MUTYABA, JOHN LIVINGSTONE2026-06-232026-06-232024-10http://41.89.96.81:4000/handle/123456789/3852Globally, cage fish farming technologies are intensive production systems highly promoted under the blue economy and seen as sustainable sources of income and employment, especially for smallholder farmers in fish-dependent communities. Despite their potential, adoption of such technologies by smallholder farmers in Uganda is still low. In addition, limited extension services in the country further challenge the sector in terms of accessing agricultural information, particularly for smallholder fish farmers. As a result, many smallholder farmers rely on social network platforms to access information about cage fish production and market. However, these platforms face challenges in adopting new technologies and accessing information on best agricultural practices and research findings from institutions. Therefore, this study was conducted to contribute empirically to the existing knowledge and to enhance adoption of cage fish farming technologies among smallholder fish farmers practicing along the shores of Lake Victoria in Uganda, ultimately aiming to improve their livelihoods. The results showed that out of the 384 sampled smallholder farmers, 98% were rearing Nile Tilapia fish species in cage technologies, while only 2% were rearing catfish. Moreover, the majority of respondents (82.8%) used metallic cage units to rear fish, while 11.5% used wooden and only 5.7% used HDE plastics. The Multinomial Logit (MNL) results indicated that age, gender, farming experience, education level, wooden cage type, extension visits, social capital, television access, and distance to the nearest market were statistically significant and associated with knowledge, attitude, and perception outcomes. The relative risk ratio values were 1.0, 12.6, 1.99, 1.0, 0.17, 0.28, 1.89, 7.32, and 1.06, respectively. Furthermore, the Double Hurdle (DH) results revealed that an increase in the experience and group membership of a cage fish farmer by one-unit change increased the probability of using social networks by 0.43 and 0.70 units, respectively. Additionally, a unit increase in extension visits and credit access decreased the probability of using social networks by 0.59 and 1.06 units, respectively. The results from Propensity Score matching (PSM) estimated an average treatment effect (ATT) of 4539.21 kilograms of fish annually over the control group and was statistically significant at t=2.32; p=0.006. The treated group harvested an average of 24627.71 kilograms of fish annually, compared to 20088.5 kilograms for the untreated group. Therefore, integrating social networks into existing policy interventions can empower smallholder farmers and facilitate sustainable agricultural development in Uganda's agricultural sector.enThesis