Agricultural education in Africa using YouTube multilingual animations: A retrospective feasibility study assessing costs to reach language-diverse populations.

There is a critical need for widespread information dissemination of agricultural best practices in Africa. Literacy, language and resource barriers often impede such information dissemination. Culturally and linguistically localized, computer-animated training videos placed on YouTube and promoted through paid advertising is a potential tool to help overcome these barriers. The goal of this study is to assess the feasibility of reaching language-diverse populations in Africa using this new type of information dissemination channel. As a case study, cost estimates were obtained for YouTube ad campaigns of a video to prevent post-harvest loss through safe food storage using sanitized jerrycan containers. Seventy-three video variants were created for the most common 16 languages in Ghana, 35 languages in Kenya, and 22 languages in Nigeria. Using these videos, campaigns were deployed country wide or focused on zones of influence that represent economically underdeveloped regions known to produce beans suitable for jerrycan storage. Using data collected from YouTube ad campaigns, language-specific models were created for each country to estimate how many viewers could be reached per US dollar spent. Separate models were created to estimate the number of viewers who watched 25% and 75% of the video (most of video without end credits), reflecting different levels of engagement. For language campaigns with both country wide and zone of influence areas of deployment, separate region-specific models were created. Models showed that the estimated number of viewers per dollar spent varied considerably amongst countries and languages. On average, the expected number of viewers per dollar spent were 1.8 (Range = 0.2-7.3) for 25% watched and 0.8 (Range = 0.1-3.2) for 75% watched in Ghana, 1.2 (0.2-4.8) for 25% watched and 0.5 (Range = 0.1-2.0) for 75% watched in Kenya, and 0.4 (Range = 0.2-1.3) for 25% watched and 0.2 (Range = 0.1-0.5) for 75% watched in Nigeria. English versions of the video were the most cost-effective in reaching viewers in Ghana and Nigeria. In Kenya, English language campaigns ranked 28 (country wide) and 36 (zones of influence) out of 37 analyzed campaigns. Results also showed that many local language campaigns performed well, opening the possibility that targeted knowledge dissemination on topics of importance to local populations, is potentially cost effective. In addition, such targeted information dissemination appears feasible, even during regional and global crises when in-person training may not be possible. In summary, leveraging multilingual computer-animations and digital platforms such as YouTube shows promise for conducting large-scale agricultural education campaigns. The findings of the current study provides the justification to pursue a more rigorous prospective study to verify the efficacy of knowledge exchange and societal impact through this form of information dissemination channel.

Machine-supported decision-making to improve agricultural training participation and gender inclusivity.

Women comprise a significant portion of the agricultural workforce in developing countries but are often less likely to attend government sponsored training events. The objective of this study was to assess the feasibility of using machine-supported decision-making to increase overall training turnout while enhancing gender inclusivity. Using data obtained from 1,067 agricultural extension training events in Bangladesh (130,690 farmers), models were created to assess gender-based training patterns (e.g., preferences and availability for training). Using these models, simulations were performed to predict the top (most attended) training events for increasing total attendance (male and female combined) and female attendance, based on gender of the trainer, and when and where training took place. By selecting a mixture of the top training events for total attendance and female attendance, simulations indicate that total and female attendance can be concurrently increased. However, strongly emphasizing female participation can have negative consequences by reducing overall turnout, thus creating an ethical dilemma for policy makers. In addition to balancing the need for increasing overall training turnout with increased female representation, a balance between model performance and machine learning is needed. Model performance can be enhanced by reducing training variety to a few of the top training events. But given that models are early in development, more training variety is recommended to provide a larger solution space to find more optimal solutions that will lead to better future performance. Simulations show that selecting the top 25 training events for total attendance and the top 25 training events for female attendance can increase female participation by over 82% while at the same time increasing total turnout by 14%. In conclusion, this study supports the use of machine-supported decision-making when developing gender inclusivity policies in agriculture extension services and lays the foundation for future applications of machine learning in this area.

Dramatically increased accessibility and decreased cost-per-person impacts are needed for scaling IPM in Africa.

While traditional scaling for integrated pest management (IPM) in Africa requires the movement of expert trainers from village to village, these efforts are often costly, time-inefficient, hampered by distance, and became impossible under COVID-19's movement restrictions (despite tremendously increased public need for IPM-scaling knowledge). One solution to this dilemma is IPM-scaling, usable by a diversity of development actors expending limited or few resources, to deliver critical information to large numbers of people with systems-approach information and communication technologies. This paper describes one such systems-approach scaling platform, Scientific Animations Without Borders, which effectively elicited end-user solution-adoption and decreased unit costs over increasing scales in three African countries during COVID-19. How to scale game-changing IPM insights 'off the shelf' and 'into people's hands in the field' is also discussed.

Just participation or just participation? A participatory justice model for more successful theory of change design, implementation, and solution uptake.

While a wide consensus acknowledges that participation is critical for the successful implementation of change that improves the livelihoods of people and communities around the world, justly securing that participation from stakeholders (at both the design and implementation stages) remains a demanding problem. This paper proposes a heuristic model for increasing participation that not only helps to investigate instances of nonparticipation but also opens up alternative intervention strategies and pathways for designers and implementers to consider toward more justly increasing participation and overcoming nonparticipation. Applied to a successful case of participation in Gurúè District, Mozambique-where an 89% solution adoption of an improved postharvest seed storage method was measured two years after initial training-this paper demonstrates the key importance of designing opportunities and motivations for participation into any solutions or innovations but especially justice as a factor for successful realization of theory of change efforts (all the more so in developing nation contexts). Applied to a second case study, this paper also explores participation despite little to no motivation to do so. Aiming to afford designers and implementers of theory of change interventions a tool for more successfully and exactly matching innovation goals with innovation outcomes, the paper also addresses broader implications for the model within theory of change approaches generally.

Large-scale rollout of extension training in Bangladesh: Challenges and opportunities for gender-inclusive participation.

Despite the recognized importance of women's participation in agricultural extension services, research continues to show inequalities in women's participation. Emerging capacities for conducting large-scale extension training using information and communication technologies (ICTs) now afford opportunities for generating the rich datasets needed to analyze situational factors that affect women's participation. Data was recorded from 1,070 video-based agricultural extension training events (131,073 farmers) in four Administrative Divisions of Bangladesh (Rangpur, Dhaka, Khulna, and Rajshahi). The study analyzed the effect of gender of the trainer, time of the day, day of the week, month of the year, Bangladesh Administrative Division, and venue type on (1) the expected number of extension event attendees and (2) the odds of females attending the event conditioned on the total number of attendees. The study revealed strong gender specific training preferences. Several factors that increased total participation, decreased female attendance (e.g., male-led training event held after 3:30 pm in Rangpur). These findings highlight the dilemma faced by extension trainers seeking to maximize attendance at training events while avoiding exacerbating gender inequalities. The study concludes with a discussion of ways to mitigate gender exclusion in extension training by extending data collection processes, incorporating machine learning to understand gender preferences, and applying optimization theory to increase total participation while concurrently improving gender inclusivity.

Breaking out: the turning point in learning using mobile technology.

Despite considerable research on YouTube as a digital media platform, little research to date has quantified the device-type used to access that online media. Analyzing access-device data for videos on one YouTube video channel-Scientific Animations Without Borders (SAWBO), which produces educational content specifically accessible to low- or non-literate, poor, or geographically isolated learners in less developed areas of the world-the results identify the historical moments between 2015 and 2017 when mobile/smartphones, both globally and by region, crossed a tipping point to surpass all other ICT devices (including desktop PCs, laptops, and other Internet-accessing technologies) as the primary device-type for accessing SAWBO videos. Specifically, data from January 2013 to June 2018 obtained for SAWBO's YouTube channel were sampled to capture and distinguish the access device-type used and then summarized in broad global and regional categories. The tipping point, as the date where the percentage of views from mobile phones was equivalent to the percentage of views from computers, were also calculated globally and by region. Besides documenting this critical global-historical moment, the results also have implications for mass digital-messaging generally and mobile-based public service learning specifically.