Informing Social and Behavior Change Programs Using Social Listening and Social Monitoring

Platforms in which social listening can be conducted include Facebook, Instagram, Reddit, YouTube, blogs, and news sites, among others. Advancements in machine learning for text analysis have expanded the capacity to rapidly collect and analyze social media data. For example, the Lexrest machine learning algorithm can identify the most representative posts in a conversation, and the Home Hexagon/Brandwatch software can classify posts by topic. Social listening enables analysts to identify key conversation terms and construct Boolean search strings using terms in all languages. Analysts can also set the listening or monitoring period of interest, target one or more social media platforms, and work with social media analysis software programs and machine learning software. For social monitoring, the sources are limited to the social media used by the relevant SBC program. Program implementers may set up hashtag-monitoring systems to track engagement and reactions to campaigns or content of interest. Information collected through social listening is analyzed for patterns in volume of posts, spikes in engagement, and drivers of conversation, sentiment, and sentiment affinities. This analysis can be similar to an analysis of qualitative data from other sources. During this process, all posts used for analysis or featured in reports must be anonymized by removing names and profile pictures. Social listening platforms do not automatically anonymize posts, but this should be done by analysts to confer identity protections or abide by human-subject requirements. Breakthrough RESEARCH is working with M&C Saatchi to apply social listening and social monitoring to inform SBC programs in Francophone West Africa. We provide examples from our experiences using these methods of the USAID-supported activities Merci Mon Héros and Resilience in the Sahel Enhanced (RISE) II. The Merci Mon Héros campaign aims to encourage young people to talk about their family planning and reproductive health realities, encourage adults to overcome restrictive social and gender norms to talk about family planning and reproductive health with young people, and stimulate discussion between young people and adults to identify, address, and shift restrictive social norms and the shame and tabooid that prevent young people from accessing family planning and reproductive health information and services. The campaign was developed by youth through an initiative supported by Breakthrough ACTION and launched in November 2019, primarily through Facebook. Campaign messages were also disseminated through other social media platforms. The campaign's social media monitoring system tracked engagement and reactions to campaign messages, and analyzed the sentiment of the conversation. The system also identified key conversation terms and constructed Boolean search strings using terms in all languages. Total potential impressions refer to the number of people who could have seen a given post based on the number of followers of those who engaged with the post. Total reach refers to the number of people who saw a post at least once. Reach is defined by Facebook as "the total number of people who had any content from your page or about your page enter their screen." Total engagement/conversation volume refers to the total number of times people engaged with a post through reactions, comments, shares, retweets, mentions, and likes. Sentiment refers to the overall positivity or negativity featured in a post, and is measured by the tone of the keywords used in a post. The MMH campaign's social media monitoring system tracked engagement and reactions to campaign messages, and analyzed the sentiment of the conversation. The system also identified key conversation terms and constructed Boolean search strings using terms in all languages. The campaign's social media monitoring system was used to inform the design and adaptation of the program, and to improve the reach and engagement of the campaign.