Scott McDonald, ARF CEO & President, issued a call for the development of ethical standards regarding the protection of consumers and research and use of their data. In addition, he stated that the ARF is reviewing whether to rescind a 2017 award given to Cambridge Analytica, given their alleged improper use of consumer information.
The continuing rise in non-response rates has led to growing concern about the representativeness of surveys. Andrew Mercer, Senior Research Methodologist at the Pew Research Center, provides an overview of the issue along with best practices in modeling representativeness, for example, think through your modeling assumptions before collecting data.
The 2016 U.S. presidential election served a big blow to the polling community, as pre-election predictions proved to be dramatically wrong. But is there really a crisis in pre-election polling? Mark Blumenthal, Head of Election Polling at SurveyMonkey, summarizes the AAPOR’s post-election investigation that demonstrated the accuracy of national polls, but weaknesses in state-level polls. Part of the challenge going forward – correcting for sample bias in an age of lower survey response rates.
Member Only AccessMeasuring the impact of in-market investments can be like finding the proverbial needle in a haystack. Combining the granularity of panel-data and the heft of Big Data can make finding that needle a lot easier. Oracle Data Cloud shows how they bring the two data sources together.
Member Only AccessAlmost 30% of consumers answer surveys on their smartphones, but survey drop outs are 20% higher on smartphones than on PCs. How to fix this? This summary of the ARF’s Original Research on Optimizing Mobile Research shows how using emojis in surveys can make a big difference, while offering tips on what to avoid.
Member Only AccessIncreasing number of people are looking to complete surveys on smartphones. Emojis are very thumb-friendly. Can they be used as an alternative to traditional response formats? Findings from ARF’s Original Research with GfK and FocusVision demonstrated that emojis can improve the consumer experience, reduce drop-off rates and deliver comparable data as text. But emojis don’t work in every situation.
Member Only Access
With so much 3rd party data available, how do you know how well that data represents your intended consumer target? Lotame and Survata partnered to link 1st and 3rd party data and boosted accuracy rates for identifying auto intenders on a publisher’s website by nearly 1/3rd.
Member Only AccessNon-probability samples have become a de facto norm in on-line survey and marketing research. But are there opportunities to improve the demographic screening tools to reduce non-probability error and come closer to Random Probability samples? This JAR article shows that additional demographic screening did not reduce bias or improve data accuracy when measured against benchmarks from large-scale Random Probability studies.
Member Only Access