Multi-country evidence of gender gaps in sustainable rice cultivation performance indicators in sub-Saharan Africa December 2025 This document reports on a study that examines gender disparities in the key sustainable rice platform performance indicators for agronomic gain (yield, net profit, labor productivity, and nitrogen and phosphorus partial factor productivity) across ten Sub-Saharan African countries. The study also identifies the most influential factors to enhance farmers’ yields and profitability and reduce gender disparities. CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps Contents | Page 1 of 34 Contents Contents 1 Abstract 2 1. Introduction 3 2. Materials and methods 4 3. Results 6 4. Conclusions 25 References 26 Appendices 28 Authors: Moukaïla Bouraima Bagri, Louis Kouadio, Ali Ibrahim, Perpetue Kouamé, Kalimuthu Senthilkumar CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps Abstract | Page 2 of 34 Abstract A comprehensive understanding of gender gaps in sustainable rice production is essential for designing interventions that promote both equity and environmental stewardship in rice- based agricultural systems. This study investigates gender disparities in sustainable rice platform agronomic performance indicators (SRP PIs)– yield, net profit, labor productivity, and nitrogen and phosphorus partial factor productivities (PFPN and PFPP, respectively) across ten sub-Saharan African countries (Burkina Faso, Burundi, Democratic Republic of Congo, Kenya, Madagascar, Nigeria, Rwanda, Sierra Leone, Tanzania, and Uganda). Data were collected from 3,081 farmers (2,042 male and 1,039 female) in irrigated lowland (IL), rainfed lowland (RL), and rainfed upland (RU) rice systems in the ten countries. The analysis revealed that gender gaps in PIs varied widely by country and production system. While the overall average yield gap (compare to the top 10 percentile) difference between male (2,587 t ha-1) and female (3,028 t ha-1) farmers was not significant, women experienced a higher average yield gap (48%) compared to men (41%), with countries such as Burkina Faso, Burundi, Kenya, Nigeria, Sierra Leone, and Uganda showing notably higher yield gaps for women. Net profit disparities were observed, particularly in IL in Burkina Faso (32%) and rainfed systems in Nigeria (28% for both RL and RU). Women’s lower use of equipment compared to men contributed to substantial labor productivity gaps, especially in Burkina Faso (28%) and Nigeria (26–46%). Additionally, over half of both male and female farmers recorded PFPN below optimal levels (40 kg grain kg-1 N), while 54% farmers exceeded desirable PFPP (250 kg grain kg-1 P), with women comprising 24% of this group. Random forest-based analysis identified labor conditions improvement, access to certified seeds, nutrients, and irrigation, and capacity building as critical steps toward enhancing female farmers’ yields and profitability and reducing gender disparities. A sister report focusing on SRP Standard using the same dataset is presented in a separate document. CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps 1. Introduction | Page 3 of 34 1. Introduction Women play a significant role in African agriculture, with their participation in farming ranging from 4% to 94% (ILOSTAT, 2019). However, persistent gender-based inequalities, including limited access to resources and agricultural inputs, and limited decision-making power, hinder the full potential of women farmers. Equalizing productivity between male and female farmers could increase global gross domestic product by $1T and lift 45 million people out of hunger (Mane et al., 2025). Once considered a luxury food, rice consumption in sub-Saharan Africa (SSA) countries averaged 20.7 kg per person per year in 2021, with projections estimating a 3% to 4% annual growth in consumption by 2030 (OECD-FAO, 2021). Despite this rising demand, local rice production currently meets approximately 60% of total consumption, with demand expected to double by 2050 to reach about 150 million tons (De Vos et al., 2023; Yuan et al., 2024). Bridging this gap becomes a critical challenge. Intensifying rice production is therefore essential to reduce import dependency, although such intensification often raises concerns about environmental sustainability due to increased input use (Gharsallah et al., 2021). Sub-Saharan Africa faces mounting challenges, including food insecurity, gender inequality, resource degradation, and climate change, all exacerbating low crop yields (Amede et al., 2023). Sustainable agriculture, which integrates economic, environmental, and social goals, is recognized as a solution to these issues, offering the potential to increase productivity, improve farmers’ income, and restore soil health (Abdallah et al., 2021). Rice, as a staple food crop, has emerged as a strategic entry point for achieving sustainable development goals in SSA, promoting not only food security but also economic development and social stability (Yadav & Kumar, 2018). Addressing gender disparities in rice farming systems is not only a social imperative but also a pathway to enhancing agricultural sustainability and resilience. While the body of literature on sustainable rice production continues to grow, gender- differentiated analyses remain limited, particularly in SSA. Integrating gender perspectives into sustainability assessments is crucial for designing interventions that address these disparities, increase productivity, and ensure equitable benefits for all stakeholders. This study assesses gender-based differences in key sustainable rice platform performance indicators (SRP PIs) for agronomic gain, i.e., yield, net profit, labor productivity, and nitrogen and phosphorus partial factor productivities (PFPN and PFPP, respectively), in SSA to identify actionable insights for policy and program design that address gender gaps. CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps 2. Materials and methods | Page 4 of 34 2. Materials and methods 2.1. Study sites and data collection The study was conducted by AfricaRice across ten SSA countries during the 2020/2021 rice harvest period, in collaboration with National Agricultural Research and Extension Services (NARES) partners. The countries involved included Burkina Faso, Nigeria, Sierra Leone (West Africa), Democratic Republic of Congo (Central Africa), Kenya, Tanzania, Uganda, Rwanda, Burundi, and Madagascar (Eastern and Southern Africa). Data were collected from 3,081 rice farming households, including 1,039 female-headed and 2,042 male-headed households, using an SRP survey tool, and covered all 41 requirements of the Sustainable Rice Platform (SRP) Standard and 12 performance indicators (PIs) (SRP, 2023). Farmers were randomly selected with the assistance of NARES partners in irrigated lowland (IL), rainfed lowland (RL), and rainfed upland (RU) systems. The sites surveyed were in three production systems in four countries (Madagascar, Nigeria, Sierra Leone, and Tanzania), two production systems in three countries (Burkina Faso, Burundi, and Democratic Republic of Congo), and one production system in the remaining countries (Kenya, Uganda, and Rwanda). 2.2. SRP performance indicators The SRP provides a comprehensive framework for assessing the sustainability of rice cultivation through 12 economic, environmental, and social PIs. This framework has been widely adopted for evaluating rice production systems, informing policy development, and designing interventions that promote sustainable intensification in Asia, Latin America, and Africa (Arouna et al., 2021; Devkota et al., 2022; Ibrahim et al., 2022; Tseng et al., 2021). In this study, we focused on the five key SRP PIs for agronomic gain: yield, net profit, labor productivity, partial factor productivity of nitrogen (PFPN), and partial factor productivity of phosphorus (PFPP) (Sajjad et al., 2025). Net profit was calculated as the gross margin minus all expenditures, including labor and variable costs. Labor productivity was measured as total rice yield per man-day of labor. PFPN and PFPP were calculated by dividing the total grain yield by the respective nitrogen and phosphorus inputs. Yield, profit, and labor productivity gaps were calculated as the percentage difference between the performance in the top 10th percentile of the farmers' SRP PIs within each country and production system, and the overall mean across countries, production systems, and gender (Ibrahim et al., 2022). The gender gap was calculated as the difference in respective performance gaps recorded for female and male farmers. It was expressed in percentage. 2.3. Statistical analysis Prior to the analysis, outliers were identified and removed from the dataset using the interquartile method within each country, as well as missing data for each interesting variable. After data cleaning, the number of observations used in subsequent analyses were 3,016 for yield, 3,007 for net profit, 2,857 for labor productivity, 1,881 for PFPN and 1,558 for PFPP. Descriptive statistics were carried out to examine data distributions across production systems and genders. For the PFPN and PFPP analysis, farmers were classified into three groups: below optimal, optimal, and above optimal range. Optimal CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps 2. Materials and methods | Page 5 of 34 PFPN and PFPP ranges were 40 - 80 kg grain kg-1 elemental N and 100 - 250 kg grain kg- 1 elemental P, respectively (Ibrahim et al., 2022). Three-way and one-way analysis of variance (ANOVA) was used to assess the differences between countries, production systems, and gender for yield, net profit, and labor productivity, as well as for their respective gaps. Least significant difference (LSD) tests were applied to compare means wherever the ANOVA was statistically significant (p < 0.05). Prior to AANOVAs, data were checked for normality and homogeneity. To identify the key drivers of yield, net profit, and labor productivity for all sites, random forest modelling was conducted. Twelve predictor variables that can affect the five PIs were used: crop calendar, record keeping, training, land leveling, use of certified seeds, production system, plot size, number of equipment, number of irrigations, N input, P input, and labor input. The most important factors determining yield and net profit identified were further subjected to multiple regression analysis to quantify their degree of influence. Multinomial logistic regression analysis was employed to identify the determinants of optimal PFPN and PFPP. All analyses were performed using R version 4.4.1 (R Core Team, 2024) within RStudio version 2025.9.1.401 (Posit Team, 2025). CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps 3. Results | Page 6 of 34 3. Results 3.1. Descriptive statistics Female participation in rice cultivation varied across the surveyed countries (Figure 1), with countries such as Sierra Leone and Madagascar having 10 to 12 percentage point differences between male and female farmers. For instance, in Sierra Leone, the proportion of female farmers surveyed was 45% (55% for males); in Madagascar, female farmers accounted for 44% of the surveyed farmers. In countries such as Burkina Faso, Nigeria, or Rwanda, notable disparities existed between surveyed female and male farmers in terms of engagement in rice farming, with male farmers dominating (Figure 1). Figure 1. Percentages of surveyed women engaged in rice cultivation in the study countries. DRC: Democratic Republic of Congo. Women cultivated 35% to 39% less land compared to men across the different production systems (Table 1). They applied 16% to 54% less nitrogen and utilized 50% less CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps 3. Results | Page 7 of 34 equipment. Such low use of agricultural equipment increased their labor contribution by 11% to 27%, depending on the production system. Table 1. Description of surveyed farmers’ rice cultivation practices. Production system Country Gender N input P input Labor input Number of equipment Number of irrigations Mean (kg ha-1) CV (%) Mean (kg ha-1) CV (%) Mean (man-days ha-1) CV (%) Mean CV (%) Mean CV (%) Irrigated lowland Burkina Faso Female 86 54 9 98 240 39 2 57 20 47 Male 65 99 12 101 119 65 3 33 13 47 Burundi Female 43 82 9 91 757 63 1 69 23 59 Male 35 90 9 96 630 53 1 57 22 65 Democratic Republic of Congo Female 26 79 2 177 282 70 2 39 10 93 Male 31 74 3 146 240 68 2 41 11 88 Kenya Female 63 63 7 136 402 74 1 58 9 66 Male 74 62 7 146 306 81 1 59 8 65 Madagascar Female 0 - 1 266 225 53 1 126 5 54 Male 0 - 2 210 196 65 1 95 5 59 Nigeria Female 85 79 10 86 89 74 3 29 19 73 Male 88 89 6 108 115 56 3 29 27 53 Rwanda Female 80 - 12 35 688 48 1 29 16 72 Male 80 0 12 41 604 59 1 32 14 69 Sierra Leone Female 0 - 3 187 24 91 1 122 - - Male 0 - 0 - 37 122 1 100 - - Tanzania Female 84 63 16 74 35 69 3 28 15 69 Male 79 65 14 82 33 67 3 28 16 67 Rainfed lowland Burkina Faso Female 43 69 7 83 128 93 3 48 0 - Male 33 123 7 106 75 97 3 41 0.235 825 Burundi Female 16 97 1 207 571 33 1 105 0 - Male 20 116 4 152 1094 45 1 93 0 - Democratic Republic of Congo Female 14 87 1 87 429 43 2 35 0 - Male 0 - 0 - 740 - 2 0 0 - CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps 3. Results | Page 8 of 34 Madagascar Female 0 - 0 - 115 79 1 118 0 - Male 0 - 0.14 849 97 55 1 114 0 - Nigeria Female 1 70 1 68 168 16 2 47 0 - Male 84 89 9 85 113 50 2 42 2 337 Sierra Leone Female 0 - 1 278 82 109 1 128 0 - Male 0 - 1 242 63 140 1 94 0 - Tanzania Female 34 141 4 199 39 52 2 59 7 161 Male 33 153 6 157 45 55 2 60 6 134 Rainfed upland Madagascar Female - - 11 - 240 - 3 - 0 - Male - - - - - - - - - - Nigeria Female 70 61 10 70 85 37 2 46 5 135 Male 63 73 7 85 111 61 2 30 8 124 Sierra Leone Female 0 - 0.01 632 145 72 1 112 0 0 Male 0 - 0.01 742 133 71 1 130 0 0 Tanzania Female 60 53 6 117 42 44 3 36 12 92 Male 42 82 7 149 41 61 2 40 13 103 Uganda Female 0 - 0.01 1114 266 51 1 54 0 - Male 0 - 0.34 834 221 45 1 63 0 - Across production systems and genders Irrigated lowland Female 53b 8a 303a 2b 13b Male 67a 8a 217b 2a 18a Rainfed lowland Female 16b 2b 104a 1a 2a Male 26a 4a 109a 2a 1a Rainfed upland Female 10b 1b 216a 1a 1b Male 25a 3a 153b 2a 3a Across all production systems Irrigated lowland 63a 8a 243a 2a 17a Rainfed lowland 23b 3b 108c 2b 2b Rainfed upland 18b 2c 181b 1c 2b Varying yield levels were recorded between female and male farmers across the three production systems, with some farmers exhibiting yields that deviated markedly from the mean in RU and RL systems (Figure 2). The greatest asymmetry in yield distribution was observed among female farmers (Figure 2). Net profits were up to USD 1,000 ha-1 for most female farmers, irrespective of the production system (Figure 3). A similar trend was observed for male farmers in RU and RL (Figure 3). Labor productivity exhibited high CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps 3. Results | Page 9 of 34 asymmetry across all production systems and both genders, with notable variation among farmers (Figure 4). Figure 2. Frequency distribution of rice yield for female and male farmers in irrigated lowland (IL; top), rainfed lowland (RL; middle), and rainfed upland (RU; bottom) systems. CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps 3. Results | Page 10 of 34 Figure 3. Frequency distribution of net profit for female and male farmers in irrigated lowland (IL; top), rainfed lowland (RL; middle), and rainfed upland (RU; bottom) systems. CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps 3. Results | Page 11 of 34 Figure 4. Frequency distribution of labor productivity for female and male farmers in irrigated lowland (IL; top), rainfed lowland (RL; middle), and rainfed upland (RU; bottom) systems. Average yields and labor productivity were higher under IL compared to RL and RU systems (Table 1). Tanzania recorded the highest yield at 6,051 kg ha⁻¹, while Sierra Leone had the lowest yield at 868 kg ha⁻¹. Similarly, Tanzania exhibited the highest values for labor productivity, while Sierra Leone, Burundi, Rwanda, and Uganda recorded the lowest values. The highest net profit was recorded in RL systems, outperforming IL and RU production systems. Sierra Leone recorded the highest net profit, and the lowest values were seen in Madagascar and Uganda. PFPN was higher in both IL and RL systems compared to RU, while PFPP remained consistent across the three production systems. At the country level, PFPN and PFPP were higher in the Democratic Republic of Congo and Burundi, while they were lower in Sierra Leone and Madagascar. CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps 3. Results | Page 12 of 34 Across all genders, male farmers recorded higher values for all five SRP PIs under IL (Table 1). Under RL, all PIs were similar between genders, except for yield, which was higher for males. Similarly, in RU systems, while net profit, PFPN, and PFPP were comparable between genders (Table 2), yield, net profit, and labor productivity were higher for males than females. Table 2. Average yield, net profit, labor productivity, nitrogen partial factor productivity (PFPN), phosphorus partial factor productivity (PFPP) across production systems and countries. Production system Country Gender Yield ± se (kg ha-1) Profit ± se (USD ha-1) Labor productivity± se (kg man-day ha-1) PFPN ± se (kg grain kg-1 N) PFPP ± se (kg grain kg-1 P) Irrigated lowland Burkina Faso Female 3640 ± 1377 435 ± 261 44 ± 18 48 ± 35 276 ± 247 Male 5100 ± 1122 690 ± 287 59 ± 23 73 ± 59 218 ± 168 Burundi Female 3817 ± 1426 1095 ± 691 6 ± 3 92 ± 52 373 ± 229 Male 4300 ± 1721 1236 ± 736 7 ± 4 114 ± 60 398 ± 252 Democratic Republic of Congo Female 3585 ± 1171 528 ± 401 42 ± 33 127 ± 61 642 ± 258 Male 3771 ± 1281 580 ± 466 36 ± 28 112 ± 60 539 ± 248 Kenya Female 4960 ± 2119 692 ± 535 14 ± 13 91 ± 54 321 ± 257 Male 5796 ± 1863 823 ± 518 22 ± 17 92 ± 51 338 ± 196 Madagascar Female 1567 ± 659 142 ± 198 24 ± 14 18 ± 44 13 ± 56 Male 1620 ± 731 179 ± 190 19 ± 12 41 ± 73 28 ± 91 Nigeria Female 3748 ± 1768 963 ± 602 62 ± 36 49 ± 40 291 ± 208 Male 5252 ± 1794 1172 ± 674 55 ± 36 48 ± 38 390 ± 223 Rwanda Female 4303 ± 1030 690 ± 417 7 ± 3 66 ± 32 339 ± 124 Male 4363 ± 949 779 ± 367 7 ± 4 69 ± 31 385 ± 202 Sierra Leone Female 608 ± 194 2886 ± 2676 10 ± 7 15 ± 14 136 ± 134 Male 190 ± 170 1081 ± 91 6 ± 8 89 ± 94 - Tanzania Female 7168 ± 1832 789 ± 377 117 ± 66 78 ± 42 310 ± 138 Male 6755 ± 2029 742 ± 345 117 ± 67 76 ± 41 355 ± 236 Rainfed lowland Burkina Faso Female 2057 ± 934 159 ± 179 29 ± 20 52 ± 44 279 ± 198 Male 2141 ± 1277 186 ± 236 15 ± 13 81 ± 76 287 ± 286 Burundi Female 1982 ± 891 439 ± 311 3 ± 1 84 ± 10 - Male 2645 ± 876 488 ± 291 3 ± 2 102 ± 79 293 ± 120 Democratic Republic of Congo Female 1800 ± 346 178 ± 109 9 ± 7 98 ± 4 - Male 2000 ± 0 607 ± 0 - - - Madagascar Female 1726 ± 444 231 ± 126 21 ± 13 - - Male 1767 ± 460 252 ± 119 22 ± 10 - 532 ± 284 Nigeria Female 2178 ± 295 734 ± 350 14 ± 5 - - Male 4037 ± 2051 1311 ± 733 54 ± 48 49 ± 35 367 ± 176 Sierra Leone Female 649 ± 561 2791 ± 1724 6 ± 6 41 ± 31 201 ± 117 Male 740 ± 552 2657 ± 1782 5 ± 6 41 ± 38 201 ± 178 Tanzania Female 4704 ± 1777 563 ± 395 90 ± 77 111 ± 76 391 ± 216 Male 4990 ± 1742 661 ± 403 86 ± 72 121 ± 78 417 ± 238 CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps 3. Results | Page 13 of 34 Rainfed upland Nigeria Female 1967 ± 1372 462 ± 276 12 ± 12 25 ± 16 243 ± 252 Male 3176 ± 1360 951 ± 686 33 ± 28 44 ± 20 356 ± 187 Sierra Leone Female 1196 ± 799 2366 ± 2061 4 ± 4 - - Male 1248 ± 693 1739 ± 1949 3 ± 4 - - Tanzania Female 4158 ± 2460 544 ± 535 75 ± 59 70 ± 41 459 ± 323 Male 3025 ± 1532 426 ± 516 45 ± 26 61 ± 8 250 ± 91 Uganda Female 2037 ± 970 218 ± 215 9 ± 6 78 ± 34 - Male 2275 ± 1135 284 ± 256 11 ± 7 120 ± 50 - Averages across all production systems Irrigated lowland Female 3940 ± 2294 b 604 ± 616 b 40 ± 47 b 63 ± 56 b 231 ± 231 b Male 4799 ± 2061 a 859 ± 617 a 47 ± 46 a 69 ± 55 a 321 ± 247 a Rainfed lowland Female 2137 ± 1807 b 1244 ± 1591 a 30 ± 48 a 67 ± 58 a 308 ± 221 a Male 2541 ± 2002 a 1114 ± 1398 a 32 ± 46 a 73 ± 68 a 320 ± 237 a Rainfed upland Female 1976 ± 1283 b 715 ± 1302 a 12 ± 23 b 49 ± 43 a 274 ± 287 a Male 2392 ± 1370 a 885 ± 1197 a 18 ± 23 a 46 ± 26 a 243 ± 187 a Means with the same letters in the columns are not significantly different for the corresponding factor at 5% threshold. 3.2. Gender-differentiated gaps in yield, net profit, and labor productivity Overall yield, net profit, and labor productivity gaps for women were 68%, 61%, and 64%, respectively, and were higher compared to men for yield (41%) and net profit (56%) gaps. However, the labor productivity gap between women and men was quite similar (66% for men) (Table 3). Despite the overall mean yield gap being statistically similar between genders (p > 0.05) across the ten study countries (Tables 3 and S1), women in IL in Burkina Faso, IL and RU in Sierra Leone, RL in Burundi, and IL and RL in Nigeria experienced yield gaps of about 50%, 19%, and 16%, 50%, 32%, and 36%, respectively (Table 3; Figure 5), with those values being larger than corresponding gaps in men. Labor productivity gender gaps were most often higher for women in the majority of countries (Table 3). Net profit differences were notably higher for women in countries such as Burkina Faso (32%), Kenya (18%), Madagascar (11%), and Rwanda (11%) in IL systems, Nigeria (28%) in both rainfed systems, and Tanzania (12%) in RL systems (Table 3). Table 3. Rice production performance gap (%) relative to the highest-performing farms by country, cropping system, and gender. Production system Country Gender Yield gap (%) Profit gap (%) Labor productivity gap (%) Irrigated lowland Burkina Faso Female 44 69 58 Male 22 47 42 Burundi Female 45 56 66 Male 38 51 61 CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps 3. Results | Page 14 of 34 Democratic Republic of Congo Female 40 65 59 Male 37 60 64 Kenya Female 45 61 78 Male 36 56 64 Madagascar Female 57 79 61 Male 52 70 67 Nigeria Female 55 65 49 Male 37 56 55 Rwanda Female 30 53 77 Male 29 47 74 Sierra Leone Female 80 59 59 Male 65 85 82 Tanzania Female 29 48 40 Male 34 48 41 Rainfed lowland Burkina Faso Female 52 73 71 Male 50 69 83 Burundi Female 50 57 82 Male 33 52 86 Democratic Republic of Congo Female 10 71 91 Male - - - Madagascar Female 40 51 65 Male 38 47 62 Nigeria Female 71 71 88 Male 46 48 58 Sierra Leone Female 74 55 76 Male 81 57 74 Tanzania Female 42 65 51 Male 39 57 51 Rainfed upland Nigeria Female 66 83 86 Male 45 63 64 Sierra Leone Female 68 62 81 Male 57 72 81 Tanzania Female 50 63 47 Male 64 79 58 Uganda Female 52 70 68 Male 46 62 62 Average across all production systems Irrigated lowland Female 47 ± 16 a 62 ± 9 a 61 ± 12 a Male 39 ± 13a 58 ± 12a 61 ± 13 a All 43 ± 14 a 60 ± 11 a 61 ± 12 a Rainfed lowland Female 47 ± 22 a 48 ± 9 a 75 ± 14 a Male 44 ± 24 a 44 ± 22 a 73 ± 16 a All 46 ± 22 a 55 ± 18 a 74 ± 15 a CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps 3. Results | Page 15 of 34 Rainfed upland Female 49 ± 23 a 56 ± 32 a 66 ± 35 a Male 43 ± 25 a 69 ± 9 a 57 ± 10 a All 46 ± 23 a 61 ± 24 a 61 ± 26 a Means with the same letters in the columns are not significantly different for the corresponding factor at 5% threshold. Figure 5. Labor productivity, profit and yield gaps (%) across production systems for all sites. For each SRP PI, means with similar letters are not significantly different (p > 0.05). 3.3. Nutrient use efficiencies The analysis revealed consistent gender-related contrasts in PFPN across production systems and countries. In IL, women most often showed slightly more favorable PFPN distributions than men, with a higher share (56%) falling below the desirable PFPN range (Table 4). PFPP patterns in this system highlighted some gender differences in countries such as Nigeria and Rwanda, where the surveyed male farmers tended to be over- represented in the above desirable PFPP range (72% and 85%, respectively), suggesting stronger P uptake efficiency or more pronounced P limitations in fields managed by women. However, exceptions exist – most notably in Madagascar– where both genders showed low PFPP (Table 4). In RL, PFPN distributions showed wider variability, with some countries (e.g., Burundi and DRC) having most farmers in the above-range category, while others showed balanced distributions across categories (Table 4). PFPP in this system is often skewed toward the above desirable range, especially among women in Burundi and the Democratic Republic of Congo, reflecting likely P limitations. CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps 3. Results | Page 16 of 34 RU systems showed the most heterogeneous nutrient use efficiency patterns, with extreme values in several locations. Many female farmers in Madagascar and Uganda, for example, fell exclusively in the lowest PFPN category. This holds true for PFPP in Madagascar (Table 4). Men, in contrast, tended to exhibit more evenly distributed efficiency profiles across categories in most upland sites. Combined across countries, IL systems tended to have more balanced PFPN, whereas rainfed systems– particularly RU – exhibited either very high or very low efficiency classes, pointing to more variable management conditions and soil fertility constraints. Table 4. Percentages of farmers in three categories of nitrogen partial factor productivity (PFPN) and phosphorus partial factor productivity (PFPP) by country and gender across production systems. Production systems Country Gender PFPN (kg grain kg-1 N) PFPP (kg grain kg-1 P) Below desirable range (<40) Within desirable range (40-80) Above desirable range (>80) Below desirable range (<100) Within desirable range (100-250) Above desirable range (>250) Irrigated lowland Burkina Faso Female 60 38 2 18 48 33 Male 56 34 10 17 50 33 Burundi Female 62 28 10 6 31 62 Male 54 31 15 5 29 66 Democratic Republic of Congo Female 57 32 11 0 7 93 Male 51 33 16 2 9 89 Kenya Female 51 33 16 6 47 47 Male 36 45 19 2 43 55 Madagascar Female 88 6 6 95 3 2 Male 85 13 2 91 1 8 Nigeria Female 67 33 0 16 41 44 Male 61 35 4 14 13 72 Rwanda Female 23 72 5 0 23 77 Male 25 71 4 0 15 85 Sierra Leone Female 100 0 0 50 50 0 Male 100 0 0 100 0 0 Tanzania Female 26 58 16 2 30 69 Male 33 56 11 2 30 68 Rainfed lowland Burkina Faso Female 58 37 5 6 53 41 Male 62 33 5 34 26 40 Burundi Female - 33 67 0 0 100 Male 67 22 11 0 40 60 Democratic Republic of Congo Female 50 0 50 0 0 100 Male - - - - - - Madagascar Female - - - - - - Male 100 0 0 0 0 100 Nigeria Female 86 0 14 0 0 0 CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps 3. Results | Page 17 of 34 Male 69 27 4 2 17 80 Sierra Leone Female 70 26 4 20 53 27 Male 70 25 5 26 59 15 Tanzania Female 52 38 10 0 36 64 Male 53 42 5 0 22 78 Rainfed upland Madagascar Female 100 0 0 0 100 0 Male - - - - - - Nigeria Female 69 23 8 50 17 33 Male 54 42 4 5 17 78 Sierra Leone Female 60 20 20 100 0 0 Male 67 33 0 0 100 0 Tanzania Female 36 55 9 0 29 71 Male 0 100 0 0 67 33 Uganda Female 67 0 33 0 0 0 Male 33 0 67 100 0 0 Averages across all production systems Irrigated lowland Female 56 35 9 35 23 42 Male 53 39 8 21 19 60 All 18 48 34 16 13 71 Rainfed lowland Female 60 31 9 8 47 45 Male 64 31 5 16 29 55 All Rainfed upland Female 58 30 12 39 22 39 Male 49 45 6 6 20 73 All 74 14 12 3 2 93 3.4. Determinants of yield, net profit, and labor productivity The relative importance of different biophysical and management variables in explaining yield, net profit, and labor productivity outcomes for female and male farmers across production systems and countries is presented in Figure 6. Several consistent gender- specific patterns emerged. For women, the five most influential factors of yield were access to N input, plot size, irrigation number, use of certified seeds, and P input (Figure 6a, left insert). For men, the five key drivers included N input, irrigation number, P input, plot size, and labor input (Figure 6a, right insert). In terms of net profitability, the use of certified seeds, N input, plot size, and irrigation number were identified as the most influential factors for women farmers (Figure 6b, left insert). For labor productivity, labor input, plot size, and N input were the three primary drivers for both genders (Figure 6c). Other key factors included the use of certified seeds and irrigation for female farmers (Figure 6c, left insert) and equipment number and record keeping for male farmers (Figure 6c, right insert). These results suggest that while other drivers were important, the focus should be on improving access to N input, expanding irrigation opportunities, and increasing the availability of certified seeds and land areas for female farmers. For males, N input, plot size, and labor input were key determinants of improved yield, net profit, and CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps 3. Results | Page 18 of 34 labor productivity. Additionally, the use of equipment and record-keeping practices were significant factors in improving male labor productivity. Figure 6. Variable importance contribution of each predictor in terms of mean decrease accuracy for (a) yield, (b) net profit, and (c) labor productivity. Data are presented for female and male farmers. The degrees of influence of the drivers of yield, net profit, and labor productivity for female and male farmers are provided in Tables 5 and 6, respectively. N input, number of irrigations, number of equipment, P input, and labor input were positively associated with yield increases for both male and female farmers. Additionally, record keeping and training were associated with yield improvements for female and male farmers, respectively. However, plot size and production system were linked to yield reductions for both genders (Tables 5 and 6). For net profit prediction, the use of certified seeds, the number of irrigations, and production system contributed to profit improvement for both male and female farmers. In contrast, labor input, number of equipment, plot size, and P input were CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps 3. Results | Page 19 of 34 associated with reduced profit. While N input was associated with reduced profit for female farmers, it contributed to increased profit for male farmers (Tables 5 and 6). Labor productivity increased with N input, record keeping, and number of irrigations for both male and female farmers, while labor input, plot size, and use of certified seeds are positively associated with labor productivity (Tables 5 and 6). For female farmers, P input was associated with increased labor productivity, while the rainfed production system (compared to IL systems) was negatively associated with labor productivity. For male farmers, the number of equipment and training were associated with increased labor productivity. Optimizing nitrogen input, expanding irrigation, and increasing access to equipment, while addressing constraints in rainfed systems, are crucial for increasing yield for both male and female farmers. Likewise, promoting the use of certified seeds and irrigation and improving labor conditions through better equipment were essential to improve profit and labor productivity among female farmers. Critical points of factors such as plot size, labor input, and equipment use must be identified to ensure effective interventions for both male and female farmers. Table 5. Degree of influence of the key drivers of yield, profit, and labor productivity among female rice farmers. SRP PIs Variable Estimate t value Yield Intercept 1974.166 12.8*** N input 14.5188 7.9*** Plot size -35.0687 -0.9 Number of irrigations 8.4113 1.6 Number of equipment 373.4556 6.9*** P input 21.2846 2.4* Labor input 0.1629 0.7 Training 256.9906 5.4*** Production system | Rainfed lowland* -1080.46 -6.7*** Production system | Rainfed upland* -900.351 -5.3*** Profit Intercept 132.3876 1.4 Use of certified seed 300.3103 10.8*** N input -0.53319 -0.5 Plot size -13.4196 -0.6 Labor input -0.09616 -0.7 Number of irrigations 1.19231 0.4 Number equipment -100.612 -3.3** P input -1.80211 -0.4 Production system | Rainfed lowland* 541.9046 6*** Production system | Rainfed upland* 372.3434 3.7*** Labor productivity Intercept 29.85739 13.4*** Labor input -0.0308 -9.7*** CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps 3. Results | Page 20 of 34 Plot size -1.78693 -3.5*** N input 0.074158 3** Use of certified seed -0.9229 -1.3 Number of irrigations 0.093145 1.3 Production system | Rainfed lowland* -4.9487 -2.3* Production system | Rainfed upland* -11.8914 -5*** Record keeping 4.650061 5.8*** P input 0.019275 0.2 *, reference = Irrigated lowland Table 6. Degree of influence of the key drivers of yield, profit, and labor productivity among male farmers. SRP PIs Variables Estimate t value Yield Intercept 2387.947 18.467*** N input 8.7843 8.034*** Number of irrigations 22.9846 6.361*** P input 14.4832 2.111* Plot size -36.1153 -2.771** Labor input 0.4049 2.07* Production system | Rainfed lowland* -1039.95 -8.367*** Production system | Rainfed upland* -1414.97 -9.559*** Record keeping 276.6707 7.385*** Number of equipment 452.8802 10.553*** Profit Intercept 210.1717 2.943** Use of certified seeds 194.6349 10.116*** N input 0.70865 1.315 Plot size -8.8233 -1.38 Labor input -0.07241 -0.77 Number of irrigations 12.0577 6.806*** Number of equipment -23.2884 -1.114 P input -2.40703 -0.716 Production system | Rainfed lowland* 420.2727 6.922*** Production system | Rainfed upland* 278.6257 3.84*** Labor productivity Intercept 22.38797 11.553*** Labor input -0.03106 -10.045*** Plot size -0.907826 -4.333*** N input 0.09677 7.161*** Number of equipment 4.129398 5.979*** Record keeping 2.757284 4.121*** Training 0.281568 0.452 Number of irrigations 0.288456 5.751*** CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps 3. Results | Page 21 of 34 Use of Certified seed -1.64916 -2.481* *, reference = Irrigated lowland 3.5. Determinants of PFPN and PFPP Figure 7 presents gender-based differences in the drivers of PFPN and PFPP. Across all panels, the ranking and magnitude of variable importance differed notably between women and men, suggesting that the determinants of PFPN and PFPP are gender specific. For men, the key predictors of PFPN tended to show stronger effect sizes (i.e., longer bars), while for women, the influence of different variables was more evenly distributed. This indicates that PFPN outcomes for male farmers were strongly shaped by a few dominant factors, whereas women’s outcomes arose from a more complex combination of multiple variables. For women, N input, P input, labor input, irrigation frequency, and number of equipment were the strongest predictors, but the effect sizes remained relatively moderate and close to each other for the last four strongest predictors (Figure 7a, left insert). This suggests that women relied on a balanced combination of nutrient management and water control. For men, N input, irrigation number, P input, production system, labor input, and number of equipment dominated, with a much higher importance level for N input. Male outcomes appeared more sensitive to intensification variables, especially N and P applications and irrigation access. Regarding PFPP, P input, use of certified seeds, irrigation number, and N input were the primary drivers for women (Figure 7b, left insert), highlighting the importance of input quality and access to certified seeds and irrigation. For men, P input, irrigation number, P input, production system, and labor input emerged as the strongest determinants (Figure 7b, right insert). This indicates that men’s PFPP was more tied to nutrient management, access to irrigation, and production systems. CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps 3. Results | Page 22 of 34 Figure 7. Variable importance contribution of each predictor in terms of mean decrease accuracy for (a) PFPN and (b) PFPP. Data were disaggregated by gender. The multinomial logistic regression analysis identified the key factors influencing the likelihood of below, optimal, and above desirable PFPN and PFPP ranges (Tables 7 and 8). Optimal PFPN and PFPP served as the reference category. For PFPN, increases in P input, plot size, training, and land leveling were associated with reduced odds of low PFPN, while rainfed lowland systems increased the likelihood of low PFPN. The odds of high PFPN were higher with increased N and P inputs, more irrigation, training in sustainable rice practices, land leveling, equipment use, and RU systems. For PFPP, higher N and P applications, land leveling, equipment use, use of certified seeds, and RL systems increased the odds of low PFPP, whereas the odds of high PFPP increased with higher N input, irrigation number, training on sustainable rice practices, and use of certified seeds. These findings highlight the need for gender-responsive actions to enhance the sustainability of rice farming. Digital innovations such as RiceAdvice and CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps 3. Results | Page 23 of 34 RiceAdvice Lite for site-specific nutrient management and adoption of good agricultural practices offer practical solutions to improve sustainable rice cultivation in SSA. Table 6. Estimated parameters from the multinomial logistic regression analysis of the drivers of implementing optimal partial factor productivity of nitrogen (PFPN). Too low (Reference = Optimum) Too high (Reference = Optimum) Variables Coef. Std Error z value p value Coef. Std Error z value p value Intercept 1.41 0.25 5.7 0 0.38 0.28 1.32 0.1866 N input 1 0.00 0.19 0.8525 1.46 0.00 -13.74 0 P input 0.95 0.01 -5.23 0 0.96 0.01 3.62*** 0.0003 Number of irrigations 0.99 0.01 -1.46 0.1439 1.04 0.01 2.27* 0.0235 Plot size 0.84 0.08 -2.23* 0.0256 1.01 0.08 -1.59 0.1118 Training 0.84 0.06 -3.07** 0.0021 0.88 0.07 -2.81** 0.005 Land leveling 0.88 0.06 -2.12* 0.0336 0.83 0.08 3.03** 0.0024 Number of equipment 0.87 0.07 -1.94 0.0523 1.26 0.08 6.56 0 Production system | Rainfed lowland 1.54 0.21 2.12* 0.0341 1.67 0.25 -1.04 0.2997 Production system | Rainfed upland 1.20 0.27 0.68 0.4969 0.77 0.36 -2.34* 0.0191 Labor input 1 0.00 -0.94 0.3497 0.43 0.0005 -0.64 0.5234 AIC 2716.129 Residual Deviance 2672.129 Chi² 510.01 Pseudo R² of McFadden 0.1603 Log-likelihood -1336.06 Table 8. Estimated parameters from the multinomial logistic regression analysis of the drivers of implementing optimal partial factor productivity of phosphorus (PFPP). Too low (Reference = Optimum) Too high (Reference = Optimum) Variables Coef. Std Error z value p value Coef. Std Error z value p value Intercept 6.61 0.62 10.66 0 1.61 0.52 3.09** 0.002 N input -0.03 0.01 -4.78 0 0.01 0 3.64*** 0.0003 P input -0.16 0.02 -6.59 0 -0.16 0.01 -10.8 0 Number of irrigations 0 0.01 0.41 0.6825 0.02 0.01 3.14** 0.0017 Plot size 0.18 0.16 1.09 0.2753 0.07 0.11 0.69 0.4924 Training 0.12 0.12 0.99 0.3222 0.26 0.08 3.33*** 0.0009 Land leveling -0.31 0.12 -2.72** 0.0065 0 0.09 0.03 0.9793 Number of equipment -0.94 0.15 -6.37 0 0.02 0.09 0.28 0.7782 Production system | Rainfed lowland (Ref. = Irrigated lowland) -1.82 0.42 -4.36 0 -1.47 0.25 -5.86 0 Production system | Rainfed upland (Ref. = Irrigated lowland) 0.5 0.58 0.86 0.388 -0.61 0.38 -1.58 0.1144 Use of certified seeds -0.37 0.14 -2.58* 0.01 0.34 0.13 2.68** 0.0074 Labor input 0 0 -4.18 0 0 0 -3.68*** 0.0002 CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps 3. Results | Page 24 of 34 AIC 1455.172 Residual Deviance 1407.172 Chi² 966.03 Pseudo R² of McFadden 0.4071 Log-likelihood -703.59 CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps 4. Conclusions | Page 25 of 34 4. Conclusions In this study, we examined gender disparities in key sustainable rice platform performance indicators for agronomic gain, i.e., yield, net profit, labor productivity, and nitrogen and phosphorus partial factor productivity, across ten SSA countries and identified the most influential factors of these five SRP PIs on farmers’ yields and profitability, and reduced gender disparities. Results indicate the persistence of gender inequity in rice production systems in the majority of SSA countries. Similarly, notable gaps in yield, net profit, and labor productivity were found across production systems, irrespective of gender. Women’s resources use efficiencies depended on a broader combination of resource access, management skills, and environmental factors, suggesting a multidimensional constraint environment. Men’s resources use efficiencies were more tightly linked to access to inputs, labor, and equipment. These patterns point to the need for gender-differentiated intervention strategies: enhancing women’s access to improved inputs, training on sustainable rice cultivation, and water management, while supporting men with labor- saving technologies and efficient input-use strategies. To strengthen women’s involvement in sustainable rice cultivation across the ten SSA countries, it is essential to address these inequalities through the implementation of agricultural development projects. Digital tools such as RiceAdvice Lite can be very useful in enhancing the sustainability of these production systems, with a particular emphasis on the participation of women. CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps References | Page 26 of 34 References Abdallah, A.-H., Abdul-Rahaman, A., & Issahaku, G. (2021). Sustainable agricultural practices, farm income and food security among rural households in Africa. Environment, Development and Sustainability, 23(12), 17668-17701. https://doi.org/10.1007/s10668-021-01407-y Amede, T., Konde, A. A., Muhinda, J. J., & Bigirwa, G. (2023). Sustainable Farming in Practiceௗ: Building Resilient and Profitable Smallholder Agricultural Systems in Sub-Saharan Africa. Sustainability, 15(7), Article 7. https://doi.org/10.3390/su15075731 Arouna, A., Devkota, K. P., Yergo, W. G., Saito, K., Frimpong, B. N., Adegbola, P. Y., Depieu, M. E., Kenyi, D. M., Ibro, G., Fall, A. A., & Usman, S. (2021). Assessing rice production sustainability performance indicators and their gaps in twelve sub-Saharan African countries. Field Crops Research, 271, 108263. https://doi.org/10.1016/j.fcr.2021.108263 De Vos, K., Janssens, C., Jacobs, L., Campforts, B., Boere, E., Kozicka, M., Havlík, P., Folberth, C., Balkovič, J., Maertens, M., & Govers, G. (2023). Rice availability and stability in Africa under future socio-economic development and climatic change. Nature Food, 4, 1-10. https://doi.org/10.1038/s43016-023-00770-5 Devkota, K. P., Beebout, S. E., & Bunquin, M. A. (2022). Setting sustainability targets for irrigated rice production and application of the Sustainable Rice Platform performance indicators. Environmental Impact Assessment Review, 92, 106697. Gharsallah, O., Gandolfi, C., & Facchi, A. (2021). Methodologies for the Sustainability Assessment of Agricultural Production Systems, with a Focus on Rice: A Review. Sustainability, 13(19), Article 19. https://doi.org/10.3390/su131911123 Ibrahim, A., Saito, K., Kokou, A., Johnson, J.-M., Diagne, M., Fagnombo, D. J., Felix, F., Sylvia, B. O., & Martial, H. (2022). Seizing opportunity towards sustainable rice cultivation in sub-Saharan Africa. Environmental and Sustainability Indicators, 15, 100189. https://doi.org/10.1016/j.indic.2022.100189 Mane, E., Macchioni Giaquinto, A., Cafiero, C., Viviani, S., & Anríquez, G. (2025). Closing the gender gap in global food insecurity: Socioeconomic determinants and economic gains in the aftermath of COVID-19. Global Food Security, 45, 100850. https://doi.org/10.1016/j.gfs.2025.100850 OECD-FAO. (2021). OECD FAO Agricultural Outlook 2021-2030. OECD Publishing, Paris, 337. https://doi.org/10.1787/19428846-en Posit Team. (2025). RStudio: Integrated Development Environment for R. Posit Software, PBC,. (http://www.posit.co/). R Core Team. (2024). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. (http://www.R-project.org/). Sajjad, M., Hussain, K., Wajid, S. A., & Saqib, Z. A. (2025). The Impact of Split Nitrogen Fertilizer Applications on the Productivity and Nitrogen Use Efficiency of Rice. Nitrogen, 6(1), 1. https://doi.org/10.3390/nitrogen6010001 SRP. (2023). The SRP Standard for Sustainable Rice Cultivation (Version 2.2). Sustainable Rice Platform. Bangkok: 2023. https://sustainablerice.org/ Tseng, M.-C., Roel, Á., Macedo, I., Marella, M., Terra, J. A., & Pittelkow, C. M. (2021). Synergies and tradeoffs among yield, resource use efficiency, and environmental footprint indicators in rice systems. Current Research in Environmental Sustainability, 3, 100070. Yadav, S., & Kumar, V. (2018). Feeding the world while caring for the planet. Direct Seeded Rice Consortium, 1(2), 1-18. CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps References | Page 27 of 34 Yuan, S., Saito, K., van Oort, P. A. J., van Ittersum, M. K., Peng, S., & Grassini, P. (2024). Intensifying rice production to reduce imports and land conversion in Africa. Nature Communications, 15(1), 835. https://doi.org/10.1038/s41467-024-44950-8 CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps Appendices | Page 28 of 34 Appendices Table S1. Summary of the ANOVA outputs for yield gap. Df Sum Sq Mean Sq F value Pr (>F) Production system 2 95 47.4 0.12 0.887 Gender 1 410 410.2 1.041 0.314 Production system × Gender 2 53 26.5 0.067 0.935 Residuals 36 14179 393.9 Table S2. Summary of the ANOVA outputs for net profit gap. Df Sum Sq Mean Sq Fvalue Pr(>F) Production system 2 265 132.5 0.482 0.622 Gender 1 198 197.8 0.719 0.402 Production system × Gender 2 1169 584.7 2.126 0.135 Residuals 35 9628 275.1 Table S3. Summary of the ANOVA outputs for labor productivity gap. Df Sum Sq Mean Sq F value Pr(>F) Production system 2 1557 778.5 2.573 0.0906 Gender 1 24 24.5 0.081 0.7777 Production system × Gender 2 188 94.2 0.311 0.7344 Residuals 35 10588 302.5 Table S4. Average yield, net profit, labor productivity, partial factor productivity of nitrogen (PFPN) and partial factor productivity of phosphorus (PFPP) across production systems. Pooled data were used. Production system Yield ± se (kg ha-1) Net profit ± se (USD ha-1) Labor productivity ± se (kg man-day ha-1) NUE ± se (kg grain kg-1 N) PUE ± se (kg grain kg-1 N) Irrigated lowland 4546 ± 2167 a 783 ± 628 b 45 ± 46 a 67 ± 55 a 246 ± 293 a Rainfed lowland 2393 ± 1949 b 1163 ± 1474 a 31 ± 47 b 71 ± 64 a 317 ± 232 a Rainfed upland 2193 ± 1343 b 805 ± 1249 b 15 ± 23 c 48 ± 32 b 325 ± 218 a CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps Appendices | Page 29 of 34 Figure S1. Frequency distribution of partial factor productivity of nitrogen (PFPN) for female (left) and male (right) farmers in irrigated lowland (IL; top), rainfed lowland (RL; middle), and rainfed upland (RU; bottom) systems. CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps Appendices | Page 30 of 34 Figure S2. Frequency distribution of partial factor productivity of phosphorus (PFPP) for female (left) and male (right) farmers in irrigated lowland (IL; top), rainfed lowland (RL; middle), and rainfed upland (RU; bottom) systems. CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps Appendices | Page 31 of 34 Figure S3: Variable importance contribution of each predictor in terms of mean decrease accuracy for (a) yield, (b) net profit, (c) labor productivity, (d) PFPN, and (e) PFPP. Pooled data were used. CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps Appendices | Page 32 of 34 Citation: Bagri B, M., Kouadio, L., Ibrahim, A., Kouamé, P., Senthilkumar, K. 2025. Multi-country evidence of gender gaps in sustainable rice cultivation performance indicators in sub-Saharan Africa. CGIAR Sustainable Farming Science Program Report. Africa Rice Center, Bouaké, Côte d’Ivoire. Acknowledgements The CGIAR Sustainable Science Program forms a part of CGIAR’s new Research Portfolio, addressing key challenges in agri-food systems by fostering efficient production of nutritious foods and safeguarding the environment to create fair employment opportunities, as we simultaneously tackle climate change, soil degradation, pests, diseases, and desertification. Its research is being implemented by CGIAR researchers from 13 CGIAR Research Centers (AfricaRice, Alliance of Bioversity and CIAT, CIMMYT, CIP, ICARDA, IFPRI, IITA, ILRI, IRRI, IWMI and WorldFish), in close partnership with National Agricultural Research and Extension Services in Burkina Faso, Burundi, Democratic Republic of Congo, Kenya, Madagascar, Nigeria, Rwanda, Sierra Leone, Tanzania, and Uganda. We would like to thank all funders who supported this research through their contributions to the CGIAR Trust Fund: https://www.cgiar.org/funders/ About CGIAR Sustainable Farming Science Program Report This research was conducted as part of the CGIAR Sustainable Farming Science Program. This research is being implemented by CGIAR researchers from Africa Rice Center (AfricaRice) in close partnership with National Agricultural Research and Extension Services in Burkina Faso, Burundi, Democratic Republic of Congo, Kenya, Madagascar, Nigeria, Rwanda, Sierra Leone, Tanzania, and Uganda. CGIAR is a global research partnership for a food-secure future. Its science is carried out by 15 Research Centers in close collaboration with hundreds of global partners. www.cgiar.org Photos © Africa Rice Center Disclaimer This working paper has not been peer reviewed. Any opinions stated herein are those of the authors and do not necessarily reflect the policies or opinions of Africa Rice Center (AfricaRice), donors, or partners. This publication is copyrighted by Africa Rice Center. It is licensed under a Creative Commons Attribution–Non-Commercial 4.0 International License. To view this license, visit https://creativecommons.org/licenses/by/4.0. Unless otherwise noted, you are free to share (copy and redistribute the material in any medium or format), adapt (remix, transform, and build upon the material) for any purpose, even commercially, under the following conditions: ATTRIBUTION: The work must be attributed, but not in any way that suggests endorsement by Africa Rice Center or the authors. NOTICE: For any reuse or distribution, the license terms of this work must be made clear to others. Any of the above conditions can be waived if permission is obtained from the copyright holder. Nothing in this license impairs or author’s moral rights. Fair dealing and other rights are in no way affected by the above. The parts used CGIAR Sustainable Farming Science Program Report | Gender-differentiated PIs gaps Appendices | Page 33 of 34 must not misrepresent the meaning of the publication. AfricaRice would appreciate being sent a copy of any materials in which text, photos, etc., have been used. ©2025 Africa Rice Center Key Words: Sustainable rice production; gender gaps; rice production systems; agronomic gain. Partners National Agricultural Research and Extension Services in Burkina Faso, Burundi, Democratic Republic of Congo, Kenya, Madagascar, Nigeria, Rwanda, Sierra Leone, Tanzania, and Uganda. About CGIAR Sustainable Farming Science Program The CGIAR Sustainable Farming Science Program will address key challenges in agrifood systems by fostering efficient production of nutritious foods and safeguarding the environment to create fair employment opportunities, as we simultaneously tackle climate change, soil degradation, pests, diseases, and desertification. CGIAR Sustainable Farming Science Program Report | Gender differentiated PIs gaps Appendices | Page 34 of 34