Nonprobability Online Samples: Promises and Pitfalls

Nov. 1, 2023

The Survey Research Center invites you to join the live webinar below to get insights into the challenges and opportunities of gathering and using nonprobability online survey samples. The webinar is hosted by AAPOR (American Association for Public Opinion Research) and is free for the Princeton University community.

Date : November 1, 2023 1:00 PM - 2:00 PM

Speakers: Carina Cornesse and Olga Maslovskaya

Requests to participate in online surveys are everywhere: advertised in our social media timelines, embedded in our online news articles, popping up when we are concentrating on the instructions to our favorite cooking recipe or booking a business trip. If we engage with these requests, we are often also asked for contact details to receive invitations to more online surveys. Nonprobability online survey data gathered in this way serves many purposes: to get a quick picture of what the public thinks about current events, to experiment with various stimuli, or to specifically target population subgroups usually overlooked or disengaged in other survey types. While these samples have their merits, using nonprobability online survey samples can also be dangerous: partisan groups may attempt to influence the results through volunteer lobbying or programing survey bots, a highly selective subgroup of people may want to earn money by participating in as many paid surveys as possible, and important subgroups of the target population may be excluded from participation by the recruitment design.

The goal of this webinar is to provide insights into the challenges and opportunities of gathering and using nonprobability online survey samples. While this topic has many dimensions (including recent debates about transparency, artificial intelligence, or data integration), we will focus on the following aspects: 1) an overview of the sample selection process and resulting biases, 2) frameworks and guidelines for evaluating fitness-for-purpose of such samples for specific research projects.

Learning objectives:

  1. For researchers aiming to conduct their own studies, we provide theoretical frameworks and best-practice guidelines which should help them make decisions about the extent to which nonprobability online samples may be useful for their research versus alternative probability-based samples.
  2. For nonprobability online data providers, we aim to provide an understanding of the concerns, which many probability sample supporters have with regard to nonprobability online samples. We also include suggestions for moving forward in this debate.


AAPOR Webinars: Free access for Princeton community members