Strengthening the Merci Mon Héros Campaign Through Adaptive Management: Application of Social Listening Methodology
Bibliographic record
Abstract
BACKGROUND: Between 2014 and 2018, the penetration of smartphones in sub-Saharan Africa increased from 10% to 30%, enabling increased access to the internet, Facebook, Twitter, Pinterest, and YouTube. These platforms engage users in multidirectional communication and provide public health programs with the tools to inform and engage diverse audiences on a range of public health issues, as well as monitor opinions and behaviors on health topics. OBJECTIVE: This paper details the process used by the U.S. Agency for International Development-funded Breakthrough RESEARCH to apply social media monitoring and social listening techniques in Burkina Faso, Côte d'Ivoire, Niger, and Togo for the adaptive management of the Merci Mon Héros campaign. We documented how these approaches were applied and how the lessons learned can be used to support future public health communication campaigns. METHODS: The process involved 6 steps: (1) ensure there is a sufficient volume of topic-specific web-based conversation in the target countries; (2) develop measures to monitor the campaign's social media strategy; (3) identify search terms to assess campaign and related conversations; (4) quantitatively assess campaign audience demographics, campaign reach, and engagement through social media monitoring; (5) qualitatively assess audience attitudes, opinions, and behaviors and understand conversation context through social media listening; and (6) adapt campaign content and approach based on the analysis of social media data. RESULTS: We analyzed posts across social media platforms from November 2019 to October 2020 based on identified key search terms related to family planning, reproductive health, menstruation, sexual activity, and gender. Based on the quantitative and qualitative assessments in steps 4 and 5, there were several adaptive shifts in the campaign's content and approach, of which the following 3 shifts are highlighted. (1) Social media monitoring identified that the Facebook campaign fans were primarily male, which prompted the campaign to target calls to action to the male audience already following the campaign and shift marketing approaches to increase the proportion of female followers. (2) Shorter videos had a higher chance of being viewed in their entirety. In response to this, the campaign shortened video lengths and created screenshot teasers to promote videos. (3) The most negative sentiment related to the campaign videos was associated with beliefs against premarital sex. In response to this finding, the campaign included videos and Facebook Live sessions with religious leaders who promoted talking openly with young people to support intergenerational discussion about reproductive health. CONCLUSIONS: Prior to launching health campaigns, programs should test the most relevant social media platforms and their limitations. Inherent biases to internet and social media access are important challenges, and ethical considerations around data privacy must continue to guide the advances in this technology's use for research. However, social listening and social media monitoring can be powerful monitoring and evaluation tools that can be used to aid the adaptive management of health campaigns that engage populations who have a digital presence.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.026 | 0.035 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.005 | 0.002 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".