Exploring the impact of social media user experiences on social cohesion & reconciliation in Puntland and South West State.
This short, preliminary report is based on data collected through HeshiisBeeg, a participatory barometer survey conducted in 2025 across 3 districts in each of Somalia’s Puntland and South West State. The survey reached 1800 individuals, proportionally spread, with between 150 and 350 people in each district; its results are representative at the district level. Heshiis Beeg is a collaboration between the Puntland Development and Research Centre, Dhaxal Reeb, the Centre for Sustainable Peace and Democratic Development and Build Up.
This report was produced by the Build Up team to catalyze discussions about the role that user experiences on social media and traditional media might play in social cohesion and reconciliation dynamics in Somalia. While the data is only representative of experiences in the 6 districts surveyed, our hope is that these insights will highlight the relevance of digital harms in the Somali context and encourage further investigation and policy discussions.
OVERVIEW OF EXPERIENCES
Overall, only 43% of survey respondents used any media (digital or traditional) in the past 4 weeks. Table 1 shows that media usage is dominated by social media, with traditional media accounting for less than 10%.
Figure 1: Percentage of survey respondents who used media in the past 4 weeks
Access to digital devices is relatively high among respondents, with 72% of respondents having access to a mobile phone they can use. This percentage is much higher in Puntland (83%) than in South West State (61%), largely due to very low access in the Baidoa district (46%). Nomadic households have the lowest access to mobile phones (30%). Furthermore, 50% say mobile phone coverage is provided very efficiently, with this rising to 70% in Puntland (versus 32% in South West State and 26% in Baidoa). People who don’t have access to a phone or to the internet also report using media less.
This paints a fairly clear picture: it seems reasonable to assume that media usage is primarily people accessing social media on their mobile devices.
Table 1 also reveals there is a significant difference in media usage between men and women. There are also far fewer people over 55 (20%) and nomadic people (16%) who use any media. However, the strongest predictor of media usage is education: those who are more educated are significantly more likely to have used media in the past 4 weeks.
Facebook usage is especially high in Puntland, with 92% of those who used any media in the past 4 weeks reporting they used Facebook.
Exposure to negative media content
Among the 43% of respondents who consumed media in the past 4 weeks, 20% witnessed or experienced something that affected them negatively multiple times per day. Table 2 gives the full breakdown and highlights gender differences, which do not conclusively show whether men or women experience worse content.
Figure 2: Percentage of survey respondents who used media in the past 4 weeks who were exposed to negative content
Respondents report that negative content was mainly about armed conflict in general, hatespeech targeting their (sub)clan and politics.
Exposure to positive media content
Among the 43% of respondents who consumed media in the past 4 weeks, 25% witnessed or experienced something that affected them positively multiple times per day. Table 2 gives the full breakdown and highlights gender differences, which again do not conclusively show whether men or women experience worse content.
| All | Men | Women | |
| None | 12.4% | 10.7% | 15.2% |
| Less than once a week | 10.3% | 12.5% | 6.8% |
| About once a week | 14.9% | 16.1% | 12.9% |
| A few times per week | 17.3% | 17.0% | 17.9% |
| About once a day | 19.6% | 17.1% | 23.4% |
| Multiple times per day | 25.4% | 26.5% | 23.8% |
Table 1: Percentage of survey respondents who used media in the past 4 weeks who were exposed to positive content
Respondents report that positive content was mainly about peace messaging campaigns to counter hatespeech, politics, religion, local news and entertainment.
Finally, respondents had very different experiences across platforms. Table 4 shows the percentage of users of each platform that reported being exposed to negative and positive content. These results suggest that Facebook is a highly emotive platform (both positive and negative); TikTok is also highly emotive (and more negative); X and YouTube are less emotive (both positive and negative); and WhatsApp is the least emotive (and more negative).
Table 4: percentage of survey respondents who used a specific social media platform in the past 4 weeks and reported they were exposed to positive or negative content on that platform
Taken together, these results suggest that the overall balance of content on social media is positive, although there is significant negative content.
IMPACT OF SOCIAL MEDIA EXPERIENCES
The findings in the previous section say nothing about the offline impact that positive or negative online content might have on people. The Heshiis Beeg survey asks a wide array of questions about people’s perceptions of others and their experience of conflict. Together, these questions are designed to capture a measure of peace and conciliation as understood by Somali stakeholders.
By examining how exposure to positive and negative content correlates with responses to relevant perceptions and experiences, we can draw general conclusions about the impact of social media on peace and conciliation. Because these conclusions are tentative and based on only six districts, they are phrased here as discussion questions to be taken forward.
- One overall observation is that respondents who reported using any media in the past 4 weeks also reported significantly more civic engagement. Discussion question: What does this tell us about the type of people who use media?
- Exposure to various kinds of environmental shocks (droughts, extreme weather, etc.) is correlated with greater exposure to both negative and positive content on social media. Similarly, exposure to conflict, to injury or death due to conflict, to violence and insecurity, to kidnapping and human trafficking, and to adversity in general are also correlated with more exposure to both negative content and positive content on social media. Discussion question: Could social media serve as a conduit for expressing both the suffering and the solidarity that come from exposure to adversity?
- In general, willingness to forgive correlates with exposure to more negative content. Similarly, the tendency to seek revenge against someone who stole from you is also correlated with exposure to more negative content. Discussion question: Does exposure to negative content cause or amplify these negative tendencies?
- Personal experience of conflict between clans is correlated with exposure to more negative content, and people who have a personal experience of clan violence report significantly more negative content on social media. Discussion question: Unlike point 3, does causation run the other way here, with the offline experiences resulting in more engagement with negative content?
- Contact with Somalis from other clans is correlated with exposure to both more negative and more positive content. Intergroup contact in general (including contact with IDPs, refugees, other clans, and subclans) is only correlated with exposure to more positive content. Discussion question: How does this inform or complicate the finding in point 4?
- Trust in the National Army correlates with greater exposure to more positive media. Discussion question: Does the National Army have a (particularly effective) social media campaign?
