Effectiveness Review: Community-Led Disaster Risk Reduction in Mongu, Zambia

This 53-page report shares information from an effectiveness review of Zambia's "Community-Led Disaster Risk Reduction" which is working to increase resilience to climatic shocks among key groups in Mongu district of western Zambia. Following analysis of the data, there is evidence that the project positively affected several characteristics assumed important for promoting resilience among the intervention population. Overall, households in the communities where the project activities had been implemented scored positively on an average of 52% of 31 characteristics of resilience considered in the review, compared to 46% in the comparison communities.

The project activities focus on strengthening the capacity of communities to manage and respond to floods and droughts, as well as encouraging livelihood diversification and asset growth. Specific endeavours include:

• Community sensitisation on early-warning information through school activities, radio messages, and community meetings;
• Efforts to strengthen community relationships with district-level early-warning institutions (e.g., weather monitoring departments); and
• Review of community preparedness plans, including integration of traditional early-warning signs.

A range of other activities, including provision of fishing nets, canal clearing, embankment building, establishment of banana plantations, and use of conservation agriculture have also been implemented.

To assess the effectiveness of this project, a quasi-experimental impact evaluation was conducted. This involved carrying out household surveys in the six communities supported by the project, as well as with households in six nearby comparison communities. In all, surveys were carried out with 491 households. The effectiveness survey analysed how the project affected 31 "resilience characteristics", which are interrelated and fall under five dimensions. The assessment for each is outlined as follows:

• Dimension 1: livelihood viability - Evidence of impact on households' dietary diversity and availability/use of early-warning information. No clear differences between project and comparison households on measures of household wealth, food security, and flood-preparedness practice.
• Dimension 2: innovation potential - Strong evidence of impact on household practice of innovative activities and household access to state innovative support. No clear evidence of differences in access to markets or access to credit facilities.
• Dimension 3: access to contingency resources and support - No clear differences between project and comparison households on measures which make up this dimension (group participation, social connectivity, savings, remittances or formal earnings, local government emergency support).
• Dimension 4: integrity of the natural and built environment - No clear differences on measures which make up this dimension (soil fertility/erosion, access to irrigation for farming, extent of crop loss due to flooding).
• Dimension 5: social and institutional capability - Evidence of positive impact on awareness of, and participation in, community-level flood preparedness plans and meetings. However, no evidence of an increase in confidence in effectiveness of local leaders/institutions.

Overall, the survey found that the project has positively affected several characteristics deemed important for promoting resilience among the intervention population. One characteristic in which there is a significant difference relates to the availability of early-warning information. Households were asked to think back to the flooding event of 2012 and to recall whether they received early-warning information that enabled them to plan and protect their livelihoods. Overall, 82% of project households reported receiving such information, compared to 67% of comparison households - a difference that is highly statistically significant. Respondents were also asked to record the source of such information; 32% reported receiving the information from community leaders, 50% from the Community Disaster Risk Management Committee, 28% from government officers, and 69% from the radio.

While there is evidence of a strong project effect on access to early-warning information, there is no evidence that the project affected whether households took any specific precautionary actions in preparation for the flooding of 2012. Less than half (47%) of project households reported taking any such action, compared to 43% of comparison households - a difference that is not statistically significant.

The intervention households outperformed the comparison households on the social and institutional preparedness activities (indices include participating in preparedness meetings, having received preparedness information, and being aware of any preparedness plan). While the supported households did not score particularly well on these latter characteristics, they are clearly better off in these regards than households in the comparison villages. It is worth pointing out that the majority of all the interviewed households said that they had not been involved in a dispute over access to water in the previous two years, hence resulting in strong performance among both groups on this measure.

The report offers the following lessons learned for the Zambia country team, and those undertaking similar projects:

• Consider further research to evaluate the effects of advocacy efforts connected to this project - As mentioned in the introduction to this report, the effectiveness review only considered the impact of the community-level interventions connected to this project. One key intended outcome from the wider project was to "improve government capacity to lead and manage comprehensive and effective disaster risk reduction and emergency response from local to national levels". As noted in the report, the proportion of households - particularly in comparison villages - who knew how to access state adaptation support, or had confidence in the quality of such support, was very low. These results indicate there is more work required in both improving the linkages between communities and local government support and in strengthening the support offer from state institutions. It is interesting to note that the results for the former were more positive in the intervention villages, indicating a positive project effect on these particular issues. Findings from further research should help highlight how and why the project has positively impacted these particular results, and how advocacy efforts across the district and beyond can best be targeted.
• Explore how to involve community members more widely in community-level drought preparedness activities and to ensure that training and early-warning information is fully disseminated - While the project was found to have had some effect on the distribution of disaster preparedness information to community members, the fact that only half of households reported receiving or being aware of such information is a cause for concern. This suggests that there is still scope for further embedding these activities in the life of the community and for ensuring that all households are involved and can fully benefit.
• Continue monitoring changes in behaviour and experiences of households in the project communities to learn whether the project activities will eventually result in higher-level changes in risk-reduction behaviour - An important observation arising from the results of this effectiveness review is that most of the characteristics on which the projects appear to have had impact are those directly connected to project outputs: receipt of early-warning information, involvement in community-level preparedness planning, and so on. There is little evidence that these successes have as yet resulted in higher-level changes in risk-management practices or households' vulnerability. In particular, households that received early-warning information before the flooding of 2012 were no more likely to take key preventative actions. Perhaps as a consequence, the losses suffered by these households in that year were no different from those in comparison communities. It is possible that more time is needed for people to build trust in the early-warning information and community-level planning structures established under this project, and so incorporate them into their household-level decision-making.

Click here to download the report snapshot and summary documents.