlines. When this is done, sample providers are “stacked” on top of one another. This method is not typically planned ahead of time and the added panels usually have not gone through the same screening process that the original sample provider may have gone through. There have been many a late Friday afternoon where researchers are scrambling to add a new sample provider to get the completes they need over the weekend. This is not a sampling best practice and may introduce risk to your study.

You are probably thinking… there has to be a better way, right!? There is! It’s called strategic sample blending and it entails blending three or more sample providers in an intentional and controlled method. The sample providers selected to be part of the blend are chosen to complement each other while reducing the overall sample bias and any potential behavioral and attitudinal impacts a single panel can have.

Strategic sample blending is better than stacking sample for a variety of reasons. First, the suppliers are selected by experts in the industry who know each panel’s unique capabilities and which projects they are the best fit for. Second, they manage risk, keeping your study from being too reliant on a single panel which, like we mentioned above, might fall short, leaving you high and dry at the last minute. Third, strategic sample blending allows for maintenance over time. If one panel in a blend isn’t performing, they can be replaced with a like panel with little to no impact to the overall blend, making it a more reliable long-term option.

We believe that strategically blending sample is the best practice for all online quantitative studies. It provides improved feasibility, better consistency, reduced risk, faster fielding times, and more. Not all techniques for combining multiple panel providers are created equal—it should be done in a strategic manner. Here at EMI, we are experts in the overall sample landscape and creating strategic sample blends.

To learn more about our strategic sample blending capabilities, check out the Strategic Sample Blending page on our website.