ICT4BXW BASELINE REPORT OCT. 2018 ICT4BXW REPORT ON BASELINE SURVEY OF BANANA FARMERS IN RWANDA 1 ICT4BXW BASELINE REPORT OCT. 2018 Table of Contents Table of Contents .................................................................................................................................... 2 Introduction ............................................................................................................................................ 3 The baseline survey................................................................................................................................. 3 Content of the survey ......................................................................................................................... 3 Sampling strategy.................................................................................................................................... 4 Selection of districts ........................................................................................................................... 4 Selection of sectors, cells, and villages ............................................................................................... 6 Sector selection .............................................................................................................................. 6 Cell selection .................................................................................................................................. 6 Village selection ............................................................................................................................. 7 Sampling of households ...................................................................................................................... 7 Sampled villages where the survey is conducted ............................................................................... 7 Baseline Survey of Farmer Promoters ................................................................................................ 8 Tentative Results from Farmers Survey .................................................................................................. 9 Demographics ..................................................................................................................................... 9 Assets ................................................................................................................................................ 10 Farming activities .............................................................................................................................. 12 Status of BXW ................................................................................................................................... 13 Knowledge of BXW and control – and prevention measures ........................................................... 15 Access to information ....................................................................................................................... 21 ICT assets and interest in ICT-based agricultural services ................................................................ 23 Tentative Results from farmer promoters Survey ................................................................................ 28 Demographics ................................................................................................................................... 28 Education and professional experience of farmer promoters ......................................................... 28 Expertise of farmer promoters ......................................................................................................... 30 ICT assets and interest in ICT-based agricultural services ................................................................ 31 APPENDIX 1 ........................................................................................................................................... 34 Annex 1: overview of mobile internet signals and symbols ................................................................. 49 2 ICT4BXW BASELINE REPORT OCT. 2018 Introduction The ICT4BXW project was officially launched in January 2018, with funding from the German development cooperation (BMZ). The goal of this project is to use citizen science and ICT to develop (cost)-effective and scalable tools for advancing the prevention and control of Banana Xanthomonas Wilt (BXW) in East and Central Africa, with initial focus on Rwanda. The project is centred on 2 major objectives – (1) To provide up-to-date knowledge on BXW diseases to extension service providers and a decision-support tool to identify labour- and cost-effective BXW Control methods; (2) Provide real- time data on BXW spread that supports governments in targeting their BXW prevention efforts in a more cost-effective way. Based on the project’s implementation framework, we conducted a baseline survey of farmers and farmer promoters (FPs) in Rwanda. The survey provides a critical input to meet the objectives and demonstrate impact over time. In the first stage (19th June – 3rd July 2018) a team of representatives from IITA and RAB’s banana programme visited eight selected districts in four of Rwanda’s provinces to select banana producing villages which would be enlisted in the survey. By design, half of the selected villages were earmarked as intervention villages for the project, and the other half were designated as control villages which would be visited two times during the project’s lifespan: once during the baseline survey, and once during the end-of-the-line survey. In this report, we provide broad overview of the sampling strategy for sector, cell and village selection in each district, brief insights from the data collected, and field notes/observations on the sampling and survey process. The baseline survey The ICT4BXW questionnaire survey was collaboratively developed, with contributions from project partners and thematic collaborators. The finalized questionnaire was formatted for deployment on mobile-based Open Data Kit (ODK) software through its ODK-Collect application. The ODK is an open- source software for collecting, managing, and using data. The software was deployed on Tablet mini- computers and this offered flexibility of collecting data with or without active internet or phone connection, and upload into cloud-based server, when internet connection exists. The seamless collection, upload, and access of data is beneficial for rapid checking of data quality and analyses. Content of the survey The ICT4BXW baseline survey covered a broad range of topics related to the banana farming system, rural livelihoods, and penetration of digital technology among smallholder farmers. Our goal was to capture a variety of relevant information that will be valuable for the project’s research, monitoring, and evaluation needs. The content of the ICT4BXW baseline survey is as follows: • Introduction of the project • Crops grown, sold and consumed, • Location data especially banana, cassava and soya • Demographics of respondent • Agricultural Inputs • Demographics of household head • Off-farm income • Household population • Food security in past year • Household assets, including livestock • Food nutrition in past 24 hours • Access to credit • Banana disease experience and • General farm land size knowledge • Banana farm land size • Cost-benefit of different BXW control 3 ICT4BXW BASELINE REPORT OCT. 2018 options • Use of mobile agriculture services • BXW prevention strategies (Inge???) • Impact of mobile phone use • Extension in general and for banana • Measurement of incidence in specifically currently BXW affected plot • Experience with ICT • Registration of respondent for follow • Experience with and use of mobile up phones • Assessment of survey quality Sampling strategy Selection of districts The baseline survey targeted 8 districts (Burera, Rulindo, Gatsibo, Kayonza, Gisangara, Muhanga, Karongi and Rubavu), within 4 provinces in Rwanda. The two criteria for selecting districts include; (1) coverage/representation of the major agro-ecological zones in Rwanda, and (2) representation of different types of banana producing farmers (Table 1). Together, the eight selected districts represent diversity in production typologies ranging from small-scale subsistence production to large-scale commercial production. Selection of districts was conducted based on expert knowledge (mainly through multiple consultations of Banana Programme Leader at RAB), and raw data from a countrywide rapid assessment of BXW status conducted by RAB in 2017-2018. Table 1: Selected districts per province and their agroecological zones Province Agro-ecological zone Districts selected Northern province Northern highlands 1. Rulindo 2. Burera Eastern province Eastern savannah 1. Kayonza 2. Gatsibo Southern province Central plateau 1. Gisagara 2. Muhanga Western province Kivu lake border 1. Rubavu 2. Karongi At district level, BXW incidence levels and distance to extension district headquarters were classified into 3 levels (Table 2). This resulted in 9 strata which guided the sampling of villages. We hypothesize that distance to the district headquarters affects access to agricultural/banana extension services. BXW incidence levels were considered important because the project aims to see reduction in disease outbreaks and need for mat uprooting, as well as improvements in banana management and BXW disease control and prevention. In each district 18 villages were sampled, 9 intervention and 9 control villages, according to the 9 strata. However, in Karongi district the number of villages was limited to 12 (6 intervention and 6 control villages). This was due to the absence of village(s) that match(es) the criteria for long distance to district headquarters. Therefore, the total number of villages included in the baseline was 138 (i.e. (7x18)+(1x12)). 4 ICT4BXW BASELINE REPORT OCT. 2018 Table 2: Sampling scheme based on disease incidence levels, distance to district headquarters, and control versus intervention. BXW incidence level Medium Low incidence High incidence incidence Village 1: Village 3: Village 5: Short Intervention Intervention Intervention distance Village 2: control Village 4: control Village 6: control Village 7: Village 9: Village 11: Medium Intervention Intervention Intervention distance Village 8: control Village 10: Village: 12: control control Village 13: Village 15: Village 17: Large Intervention Intervention Intervention distance Village 14: Village 16: Village 18: control control control 5 Distance from district headquarters ICT4BXW BASELINE REPORT OCT. 2018 Selection of sectors, cells, and villages Sector selection In each district sectors were selected based on expert input from the district and sector agronomists. The total number of sectors from which villages were selected ranges between three and five depending on the district. The selection criteria include: 1.) Distance between the sector and the district headquarters (Figure 1), 2.) Dominance of banana in the sector (the sector had to be considered a banana producing area, meaning that banana is produced for both household consumption and commercial value) and 3.) presence of BXW: the disease had to be present in the sector. There were no restrictions to the total number of villages that could be sampled from one sector. Figure 1: Map with overview of all districts from which villages were sampled. Colour codes represent short (blue), medium (yellow) and long (red) distance to the district's headquarters. Cell selection Cells were selected based on the same criteria as the sectors, based on expert input from sector and cell agronomists. The sampling team aimed for a good balance between geographical spread in the district and efficient access to the cells. There were no restrictions to the total number of villages that could be sampled from one cell. 6 ICT4BXW BASELINE REPORT OCT. 2018 Village selection Three criteria were considered when selecting villages: (1) distance between the village and the district headquarters (figure 1). A three-point scale was used (close, medium, far). (2) Level of BXW incidence. A three-point scale was used (low, medium, high). Incidence levels were determined based on reports from sector and cell agronomists and field observations from RAB banana experts and technicians when passing through the village. Incidence levels are fluid and there is a chance that some villages move up or down on the scale between the time of sampling and the time of survey execution. This is due to ongoing disease eradication efforts (organized mat and plantation uprooting) and potential increase in incidence levels due to continued poor management and/or insect vectors. (3) Distance between villages. The sampling team aimed for selection of villages with a minimum distance of 5km or a non-intervention and non-control village in between two selected villages. In some districts these requirements were relaxed to reach the required total number of villages. Villages were randomly assigned to the intervention or control group. Sampling of households In both intervention and control villages a total of five households were sampled based on the gender of the household head. A weighted sample was taken from each stratum: 60% from the male-headed (3 households) and 40% from the female-headed stratum (2 households). Male enumerators were assigned to survey the male-headed households, and female enumerators surveyed female-headed households (Figure 2a). Figure 2: Graphical representation of household sampling strategy. Sampled villages where the survey is conducted A simple protocol was followed to sample villages, considering three levels of distance to the district headquarters (short, medium, large), and three levels of BXW incidence at village level (low, medium, high). For each category, two villages were sampled, one was the intervention village and the other was designated as the control village. In total there should be 72 intervention villages (9 per district) and 72 control villages (9 per district). However, due to unavailability of villages that 7 ICT4BXW BASELINE REPORT OCT. 2018 falls within the ‘long distance to the district headquarters’ category in Rubavu, the baseline was conducted only in 69 intervention and 69 control villages. In each village, 5 households were interviewed, resulting in a total of 690 households surveyed (Figure 2b). Further details on the overview of the villages selected in each district, including observations from field visits can be found in the Appendix section of this report. N Figure 2b: Map showing villages where baseline survey of banana farmers was conducted in Rwanda Baseline Survey of Farmer Promoters Upon completion of baseline survey of farmers, a separate survey of FPs (Farmer Promoters) was conducted to assess their technical knowledge about BXW, ICT competencies and perspectives on delivery of extension service among farmers for BXW control. The major goal is to understand the critical gaps existing in the expertise and capabilities of the “next-users” of anticipated BXW tool and build on the major insights during the participatory and inclusive technology design (PITD) process. The survey, which lasted from late August to early September 2018, was conducted remotely over phone call and included 136 FPs across the surveyed villages. Following similar approach adopted for the farmers’ survey, questionnaire was developed and deployed on smartphone-based ODK, and the enumerator recorded each FP’s response to each question on the ODK form during each phone call session. The FPs within the intervention villages were assigned unique barcode IDs and cards which can be used to link information as the project intervention progresses. 8 ICT4BXW BASELINE REPORT OCT. 2018 Tentative Results from Farmers Survey Demographics The household survey covered 692 respondents, with 40% female and 60% male. Most of the surveyed respondents are in their middle ages (30 – 60 years;Figure 3). The size of the household varies between 1 and 18, with an average of 5 (Figure 4). Most respondents speak only Kinyarwanda, although some are knowledgeable in a second language, usually French (10%; Figure 5). With regards to education, 85% of the respondents has an education level of primary school or lower. Age classes of respondents differentiated by gender 35% 30% Female respondents 25% Male respondents 20% 15% 10% 5% 0% 20-30 31-40 41-50 51-60 61-70 71-80 81-90 Age class (years) Figure 3: Respondents' age differentiated by gender. Distribution of household size 18 15 12 10 8 6 4 2 0 0 20 40 60 80 100 120 140 Number of households Figure 4: Distribution of household size. 9 Household size ICT4BXW BASELINE REPORT OCT. 2018 Languages spoken by farmers 100% 80% 60% 40% 20% 0% English French Kinyarwanda Kiswahili Figure 5: Percentages of farmers knowledgeable in English, French, Kinyarwanda and Kiswahili. Assets On average, farmers own 85% of the land they cultivate with the average cultivated land being 2,6 ha. Male farmers more often own land than female farmers, indicating that female farmers tend to rent land in addition to the land in their possession (Figure 6). Majority of the farmers (69%) have no means of transportation (Figure 7). More than half of the farmers has access to credit, in most cases provided by microfinancing (Figure 8). Distribution of proportion of owned land differentiated by gender 80% 70% 60% 0% 50% 5% 40% 25% 30% 50% 20% 75% 10% 100% 0% Female respondents Male respondents Figure 6: Distribution of the land farmers possess differentiated by gender. 10 ICT4BXW BASELINE REPORT OCT. 2018 1% Means of transportation 3% 0% No transport 27% Bicycle Motorbike Wheelbarrow 69% Car Figure 7: Means of transportation in possession of the surveyed farmers. Source of credit 1% 0% 6% No credit Formal credit Informal credit 46% Microfinance 39% NGO Friends/family Other 4% 4% Figure 8: Access to and source of credit. 11 ICT4BXW BASELINE REPORT OCT. 2018 Farming activities Most of the farmers own animals, mainly Cattle and goats (64% and 43%, respectively; Figure 9). Number of animals per farm are however low with an average of 1,6 cattle and 2,66 goats. The most commonly cultivated crop is food banana, followed by beer banana. It can be noticed that beer banana is more commonly cultivated by male farmers (Figure 10). The required labour for banana production is usually provided by the household head (mainly males) and supplemented by labour contribution of the spouse. In addition, 41% of the farmers hire additional labour (Figure 11). Possession of animals 70% 60% 50% 40% 30% 20% 10% 0% Figure 9: Percentages of farmers owning cattle, goats, sheep, pigs, rabbits, chicken, bees, fish, Guinea pigs or other animals. Principal crop female Principal crop male respondents 0.36% respondents 0.36% 0.48% 1.09% 4.36% 13.04% 0.48% amaranth avocado 23.73% bush beans beerbananas 85.14% 70.94% foodbananas Figure 10: Principal crops cultivated by female and male farmers. 12 ICT4BXW BASELINE REPORT OCT. 2018 Labour for banana production 100% 80% 60% 40% 20% 0% Household Male Female Reciprocal Hire head spouse spouse labour (M/F) Figure 11: Source of labour for banana production. Status of BXW 67% of the farmers has ever experienced BXW in his/her farm and 59% did so in the past year (Figure 12). No difference can be recognized between male and female farmers. Except for BXW, the most common disease on banana is Fusarium wilt (Figure 13). Infection in a farm seems to be strongly correlated with infection in neighbouring farms, as 95% of the infected farms also have neighbouring farms which are infected (Figure 14). Farmers experiencing BXW 80% 70% 60% 50% 40% Yes 30% No 20% 10% 0% Ever experienced BXW Exeperienced BXW in the past year Figure 12: Percentages of farmers having ever experienced BXW and having experienced BXW during the past year. 13 ICT4BXW BASELINE REPORT OCT. 2018 Experience with other pests and diseases in banana 25% 20% 15% 10% 5% 0% Fusarium wilt Bunchy top Weevils Nematodes Other disease Figure 13: Pests and diseases on banana, other than BXW, ever experienced by farmers. Neighbours' experience BXW compared to farmers' experience Yes 94.62% 5.38% yes No No 44.00% 56.00% 0% 20% 40% 60% 80% 100% Neighbours experienced BXW in past 2 years? Figure 14: Relationship between ever infected farms and infection in neighbouring farms during the past 2 years. 14 Farmer EVER experienced BXW? ICT4BXW BASELINE REPORT OCT. 2018 Knowledge of BXW and control – and prevention measures Most of the farmers seems to be well-aware of prominent symptoms of BXW. The very early symptom of leaf discoloration is only known by 27% of the farmers (Figure 15). No strong differences occur between the knowledge of male and female farmers. Farmers with a history of experiencing BXW do however tend to have a better knowledge as indicated in Figure 16 for ‘brownfruit’ symptom. With regards to knowledge on control methods, results strongly differ. Certain methods (CMU and burning/destroying) are well known whereas others (SDSR and CFU) are known by less than a quarter of the consulted farmers (Figure 17). As could be seen for symptoms, male farmers have a better knowledge on BXW control methods (Figure 18). This is translated in the control farmers perceive in the decision process of controlling BXW. 22% of female respondents indicates to have no control in this process, whereas this is only the case for 5% of the male respondents (Figure 19). Knowledge on BXW symptoms 100% 80% 60% 40% 20% 0% Brownfruit Withering of Oozing or Leaf Early fruit Fruit rotting leaves debris discoloration ripening Figure 15: farmers' knowledge of both early and late symptoms of BXW. 15 ICT4BXW BASELINE REPORT OCT. 2018 Knowledge on BXW symptoms differentiated by BXW experience Ever experience 50.54% 49.46% Yes d BXW No Never experience 29.78% 70.22% d BXW 0% 20% 40% 60% 80% 100% 120% Farmer knows 'brownfruit' symptom? Figure 16: Farmers' knowledge on ‘brown fruit’ symptom, differentiated by gender. Knowledge on BXW control methods 70% 60% 50% 40% 30% 20% 10% 0% Figure 17: Farmers' knowledge on single-diseased stem removal (SDSR), complete mat uprooting (CMU), complete field uprooting (CFU), burning and destroying of infected stems and other control methods. 16 ICT4BXW BASELINE REPORT OCT. 2018 Knowledge on control method differentiated by gender Male 35.84% 64.16% respondents Yes No Female 13.04% 86.96% respondents 0% 20% 40% 60% 80% 100% Farmer knows SDSR? Figure 18: Farmers' knowledge on SDSR as a control method, differentiated by gender. Perceived control of respondent in deciding how to control BXW 40% 30% 20% 10% 0% Female respondents Male respondents Complete control More control Equal control Some control No control Figure 19: Gender-related perceived control by respondents in the decision process of controlling BXW. Knowledge on the causes of BXW is limited. Only 57% of the farmers is aware that insects and tools can transmit BXW and less than 20% is aware of other important transmitters like animals and infected plant material (Figure 20). These results are translated in farmers’ knowledge on prevention measures, as can be seen in Figure 21. The most well-known prevention methods are male flower debudding and tool-related practices, whereas very few farmers are aware of the importance of selecting healthy plant material or preventing the roaming of animals. Whilst the knowledge on causes and prevention methods of BXW is similar for male and female farmers, control in decision process is not equal. Female respondents indicate to have less control in deciding on prevention measures than male respondents (Figure 22). 17 ICT4BXW BASELINE REPORT OCT. 2018 Knowledge on causes of BXW 60% 50% 40% 30% 20% 10% 0% Figure 20: Farmers' knowledge on potential causes of BXW. Comentado [mv1]: Should I add a graph on the Knowledge on BXW prevention measures APPLICATION of prevention measures? It gives the exact same trend as can be seen in this one. 0.60 0.50 0.40 0.30 0.20 0.10 0.00 Figure 21: Farmers' knowledge on BXW prevention measures. 18 ICT4BXW BASELINE REPORT OCT. 2018 Perceived control of respondent in deciding how to prevent BXW 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% Female respondents Male respondents Complete control More control Equal control Some control No control Figure 22: Gender-related perceived control by respondents in the decision process of preventing BXW. The most important overall challenge for farmers to implement BXW prevention and control measures seems to be a lack of knowledge on these measures (Figure 23). Female and male farmers however experience different challenges. Male farmers indicate labour, cost and access to material as important challenges as well, whereas these are not at all perceived as important challenges by female farmers (Figure 24). As indicated earlier, few farmers are aware of the importance of selecting healthy plant material. Accordingly, it is common practice to share and receive plant material and not buying and selling it ( Figure 25). This is true for both male and female farmers. When looking into the source of plant material, the usual origin is through a fellow farmer (Figure 26). Challenges for BXW control and prevention in order of importance 40% 35% 30% 25% 20% 15% 10% 5% 0% Knowledge Labour Other Product Cost Time Access to loss material Figure 23: Most important challenges for the implementation of BXW control and prevention methods, as indicated by farmers. 19 ICT4BXW BASELINE REPORT OCT. 2018 Most important challenges for BXW control and prevention differentiated by gender 25% 20% 15% 10% 5% 0% Labour Cost Time Production Access Knowledge Other loss material Female respondents Male respondents Figure 24:Most important challenges for the implementation of BXW control and prevention methods, as indicated by male and female farmers. Received/shared plantmaterial in the past 2 years 39.28% yes 60.72% no Figure 25: Percentage of farmers having received or shared banana plant material during the past 2 years. Source of banana plantmaterial 50% 40% 30% 20% 10% 0% Bought from Bought from bought from Received from Received from Received from another farmer cooperative NGO another farmer NGO gov Figure 26: Sources of farmers' banana plant material. 20 ICT4BXW BASELINE REPORT OCT. 2018 Access to information 61% of the consulted farmers has ever received extension services (Figure 27). No notable difference between male and female farmers exists. The most important providers of extension services, in case received, are government agronomists, farmer promoters and Farmer Field School facilitators (Figure 28). In addition to extension services, farmers receive information from a group. 49% of the respondents is member of a group (Figure 29). Nonetheless, 73% of the farmers identifies the radio as the most important source of agriculture-related information (Figure 30). Other important sources are extensionists (66%) and fellow farmers (36%). Access to extension services Never received extension 38.84% services received extension 61.16% services Figure 27: Percentage of farmers having received extension services. Providers of received extension services 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% Gov RAB NGO FFS Farmer Agrodealer agronomist facilitator promotor Figure 28: Sources of the received extension services. 21 ICT4BXW BASELINE REPORT OCT. 2018 Group membership 60% 50% 40% 30% 20% 10% 0% Farmer Field Twigire group Womens' Cooperation None School group Figure 29: Farmers' membership of an agriculture-related group. Sources of information on BXW control and prevention in order of importance 80% 70% 60% 50% 40% 30% 20% 10% 0% Figure 30: Most important sources of the information farmers receive on BXW control and prevention measures. 22 ICT4BXW BASELINE REPORT OCT. 2018 ICT assets and interest in ICT-based agricultural services Radio and basic phone are the most commonly used ICT-devices used by farmers (Figure 31). One out of every ten of the farmers has not used any ICT-device within 30 days prior to the survey. Focusing on phones, 72% of the farmers owns a basic phone and 3% own smartphone(s) (Figure 32) while the most common subscribed service provider is MTN (Figure 33).The likelihood of owning a phone is clearly determined by both gender and age. Male and young farmers are most often in possession of a phone (Figure 34; Figure 35). About 1 in 5 farmers has used a phone for any agriculture-related service and this is again most common among young farmers (Figure 36). Use of ICT-devices in past 30 days 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Figure 31: ICT-devices used by farmers during the past 30 days. Phone ownership 3% 28% Smartphone Basic None 72% Figure 32: Phone ownership among farmers. 23 ICT4BXW BASELINE REPORT OCT. 2018 Sim card provider 100% 50% 0% MTN TIGO Airtel Figure 33: SIM card service providers on farmers' phones. Phone ownership differentiated by gender Male 79% 21% respondents Female 63% 37% respondents 0% 20% 40% 60% 80% 100% Owns a phone Does not own a phone Figure 34: Percentages of male and female farmers owning a phone. Phone ownership differentiated by age 81-90 33% 67% 71-80 60% 40% 61-70 62% 38% 51-60 70% 30% 41-50 74% 26% 31-40 81% 19% 20-30 77% 23% 0% 20% 40% 60% 80% 100% Owns a phone Does not own a phone Figure 35: Percentages of farmers in different age classes that owns a phone 24 Age class (years) ICT4BXW BASELINE REPORT OCT. 2018 Use of phone for agricultural services in past 30 days 81-90 100% 71-80 11% 89% 61-70 11% 89% 51-60 12% 88% 41-50 21% 79% 31-40 26% 74% 20-30 21% 79% 0% 20% 40% 60% 80% 100% Yes No Figure 36: Percentages of farmers in different age classes which have used their phone for agricultural services during the past 30 days. Three out of every five farmers believed that electronic devices could help in preventing and controlling BXW. Nonetheless, they currently experience several barriers to the usage of such services, the most important barriers include non-awareness of mobile-based tool existence and lack of know-how on their use (Figure 37). For farmers who indicated that they would use mobile-based services, their preferred mode of communication were voice calls (82%), radio (52%) and SMS (32%) (Figure 38). Majority of the farmers (72%) indicated willingness to pay for such services and the amount that they would be willing to spend varies between 100 and 600 RWF (Figure 39; Figure 40). Barriers to use of ICT-based agricultural services in order of importance 70% 60% 50% 40% 30% 20% 10% 0% Figure 37: Most important barriers to using ICT-based agricultural services as indicated by farmers. 25 Age class (years) ICT4BXW BASELINE REPORT OCT. 2018 Preferred channels for exchange of information 100% 80% 60% 40% 20% 0% Voice calls Radio SMS Voice Video Social USDD message media Figure 38: Channels farmers would prefer to receive ICT-based agricultural services. Willing to pay for ICT-based agricultural services no 28% yes 72% Figure 39: Percentage of farmers that are willing to pay for ICT-based agricultural services. Willingness to pay for ICT-based agricultural services >1000 600-1000 300-600 100-300 <100 0% 10% 20% 30% 40% 50% Figure 40: Farmers' willingness to pay for ICT-based agricultural services in monetary terms. 26 Amount willing to pay (RWF) ICT4BXW BASELINE REPORT OCT. 2018 BXW incidence and initial heat map Approximately 60% of the farmers reported that they have observed BXW in their farm within the past 12 months, and similar percentage indicated that they have observed BXW in the past 2 years and 6 years. Heat map rendition of the reported BXW frequency (Figure 41a-c) suggests that the severity of BXW may have changed within the surveyed districts when comparing BXW incidence rate in the past 2 years to past 12 months. The severity of BXW in the past 2 years seems comparable to the past 6 years, with prominent hot-spots in the north-central and south-eastern districts. a. Past 12 Months N b. Past 2 Years c. Past 6 Years Figure 41a-c: Heat map showing hotspots of reported BXW incidence among surveyed Banana farmers 27 ICT4BXW BASELINE REPORT OCT. 2018 Tentative Results from farmer promoters Survey Demographics The farmer promoters’ survey was carried out among 138 farmer promoters, 112 male and 26 female respondents. Majority of the FPs (~57%) are 30-50 years old, while 38% are over 50 years old, and the rest are less than 30 years old. Similar aged-based distribution was observed in both gender (¡Error! No se encuentra el origen de la referencia.2). Age categories of female FP Age categories of male FP 20 70 63 16 60 15 50 43 9 40 10 30 5 20 1 10 6 0 0 a[0-30] b[30-50] c[50-> a[0-30] b[30-50] c[50-> Age class (years) Age class (years) Figure 42: Age distribution differentiated by gender. Education and professional experience of farmer promoters Majority of the farmers (66%) have attended primary school (Figure 413) and the distribution of educational status was similar among male and female FPs (Figure 424). Most the FPs (64%) have acquired more than 4 years of professional experience (Figure 435), although this is differentiable by gender. A highest proportion (38%) of the female FPs have less than 2 years of experience, in contrast to male FPs who mostly have greater than 6 years’ experience. This reflects the recent inflow of women into the profession of female promoters (Figure 446). Education level of FP 100 91 Lower_sec 80 no_school 60 Primary 40 Upper_sec_gen 15 18 20 11 Upper_sec_teach 2 1 0 Upper_sec_tech FP Figure 413: Education level of farmer promoters: lower secondary (Lower_sec), no school (no_school), primary (Primary), general upper secondary (Upper_sec_gen), teachers’ upper secondary (Upper_sec_teach), technical upper secondary (Upper_sec_tech). 28 ICT4BXW BASELINE REPORT OCT. 2018 Education level of female FP Education level of male FP 16 15 80 76 14 70 12 60 10 50 8 40 6 4 30 4 3 3 20 15 8 11 2 1 10 1 1 0 0 Figure 424: Education level of farmer promoters differentiated by gender. Years of experience of farmer promotors 60 51 50 40 32 27 28 30 20 10 0 <2 2-4 4-6 >6 Years of experience Figure 435: Years of professional experience of farmer promoters. 29 ICT4BXW BASELINE REPORT OCT. 2018 Years of experience differentiated by gender 50 43 40 <2 2-4 30 28 24 4-6 20 17 >6 10 10 8 4 4 0 Female farmer promotors Male farmer promotors Years of experience Figure 446: Years of professional experience of famer promoters differentiated by gender. Expertise of farmer promoters Half of the FPs rated their expertise with BXW as “Good” (Figure 457) and 22 FPs indicated that they have a poor or very poor experience. The distribution of expertise among female and male farmer promoters is considerably similar. The experience of FPs with Banana was similar to their experience on BXW, and although this is differentiated by gender category, the distribution is comparable (Figure 48; Figure 4649). This suggests that expertise on banana and BXW are strongly related. FP expertise on BXW 3 17 19 Very poor Poor 30 Neutral 69 Good Very good Figure 457: Expertise levels of farmer promoters on BXW. 30 ICT4BXW BASELINE REPORT OCT. 2018 FP expertise on banana 1 9 16 Very poor Poor 41 Neutral 71 Good Very good Figure 48: Expertise levels of farmer promoters on banana production. FP expertise on banana differentiated by gender 70 58 60 Female farmer 50 promotors 40 33 Male farmer 30 promotors 20 12 13 8 8 10 4 1 1 0 Very poor Poor Neutral Good Very good Figure 469: Expertise levels of farmer promoters on banana production, differentiated by gender. ICT assets and interest in ICT-based agricultural services Almost all FPs possess a phone, which are mostly basic phone, except for 5 respondents who indicated that they have a smartphone (Figure 50). The most commonly used ICT-devices are the radio and basic phone (Figure 51). This is true for both male and female farmer promoters (Figure 52). Less than half of the farmers use their phone for job-related purposes, mostly by calling (Figure 483). If an ICT-based agricultural service would exist, the preferred media channel for farmer promoters would be voice calls or SMS (Figure 494). There is some interest as well in USSD codes or video communication. 31 ICT4BXW BASELINE REPORT OCT. 2018 FP own ICT-devices 5 5 own smart phone own basic phone No phone owned 132 Figure 50: Farmer promoters owning smart and/or basic phones. Use of ICT devices 160 134 140 130 120 100 80 60 35 40 26 20 1 5 1 2 0 0 Figure 51: Farmer promoters who have previously used ICT-devices. 32 ICT4BXW BASELINE REPORT OCT. 2018 Use of ICT-devices differentiated by gender 120 105 108 100 80 Female 60 Male 40 3225 25 26 20 3 1 0 1 1 4 0 1 0 2 0 0 0 Figure 472: Farmer promoters using ICT-devices by gender. FP using phone functions for their job 120 113 100 80 60 40 20 5 0 0 0 0 call sms whatsapp email facebook Figure 483: Farmer promoters using phone functions for their job. FPs' preferred media for exchanging information and knowledge about BXW 140 120 100 80 60 40 20 0 sms usdd voice voice call social agriculture Video Radio Others message media application Figure 494: FPs' preferred media for exchanging information and knowledge about BXW. 33 ICT4BXW BASELINE REPORT OCT. 2018 APPENDIX 1: Overview of village sampling and field notes during the baseline survey Eastern province Kayonza district Incidence Low Medium High Distance Short Intervention Village: Kamajigija Rurama Rusera Cell: Rukara Rusera Rusera Sector: Rukara Kabarondo Kabarondo Control Village: Butimba II Kinkoronko Butogbagire Cell: Kawangire Gikaya Rusera Sector: Rukara Nyamirama Kabarondo Medium Intervention Village: Kinunga Kinunga II Nyabikenke II Cell: Rwiminshiya Rwiminshiya Kabura Sector: Rukara Rukara Kabarondo Control Village: Akabare I Muzizi Agashar Cell: Rwiminshiya Rukara Kabura Sector: Rukara Rukara Kabarondo Large Intervention Village: Murambi Kabeza Rubira Cell: Kabura Kabura Kabarondo Sector: Kabarondo Kabarondo Kabarondo Control Village: Nyabikenke Agatare Gisoro Cell: Kabura Kabura Kabura Sector: Kabarondo Kabarondo Kabarondo 34 ICT4BXW BASELINE REPORT OCT. 2018 Map showing geographical spread of sampled villages in Kayonza district Field notes Kayonza is an important banana producing district with bananas produced throughout the district except for the areas close to Akagera National park. BXW is present in all banana producing areas and the sampling team found it most challenging to locate villages with low BXW incidence. The district has a flat terrain and road conditions are generally good, including off principal tarmac road. Kabarondo sector - This sector covers “near” to “far” distance gradient from the district headquarters. - The sector has all gradients of BXW incidence. - Initially the tarmac road from Kayonza towards the Tanzania border can be followed. In the sector you drive one murram roads. Primary murram roads are in good condition, secondary ones are poor but motorable without using 4x4. - Phone connectivity on both Airtel and MTN is medium to poor in most cells. - Internet connectivity in Kabarondo is poor. WhatsApp can send and receive messages occasionally. Use of online Google Maps is not possible. In Rusera (near district headquarters) connectivity is better. Google Maps is slow but can be used. WhatsApp works well in this cell. - Rusera cell is located directly next to the principal tarmac road to Kayonza. - Rusera has good phone connectivity, especially for MTN. - Rusera is grid connected with a direct line from Kayonza. 35 ICT4BXW BASELINE REPORT OCT. 2018 Nyamirama sector - Gikaya cell present in sample with one village in short distance to district headquarters and medium BXW incidence gradient. - Generally medium to high BXW incidence in the cell - Cell located close to the principal road to Kayonza. Roads in Gikaya are unpaved yet murram conditions are good. - Phone reception is good - Internet reception is good enough to send and receive via WhatsApp and use online Google Maps - The cell is grid connected Rukara sector - Included in sample for short distance to headquarters and low BXW incidence (2 villages) and medium distance to headquarters and low to medium BXW incidence (4 villages). - To reach the cell you drive on the principal tarmac road from Kayonza past Lake Muhazi. Roads in the sector are murram and in good condition. - The further away from the principal tarmac road the poorer the cell reception becomes. Receiving and sending WhatsApp messages becomes difficult and Google Maps no longer works. - In the center of the cell there is no powerline visible and villages are likely off-grid. - Once reaching Muzizi village the powerline returns, with improvement in phone and internet reception. From this village it is easy to reach the tarmac road again. Gatsibo Incidence Low Medium High Distance Short Intervention Village: Nyabikenke Rushashi Rukiri Cell: Nyabigiri Cyabusheshe Nyamirama Sector: Gitoki Gitoki Gitoki Control Village: Kwishaba Sata Kinteko Cell: Nyabigiri Karubungo Nyamirama Sector: Gitoki Gitoki Gitoki Medium Intervention Village: Rushenyi Rubare Mataba Cell: Rwarenga Nyagakombe Rwarenga Sector: Remera Remera Remera Control Village: Kabuye Byimana Agacyamo Cell: Rwarenga Murambi Muambi Sector: Remera Murambi Murambi Large Intervention Village: Rukira Rugarama Kizinga Cell: Nyagisozi Nyagahanga Nyagahanga Sector: Kageyo Gatsibo Gatsibo Control Village: Gakeri Nyagisozi Kageyo Cell: Nyagisozi Nyagisozi Nyagisozi Sector: Kageyo Kageyo Kageyo 36 ICT4BXW BASELINE REPORT OCT. 2018 Map showing geographical spread of sampled villages in Gatsibo district Field notes Generally, Gatsibo is flat with small elevation changes. Bananas are an important crop in the district and are farmed commercially and at large scale in some of the district’s sectors. Primary and secondary roads are unpaved but road conditions are good. The district shows all levels of BXW incidence. District agricultural officer appears well informed about BXW presence in the district. Gitoki sector - All six villages in the short distance to district headquarters category are located in this cell. - The cell has good murram roads - There is full phone reception in Gitoki - Mobile internet works, WhatsApp can be used for sending and receiving messages including pictures. - The sector has large scale, commercial banana farms. BXW is mostly present at medium to high incidence rates. Remera sector - Four of the six villages in the medium distance to headquarters category were sampled in this sector. - Remera has good murram roads. - Phone reception is generally good in Remera - Internet has medium reception. WhatsApp works, Google Maps can be used to limited extend. - All incidence levels of BXW are present in the sector which is reflected in the sample. Murambi sector - Murram roads in good conditions. - Sector typically has high incidence of BXW - Phone connectivity is very poor 37 ICT4BXW BASELINE REPORT OCT. 2018 - Internet connectivity is nearly absent (E level maximum) Gatsibo sector - Two villages were sampled in this sector, representing the far from district headquarters medium and high incidence categories. - Murram roads in good condition - Internet connectivity is poor - Generally, the sector has low BXW incidence except for the southern area. Northern province Rulindo Incidence Low Medium High Distance Short Intervention Village: Gaseke Rugando Karambi Cell: Barari Nyirabirori Mirezero Sector: Tumba Tumba Tumba Control Village: Marembo Ruvumba Kagusa Cell: Mizezero Taba Gahabwa Sector: Tumba Tumba Tumba Medium Intervention Village: Cyasenge Ndusu Giseko Cell: Karama Rebero Butunzi Sector: Buyoga Kinihera Kinihera Control Village: Kagozi Bwishya Mutoyi Cell: Ndarage Karegamazi Karegamazi Sector: Buyoga Kinihera Kinihera Large Intervention Village: Rusugati Buberano Akamanamana Cell: Sayo Gitatsu Mumama Sector: Kisaro Kisaro Kisaro Control Village: Songa Murambi Nyantabo Cell: Kamashenyi Muhuga Kigarama Sector: Kisaro Kisaro Kisaro 38 ICT4BXW BASELINE REPORT OCT. 2018 Map showing geographical spread of sampled villages in Rulindo district Field notes Rulindo is mountainous and therefore moving around in the district is slow. Road conditions are good however with new murram feeder roads constructed throughout the district. Cell reception in the district is generally good, internet medium to poor (E to H+ network). From Bushoki it is easy to link with the main road coming from Musanze towards Kigali. Tumba - Tumba sector is the district next to Bushoki and to get there it takes a few minutes only, mostly on a tarmac road. - The sector has all types of BXW incidence levels - Phone reception is good in Tumba - Internet gives H+ reception which is enough to use WhatsApp, online Google Maps, and simple browsing. Buyoga - This sector generally has a low BXW incidence since all affected plantations were uprooted just last year. - Two villages were sampled in this sector, representing the medium distance low BXW incidence category (control and intervention). In these two villages too BXW affected plantations were uprooted last year. - The sector has good phone reception, and internet connectivity ranges between E and H+. - Road conditions are good, with new murram feeder roads as primary roads. Kinihira - For the medium distance medium and high incidence category villages were sampled in Kinihira sector. - BXW is present throughout this sector. 39 ICT4BXW BASELINE REPORT OCT. 2018 - In the recent past affected plots have been uprooted, yet it appears that BXW returned. - Banana production in this sector is mostly on smaller plots. - The sector has acceptable cell reception, basic phone use is possible. Kisaro - The team started sampling in this northern sector as it can be reached from Gatsibo after crossing through Gicumbi. This sector is furthest from Bushoki town, where the district headquarters are located. - Only villages in the far distance category could be sampled, those located in the medium distance category are all in the highlands where there is no banana production. - There is a brand-new tarmac road that leads to the sector office. Inside the sector roads are of murram quality. - Kisaro has good phone reception (making calls and using WhatsApp is possible, internet gives H signal). Burera Incidence Low Medium High Distance Short Intervention Village: Gitovu Ndago Murambo Cell: Ndago Ndago Kaganda Sector: Rusarabuye Rusarabuye Kinyababa Control Village: Burehe Gashiru Gatare Cell: Ndago Musasa Bugamba Sector: Rusarabuye Gitovu Kinyababa Medium Intervention Village: Nyarubuye Cyogo Buhembe Cell: Ntaruka Bugamba Mariba Sector: Kinoni Kinyababa Gitovu Control Village: Kabaguma Kigugu Shyamba Cell: Nkumba Nkenke Musasa Sector: Kinoni Kinoni Gitovu Large Intervention Village: Rwambeho Nyagafunzo Karambo Cell: Rurembo Gafuka Nkumba Sector: Rugarama Kinoni Kinoni Control Village: Basumba Nyakiriba Birwa II Cell: Gafuka Rurembo Nkenke Sector: Kinoni Rugarama Kinoni 40 ICT4BXW BASELINE REPORT OCT. 2018 Map showing geographical spread of sampled villages in Burera district Field notes - There are no tarmac roads in Burera. The district’s principal road is murram and in bumpy condition. - The district headquarters is located central in the district, an hour drive from the tarmac road between Kigali and Musanze/Cyanika (both from north and south end). - All banana producing sectors in Burera district are located around Lake Burera. The sectors can be reached by following the principal road. - Banana production ranges from small scale to large, consolidated plantations (latter especially in the western sectors). - Sector offices are located not too far from the principal road. - Reaching villages by car is sometimes difficult, via dirt tracks. Some villages can only be reached by foot. - Mobile network coverage is medium to poor. Not all areas have cell reception for either calling or use of internet. Sending and receiving of WhatsApp messages is intermittent). Western province Rubavu Incidence Low Medium High Distance Short Intervention Village: Schwemu Rukukumbo Nkama Cell: Gisa Kabilizi Kabilizi Sector: Rugerero Rugerero Rugerero Control Village: Bambiro Ndobogo Byima Cell: Gikombe Rwaza Rwaza Sector: Rubavu Rugerero Rugerero 41 ICT4BXW BASELINE REPORT OCT. 2018 Medium Intervention Village: Kanyamagare Hanika Kanembwe Cell: Ryabiziga rwangara Busigari Sector: Cyonzarwe Cyonzarwe Cyonzarwe Control Village: Mukingo Butango Kanyentambi Cell: Makurizo Cyonzarwe Kinyanzovu Sector: Cyonzarwe Cyonzarwe Cyonzarwe Large Intervention Village: Cell: Sector: Control Village: Cell: Sector: Map showing geographical spread of sampled villages in Rubavu district Field notes In Rubavu district, the sampling team experienced several challenges when sampling villages. Firstly, it was difficult to sample villages in the short distance to headquarters low BXW incidence category. The team observed a lot of high and medium incidence plots near Rubavu town where the district headquarters are located. Those villages assigned by informants as low incidence were assessed as at 42 ICT4BXW BASELINE REPORT OCT. 2018 least medium incidence by the team. Secondly, all far distance to district headquarters sectors were unsuitable for sampling. These sectors are either located in Volcanoes National Park or highland areas that don’t produce bananas. The team tried to alternatively sample from the furthest medium distance sectors as a replacement for the far distance villages. They did not succeed because interactions with the sector agronomists taught them that those are again sectors at high altitude where any significant banana production is absent. The district benefits from rich volcanic soils. Farmers don’t manage the size of their banana mats. The team observed mats with up to 15 stems, where 4-5 stems would be normal. Spacing between mats is also limited. As a result, there is high chance of disease transmission. Rubavu and Rugerero sectors - Generally, in Rubavu and Rugerero sectors the BXW incidence is high. The team observed entire mats with clear BXW infection symptoms. - In Rubavu and Rugerero sectors banana production is of peri-urban nature, with medium sized plots directly adjacent to the homesteads. - When asking villagers why they did not remove visibly BXW affected banana mats next to the homestead, they stated to be waiting for the Umuganda groups that remove BXW infected mats from villages during the dry season. - BXW infection takes place both due to insect vectors (recognizable from shredding male buds and fruit decolouration and deformation) and tool contamination (brown, withering leaves without fruit deformations). - Male buds not removed in plantation, neither from healthy nor from diseased stems. Thus, high chance of insect vector transmission. - Villages are easily accessible on murram/dirt roads from the principal Rubavu road. - Cellphone connectivity is good in this peri-urban area. However, phones sometimes switch to a DR Congo network. Banana fruits affected by BXW due to insect vector transmission of the disease. The leaves do not yet show disease symptoms. Cyanzarwe - Sector agronomist very well informed about presence and incidence rates of BXW in her sector. - Wide variety of incidence rates in this sector - In infected plantations management is generally poor and there is visible high chance of tool transmission (e.g. due to skinning of banana stems in infected plots using tools). - Besides tool transmission there is visible insect vector transmission (shredding male buds) and chance of animal transmission (roaming goats, sheep, and cattle were observed). - Roads in the sector are all murram and in good condition. Most of the sampled villages are located along these roads. Sometimes a short walk to the village may be required. - Cell reception in the sector is variable, at times absent or poor. Additionally, the phone switches to a DR Congo network and roaming mode which makes the use of internet or calls expensive or impossible (e.g. the team did not manage to reach the sector agronomist by phone since it was not operating on the Rwandan network). 43 ICT4BXW BASELINE REPORT OCT. 2018 Observed presence of both fusarium wilt (left) and BXW (right) in Cyanzarwe sector Karongi Incidence Low Medium High Distance Short Intervention Village: Gitega Gisayo Josi Cell: Kayenzi Gasura Gitarama Sector: Bwishuyra Bwishuyra Bwishuyra Control Village: Nyamarebe Majuri Gomba Cell: Burunga Burunga Gitarama Sector: Bwishuyra Bwishuyra Bwishuyra Medium Intervention Village: Muramba Gitwa Bwakira Cell: Nkoto Kigarama Muhororo Sector: Murambi Gishyita Murambi Control Village: Kagano Kananira Mpatsi Cell: Cyanya Muhororo Cyanya Sector: Gishyita Murambi Gishyita Large Intervention Village: Ngugu Gisoro Kubutare Cell: Munanira Munanira Kigarama Sector: Gishyita Gishyita Gishyita Control Village: Uwingabo Mataba Nyakabuye Cell: Ngoma Ngoma Munanira Sector: Gishyita Gishyita Gishyita 44 ICT4BXW BASELINE REPORT OCT. 2018 Map showing geographical spread of sampled villages in Karongi district Field notes Similar to Rubavu, the sampling team struggled to identify villages in the large distance to headquarters category. Most of the sectors belonging to this category are highland areas that do not grow banana (southern part of the district). Those in Gishyita sector (eastern Karongi) are mostly coffee growing. Sector agronomist of Murambi stated that the large distance villages are BXW free. Murundi sector is completely BXW free according to its sector agronomist. The latter was perceived as unlikely by RAB technician on the sampling team and would rather be a sign that the agronomist is not well informed about BXW symptoms and/or reports about BXW presence in his sector. Regardless, the team decided not to sample from the sector. Bwishuyra sector - Sector office located in Kibuye town and easily accessible. - Sector easily reached from district headquarters via a new principal tarmac road. - Phone reception for calls and internet medium to good. Gishyita sector - Sector located in medium distance from district headquarters category. - Sector easily reached from Kibuye town on new principal tarmac road. - Sector office located directly next to the tarmac road. - Banana producing villages mostly located along the lake shores. - Team had to sample villages close to one another because especially far distance category villages do not all grow banana (Those on the peninsula grow coffee). - Cell phone reception medium to poor. In some areas there is no internet signal. Murambi sector - According to the map this sector is comparable to Gishyita, with both medium and far distance villages. In reality the comparability stops at distance. The principal road leading to Gishyita is 45 ICT4BXW BASELINE REPORT OCT. 2018 entirely unpaved, ranging from murram to dirt roads, and is generally in poor condition. This makes this sector difficult to access - The sector office is located at the far end of the sector. - Sector agronomist appeared not to know BXW symptoms well and was not able to clearly point out the presence of BXW in her sector. - According to the agronomist there is no BXW in any of the villages belonging to the far from district headquarters category. - Cell phone reception in the sector is good, both calling and use of WhatsApp is possible (H internet signal). - The team sampled three villages that fall in the medium distance category and are located near to the principal road. Southern province Muhanga Incidence Low Medium High Distance Short Intervention Village: Songa Nyakaguhu Rubuye Cell: Mbare Kinini Mbare Sector: Shyogwe Shyogwe Shyogwe Control Village: Mapfundo Gasharu Matsinisi Cell: Mubuga Mubuga Mubuga Sector: Shyogwe Shyogwe Shyogwe Medium Intervention Village: Karambo Gasiza Gitwa Cell: Butare Kavumu Sholi Sector: Kabacuzi Kabacuzi Kabacuzi Control Village: Mataba Ngando Ntonde Cell: Kiboga Nsanga Kanyana Sector: Rugendabari Rugendabari Rugendabari Large Intervention Village: Musarara Butare Rwesero Cell: Gitega Rukyiniro Gisharu Sector: Kibangu Kibangu Kibangu Control Village: Musekera Matoshya Musambagiro Cell: Ryakanimba Mubuga Gitega Sector: Kibangu Kibangu Kibangu 46 ICT4BXW BASELINE REPORT OCT. 2018 Map showing geographical spread of sampled villages in Muhanga district Field notes (no field notes) Gisagara Incidence Low Medium High Distance Short Intervention Village: Rubazi Akagarama Rususa Cell: Ruturo Mugombwa Nyabisagara Sector: Kibirizi Mugombwa Mukindo Control Village: Rwuya Nyarukeri Munyegera Cell: Duwan Baziro Runinya Sector: Kibirizi Mugombwa Mukindo Medium Intervention Village: Shenyeri Gitwa Itaba Cell: Kibirizi Bwiza Runinya Sector: Kibirizi Kansi Mukindo Control Village: Nyabununi Ruhuha Nyakazana Cell: Ruturo Akaboti Mukiza 47 ICT4BXW BASELINE REPORT OCT. 2018 Sector: Kibirizi Kansi Mukindo Large Intervention Village: Gasagara Nyabitare Joma Cell: Kibirizi Rwanza Gitega Sector: Kibirizi Save Mukindo Control Village: Akabuhuzu Gahora Akagarama Cell: Ruturo Rwanza Mukiza Sector: Kibirizi Save Mukindo Map showing geographical spread of sampled villages in Gisagara district 48 ICT4BXW BASELINE REPORT OCT. 2018 Annex 1: overview of mobile internet signals and symbols Overview of different mobile internet signals. Adapted from http://rechargehut.blogspot.com 49