Questions & explanations
1. What challenges might arise when implementing PM&E, and how can they be addressed?
Challenges include farmers' time constraints, low literacy, and power dynamics where dominant voices overshadow others. To address time, integrate PM&E into regular meetings rather than extra sessions. For low literacy, use visual tools like drawings, symbols, or oral reporting. To handle power dynamics, use techniques like separate groups for women and men, or anonymous voting. Another challenge is that farmers may give socially desirable answers. Build trust by ensuring no negative consequences for honest feedback. Also, facilitators need training in participatory methods. Finally, donors may want standardized data. Negotiate with donors to accept a mix of participatory and quantitative data. For example, use PM&E for learning and supplement with a simple survey for key indicators.
2. How do you combine qualitative and quantitative data in a mixed methods analysis?
One common way is 'triangulation': compare findings from both types to see if they agree. For example, if survey data shows high adoption, but interviews reveal farmers are not using the practice correctly, the adoption rate may be misleading. Another way is 'complementarity': use qualitative data to explain quantitative results. For instance, if a survey finds no yield increase, interviews might show that farmers used the seeds but faced a pest outbreak. Also, 'expansion': use one method to explore questions raised by the other. For example, after a survey finds that women adopt less, conduct interviews to understand barriers. Finally, 'development': use qualitative findings to design a better survey. The key is to integrate findings in the discussion, not keep them separate.
3. What are the main threats to validity in quasi-experimental designs, and how can they be addressed?
Main threats include selection bias (treatment and control groups differ), history (external events affect one group), and maturation (natural changes over time). For example, if the treatment village has better soil, yield differences may not be due to extension. To address selection bias, use propensity score matching: match each treated farmer with a similar untreated farmer based on observable characteristics. For history, use a control group that is as similar as possible and collect data on external events. For maturation, use a difference-in-differences method: compare the change in treatment group to change in control group. Also, collect baseline data before the program. Randomization is best but often not feasible, so these methods help approximate causal impact.
4. What are common challenges in managing a PPP for extension, and how can they be addressed?
Common challenges include conflicting goals, unequal power, and lack of trust. Private partners may prioritize profit over farmer welfare, while public partners may be slow and bureaucratic. To address this, a clear memorandum of understanding should define roles, funding, and dispute resolution. Regular joint meetings help align objectives. Another challenge is sustainability: when private funding ends, services may stop. To avoid this, build farmer groups that can pay for some services or link to credit. Also, monitor outcomes independently to ensure both sides benefit. For example, if a private company provides free training but then offers expensive inputs, farmers may become dependent. Transparent contracts and farmer feedback mechanisms can prevent exploitation.
5. What is participatory monitoring and evaluation (PM&E) in extension?
Participatory monitoring and evaluation (PM&E) is a process where farmers and other stakeholders actively help to track and assess an extension program. Unlike traditional M&E done by outsiders, PM&E involves farmers in setting indicators, collecting data, and making decisions. For example, farmers might decide to measure 'number of new practices tried' instead of just 'yield'. They collect data through simple records or group discussions. Then they analyze results together and suggest improvements. PM&E builds ownership and accountability because farmers see their own progress. It also provides local knowledge that outsiders might miss. However, it requires training and time. PM&E is often used in farmer field schools and community-based programs.
6. How does a PPP extension model differ from a purely public extension system?
In a purely public system, the government funds and delivers all extension services, often free to farmers. A PPP model involves a private partner that may charge for some services or expect something in return, like crop purchases. Public systems aim for broad coverage and equity, but may lack resources and innovation. PPPs can bring private sector efficiency, technology, and market links. However, PPPs may focus on commercially viable farmers, neglecting poorer ones. Public systems are accountable to citizens, while PPPs are accountable to shareholders. Therefore, PPPs need strong regulation to ensure public goals are met. For example, a public system might teach general crop management, while a PPP might focus on a specific crop for export.
7. How can you compare the cost-effectiveness of two different extension methods using CBA?
To compare two methods, calculate the benefit-cost ratio (BCR) or cost per unit of outcome for each. For example, method A is farmer field schools costing $30,000 with BCR 1.67; method B is mass media campaigns costing $10,000 with BCR 1.2. Method A has higher BCR, but method B reaches more farmers per dollar. So you also compute cost per farmer adopting a practice: if method A costs $600 per adopter and method B costs $100 per adopter, method B is more cost-effective for adoption. But consider quality of adoption: method A may lead to deeper learning. So use multiple criteria: BCR, cost per adopter, and sustainability. The best method depends on goals: if budget is tight, choose method B; if long-term change is needed, method A may be better.
8. What is a quasi-experimental design, and why is it used in extension research?
A quasi-experimental design compares a group that receives an extension program (treatment) with a group that does not (control), but without random assignment. Random assignment is often impossible because programs target specific areas. For example, a new training is offered in one village, and a similar village is chosen as control. Researchers then measure outcomes like yield before and after, and compare changes between the two villages. This helps estimate the program's impact. However, without randomization, the groups may differ in ways that affect results (selection bias). To reduce bias, methods like propensity score matching or difference-in-differences are used. Quasi-experiments are practical for real-world extension evaluation.
9. Give an example of a mixed methods study evaluating an extension program.
Suppose an extension program introduces drought-tolerant seeds. The quantitative part: a survey of 200 farmers (100 who received seeds, 100 who did not) measures yield and income. The qualitative part: in-depth interviews with 20 farmers from each group explore how they used the seeds, their perceptions, and challenges. The survey shows a 15% yield increase. The interviews reveal that some farmers did not use recommended planting spacing because they lacked labor. So the program could add training on labor-saving techniques. Mixed methods thus explain the numbers and suggest improvements. The study might also include focus groups with women to understand gender differences. This combination provides evidence for both impact and process.
10. How does the Harvard Analytical Framework help identify gender roles in farming?
The Harvard Analytical Framework uses four tools: activity profile, access and control profile, influencing factors, and project cycle analysis. The activity profile lists who does what tasks, like planting, weeding, and marketing. The access and control profile shows who has access to resources like land, credit, and extension services, and who controls them. For example, men may control land even if women work on it. Influencing factors include cultural norms and laws. By filling these profiles, extension planners see that women often have less access to training. They can then schedule training when women are free and use female extension agents. This framework is simple and widely used, but it may not capture power dynamics deeply.
11. What are the limitations of using ROI to evaluate extension programs?
ROI focuses only on monetary benefits, ignoring important outcomes like environmental health, women's empowerment, or social capital. For example, a program that reduces water pollution has no direct market value but is valuable. Also, ROI assumes all benefits can be measured accurately, but extension impacts are often diffuse and long-term. Farmer income may increase due to other factors like good weather, not just extension. ROI also ignores distribution: benefits may go to richer farmers while poorer ones gain little. Moreover, costs like farmer time are often undervalued. To address these, combine ROI with qualitative methods, like interviews, to capture non-monetary impacts. Also, use sensitivity analysis to test assumptions.
12. What are the benefits of PM&E compared to traditional M&E?
PM&E gives farmers a voice, so the program reflects their real needs. Traditional M&E often collects data that outsiders think is important, like yield, but ignores social impacts like women's confidence. PM&E captures these because farmers choose indicators. It also builds capacity: farmers learn to analyze data and make decisions. This leads to more sustainable changes because farmers own the process. Additionally, PM&E increases accountability: farmers can question why a program is not working. Traditional M&E may be seen as a top-down report. However, PM&E takes more time and may be less rigorous statistically. Combining both approaches often works best: use PM&E for learning and traditional M&E for reporting to funders.