Does brand sentiment vary by social media platform? According to research by David Scweidel and Wendy Moe, the answer appears to be yes. The authors discuss the results of their 2014 study, in which they modeled previous data collected from different social media venues (blogs, Facebook, Twitter, etc.), in an effort to determine if consumer brand sentiment varies by venue type. In this article, Scwheidel and Moe uncover the risks of using social media metrics without accounting for differences across venues. Previous studies had already suggested that what people say in a post is related to where they post. In this study, the authors set out to examine this further and to find out why blogs have the most positive posts, forums the second most positive posts, and microblogs the least. The result? The article points to two factors: 1) Consumers often choose to participate in online communities whose member share their interests and opinions; and 2) The limitation on numbers of characters allowed has impact on the opinion expressed. Forums and blogs, (Facebook) because they are lengthier, allow for more expression. Additionally these sites actually expose posters to more varying amounts of social dynamics, including peer pressure, etc. On micro blogs, (Twitter) where text is limited, consumers tend to post more extreme opinions so that they can convey their perspective succinctly. The results from the authors’ modeling shows that the inferences marketing researchers obtain from monitoring social media are dependent on where they “listen.” For example, Facebook tends to be more positive, and Twitter more negative in the opinions expressed, based on the dynamics stated in the above paragraph. Common approaches that either focus on a single social media venue or ignore differences across venues in aggregated data, can lead to misleading brand sentiment metrics. The authors conclude “… the current research demonstrates the potential for social media monitoring to supplement “market” research programs, but further investigation using both social media and survey data from a range of categories is essential before market researchers can rely exclusively on social media for customer insights.” 1Muniz, Albert M., Jr., and Thomas C. O’Guinn (2001), “Brand Community,” Journal of Consumer Research, 27 (4), 412–32. This synopsis was written by Lynn Croft, independent marketing and market research consultant. With 15 years of experience at companies such as Genzyme, Bayer Corporation, Shire, and Eli Lilly, Lynn has expertise in market research, market analysis regarding product launches, pricing and lifecycle management. [Want to learn more about social media monitoring, social sample sources, and more? Get a practical perspective on how you can use social media in your market research projects in our 90-minute, live, online Social Media Meets Market Research class. MRA approved for 1.5 hours of PRC credit.]
Kathryn Korostoff is founder and lead instructor at Research Rockstar. Over the past 25 years, she has personally directed more than 600 primary market research projects and published over 100 bylined articles in magazines. She is also a professor at Boston University, where she teaches grad students how to analyze and report quantitative data.