As a sociology undergrad, I learned all about sampling theory – from probability and non-probability strategies, and their subsets from random and stratified to snowball and convenience, through sampling size and recruitment approaches. This was back in the day when you would talk about modes such as mail, telephone, and mall intercepts.
Come the mid-2000s, when I was engaging my doctoral research around web surveys, I curiously followed the rapid development of online panels. Broadly speaking there were two key approaches. The first was access panels, through companies such as SSI and Greenfield, where people are recruited into a database, profiled, and subsequently sent targeted surveys. The other was river sampling, where visitors to a high-traffic internet site were displayed survey invitations.
As with any development, there was a lot of initial debate about their validity. Of course, it is important to remember how young the web was at that time, so concerns around coverage bias due to internet penetration rates were very real. At the same time, there was a lot of enthusiasm for this promising new way of reaching people.
Fast forward to 2020 and the world of internet sample has grown into a complex, technologically sophisticated field where conversations around APIs (application programming interface), EPCs (earnings per click), and a myriad of other acronyms are the norm. There are many approaches to recruiting including panel pools, communities as well as employing web intercepts. There are also old and new concerns about data quality, representativeness, incentives, and more.
While I’ve been following of on-going conversation about sample, and it is a prominent consideration as I run studies, I’ve not delved into the intricacies of this world, relying on the extremely competent hands of the FocusVision project managers sample providers to guide me on best approaches.
This is well and good. It is important to lean upon experts to help make the best decision. That said, I also realize that if I don’t have a better grasp of the inner workings, then how I can truly make informed decisions about my sample, my questionnaire design, and my data.
As a company that provides research tools for all different kinds of research from surveys to online focus groups, we know that sample quality is also very much on your mind because we asked you. In our recent 2020 Research Trends study, which we’ve run for three years now, sample quality once again tops the charts, with 49% of you citing it as a frustration.
It is with these thoughts that I asked our FocusVision Decipher Sample Marketplace partners – Branded, InnovateMR, Lucid, and Market-Cube – to join us for an educational discussion on sample. The conversation far exceeded my expectations, with insights into the workings as well as highlighting a clear need to continue the dialogue.
If you missed it, the webinar is available to view on-demand. And in the new year, we’ll be back with a follow-up session. I hope to see you then.
First published on FocusVision.com
コメント