Online Program

Social media as qualitative research: Reports of a data mining experiment

Tuesday, November 5, 2013 : 5:30 p.m. - 5:50 p.m.

Iana M. Simeonov, UCSF mLabs, University of California San Francisco, San Francisco, CA
As people broadcast information and solicit advice about health through social media, the medium offers volumes of rapid, real-time information.

We wanted to explore if the freely available information on social media would offer similar insights to and possible replace costly and time-consuming research using focus groups. We also sought to determine if such channels offered actionable information and entry points for health outreach or promotion opportunities.

We used two social media data mining services, Radian6 and Lithium, and a combination of free trials and a paid subscription. Both services search billions of Twitter feeds, blogs, online discussion boards, forums and news articles for key terms and use proprietary algorithms to evaluate and categorize posts. Using these services, we were able to analyze and rank words included along with our key terms for positive and negative sentiment. Sentiment analysis allowed for both big-picture insights and detailed observation of specific posts, shedding light on conversations and influencers.

Eschewing neutral posts, those that skewed positive or negative mirrored focus group insights with surprising accuracy. A key, and highly actionable, negative sentiment surfaced in social media posts as it had in focus groups. The critical difference was that unlike with focus groups, in the social media space we were immediately able to join and help steer the conversation.

Social media offers significant opportunities for understanding consumer health behavior that are more cost-effective than traditional research methods. Observing health “crowdsourcing” and unfiltered experience also affords opportunities for direct engagement.

Learning Areas:

Communication and informatics
Implementation of health education strategies, interventions and programs
Public health or related education
Public health or related research

Learning Objectives:
Describe how social media text mining works Indicate uses of data mining for public health research Identify the right social media monitoring tools for research projects Formulate a simple strategy for social media data mining

Keyword(s): Data Collection, Social Marketing

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: Over 10 years of experience in research, program and development for underserved populations, multi-year APHA, NIH and CDC conference presenter, responsible for insights and innovation in mobile technology for research and practice at one of the top 5 medical schools in the U.S. Creator of bilingual text-messaging campaigns, online games, apps, digital and mobile tools for public health programs and projects nationwide.
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.