Gendered analysis of fish feed ingredients and sourcing practices in Western Kenya’s aquaculture

Abstract

Introduction:

Fish feed management plays a central role in aquaculture productivity, but it is rarely analysed through a gender lens. This study examines how women and men differ in their use and sourcing of fish feed ingredients in smallholder aquaculture systems in Western Kenya. Drawing on feminist economics, it explores how everyday roles, access to resources, and mobility shape feed practices.

Methods:

The analysis uses survey data from 213 fish farmers, including 112 women and 101 men, across six counties in Western Kenya. A binary logit model was used to assess factors associated with the use of selected feed ingredients, including poultry waste, plant leaves, and maize powder, while controlling for socio economic characteristics, mobile phone ownership, farming system, and county.

Results:

Clear gendered patterns emerged. Women reported higher use of maize powder, poultry waste, cassava waste, chicken manure, feathers, and other household organic by products. Men were more likely to use plant leaves, kitchen leftovers, rice bran, and a broader mix of other inputs, often accessed through male networks. Differences were also evident across farming systems. Semi intensive earthen ponds relied on more purchased and diverse inputs, intensive systems focused on energy dense staples such as maize and cassava, and extensive systems drew heavily on locally available organic materials. Men were more commonly involved in applying plant leaves and chicken manure in these systems. Regression results show that gender remains significantly associated with the use of poultry waste and plant leaves, but not maize powder, suggesting that some differences are not fully explained by location or production system.

Discussion:

These findings point to feed practices as embedded in gendered divisions of labour and resource access rather than neutral technical choices. Women’s roles in cooking, managing household waste, and caring for backyard poultry provide regular access to organic by products that can be used as feed. Men’s greater control over land, cash, and mobility supports access to field based and market sourced inputs. This has practical implications for extension and advisory services, which need to recognise these differences rather than assume uniform practices. It also highlights the importance of monitoring how changes in feed practices affect women’s time use, decision making, and control over income.

Introduction

There is a dearth of literature on the gender differences in the use of fish feed ingredients (Ganguly et al., 2021). The collection of sex-disaggregated data from past fish feed studies has been inconsistent, resulting in a lack of understanding of women’s contributions to aquaculture production. This study examined fish farming systems and the diverse range of ingredients employed in Western Kenya to shed light on gender differences in the use of novel feed ingredients.

Aquaculture development in Kenya has shown steady growth, driven by supportive government policies and public investments in the sector. According to the Kenya National Bureau of Statistics, Kenya’s annual aquaculture production is approximately 18,000 tons (KNBS, 2021). By 2021, aquaculture production in Kenya had soared to 21,285 metric tons, according to the World Bank (2023). This expansion brought about numerous benefits for the nation. First, they play a vital role in enhancing food security by providing a steady source of fish protein to meet the population’s dietary needs. It also serves as a substantial source of income for local communities engaged in fish-farming activities, thereby contributing to poverty alleviation. Importantly, aquaculture production provides alternative sources of fish, thereby reducing the need to exploit capture fisheries. This, in turn, reduces the ecological pressure on natural fishery stocks in rivers, lakes, and the ocean, thereby contributing to ecosystem conservation.

This aquaculture expansion is closely tied to the high demand for quality and cost-effective fish feed, particularly among small-scale fish farmers in rural Kenya. Fish feed directly influences optimal growth, health, and quality (Craig and Helfrich, 2017). High-quality feeds improve feed conversion efficiency and lead to higher production yields than those from capture fisheries (FAO, 2018; Kok et al., 2020), thereby enhancing the profitability of aquaculture operations (Munguti et al., 2021a). However, the cost of fish feed is rising considerably. Fish feed typically accounts for 40–60% of production costs (Munguti et al., 2021c), significantly reducing the profitability and viability of aquaculture enterprises (Enyidi et al., 2017; Singh et al., 2018). High costs are further exacerbated by inflation, high taxation, and other economic factors in Kenya. Given the crucial role of aquaculture in the economy, finding solutions for reducing fish feed costs is imperative.

In Kenya, smallholder farmers rely on farm feeds made from locally available agricultural by-products, including vegetables and animal proteins, as a supplementary feed. Other home-made feeds generated using fertilisers or manure (Munguti et al., 2021b) are nutrient-rich and easily ingested and digestible by fish. Commercially produced feed is imported from regional or international markets or locally manufactured using cereal by-products (Munguti et al., 2021b). Poor implementation and lack of monitoring mechanisms for fish feed have led to the domestic production of low-quality feed (Munguti et al., 2021c). Currently, existing information networks need to be more robust at promoting the exchange of knowledge on feed ingredient availability, quality, pricing, and authorised suppliers (Munguti et al., 2021c). Therefore, the need for high-quality, safe, and affordable fish feed ingredients, seeds, and fingerlings remains a significant challenge for smallholder fish farmers.

At the same time, gender dynamics shape access to resources, decision−making, and labour roles in aquaculture. Existing studies rarely collect gender−disaggregated data on fish feed production, procurement, and management, making it difficult to identify women’s contributions or design interventions that work for them. Given the central role of household food waste in many farm−made feeds, and the traditional responsibility of women for managing kitchens and home gardens, it is important to understand current feed−use patterns from a gender perspective. This study responds to that gap by examining gender differences in the use, access, and preferences for fish feed ingredients. The paper reviews the literature, then describes the methodology and sample, before presenting the findings and their implications.

Literature review

Gender and aquaculture

Gender disparities significantly affect farming practices and the adoption of technology in aquaculture. Despite statutory laws in Kenya that support women’s land ownership, cultural norms often prevent them from inheriting land, leaving them dependent on male family members for indirect access (Kruijssen et al., 2018). Women often face limited access to essential resources, including land, capital, profitable markets, high-value fish, and assets such as freezers or transportation (Fröcklin et al., 2013; Githukia et al., 2020). Disparities in fish pond ownership and resource access, typically controlled by male household heads, also contribute to lower women’s participation in aquaculture and limit their access to credit facilities that often require land as collateral (Barclay et al., 2022; McClenachan and Moulton, 2022; Torre-Castro, 2019). Even when women do access capital, they do so on unequal terms, setting them up for hardship and limiting their ability to invest in improved farming methods, such as higher-return cage farming, or to purchase essential inputs (McClenachan and Moulton, 2022).

Gender and cultural norms further restrict women fish farmers. Cultural norms dictate labour and household duties (Chepkirui et al., 2023). Women’s mobility and ability to build business contacts are often restricted by household and childcare responsibilities, which can also affect their flexibility in sales and distribution (Kantor and Kruijssen, 2014; Fröcklin et al., 2013). Even when spending similar amounts of time on trade, women may earn less than men, and their work is insecure (Fröcklin et al., 2013; Barclay et al., 2021). Brugere et al. (2023) caution that aquaculture should be ‘humanised’ to prevent irreversible social harms by removing exploitation, inequitable benefit distribution, and weak labour protections, which disproportionately affect women.

Women’s contributions, such as making and mending fishing nets, pond maintenance, including processing, grinding, mixing, pelleting, and packaging fish feed, are often considered extensions of household duties that remain largely invisible and undervalued (Barclay et al., 2021; Chepkirui et al., 2023; Awuor, 2021; Chepkirui et al., 2023). Furthermore, lower levels of education and literacy among women can impede their access to information and resources, making it difficult for them to adopt innovative agricultural practices and mechanised farming technologies, which could generate more revenue (Barclay et al., 2021).

Formal fisheries management often overlooks the key roles of women in local trade and processing, focusing instead on male-dominated capture fisheries and export markets (Munguti et al., 2023). Sociocultural norms also contribute to male dominance and women’s underrepresentation in governance structures, such as cooperatives, hindering their active participation in decision-making, even when formally members (OCDC International Cooperative Research Group, 2019; Kruijssen et al., 2018; Lentisco and Lee, 2015). A study using the Abbreviated Women’s Empowerment in Fisheries and Aquaculture Index (A-WEFI) found that, while both men and women were largely empowered, women reported a lack of agency in production, resources, time use, and leadership (Adam et al., 2023).

However, when the extension is gender responsive, more benefits accrue to women. Brugere et al. (2020) found that extension training, when delivered to women, helps them acquire new skills while fostering group solidarity and confidence. These social dimensions were noted to be as important as the technical benefits. Similarly, Kantor and Kruijssen (2014) found that when extension services reach women, they reduce market barriers and increase women’s sales and earnings by improving product quality and bargaining power. Meanwhile, Ghosh (2021) illustrates how gender-responsive, hands-on extension can boost both feed use and women’s market entry.

Successful gender-responsive extension practices intentionally include women and adapt approaches to their specific needs and contexts. Examples from Bangladesh, Egypt, and India demonstrate this through the deliberate recruitment of women for training, as evidenced by 55% female participation in Bangladesh’s AIN program (WorldFish, 2020). Training local sellers and establishing community centres in Bangladesh also made advice and inputs more accessible by reducing travel and time burdens for women. Furthermore, providing tailored business skills, hygiene training, and safer equipment to women fish retailer groups in Egypt, along with linkages to suppliers, reduced market barriers and boosted their sales and earnings by improving product quality and enhancing their bargaining power (Kantor and Kruijssen, 2014). In fact, Torre-Castro (2019) asserts that sustainability necessitates the conscious inclusion of all genders and social groups in fish and water governance.

Beyond access, these practices also empower women through skill development and provide them with economic opportunities. In West Bengal, India, women’s self-help groups were trained to produce and sell formulated fish feed using local ingredients, creating new income streams and improving local access to affordable feed (Ghosh, 2021). Similarly, Bangladesh’s BANA program partnered with private companies to deliver farmer training and advisory services that specifically included women, shifting training to local nodes and adapting delivery to ensure their attendance, thereby expanding their reach to provide advice and connect with suppliers (WorldFish, 2020). In Zanzibar, participatory training on tubular nets for women seaweed farmers not only provided technical benefits but also helped them acquire new skills, such as boat use and navigation, leading to increased income, self-esteem, and a stronger sense of community (Brugere et al., 2020). These examples demonstrate how gender-sensitive design and extension services can empower women, increase their incomes, and enhance the functioning of value chains.

Fish feed ingredients and gender

Aquaculture feeds have traditionally relied on fishmeal and fish oil, but recent efforts towards sustainability have prompted research into alternative dietary sources (Kwasek et al., 2020). Natural foods, such as algae commonly found in ponds, have been explored as part of this shift, although they result in slower fish growth (Weimin and Mengqing, 2007). Weimin and Mengqing (2007) proposed fertilising ponds with urea and diammonium phosphate (DAP) to enhance productivity and nutrition.

Munguti et al. (2012) studied the nutritional composition of fish feed ingredients and identified potential sources of both animal- and plant-based proteins for fish farming in Kenya. Currently, Omena (Rastrineobola argentea) and Ochonga (Caridina nilotica) are primary sources of animal proteins (Munguti et al., 2021b). Additionally, cereal by-products and oilseeds are utilised in feed formulation; however, their availability is influenced by the harvesting season, which can be inconsistent due to climate change (Munguti et al., 2014). Craig and Helfrich (2017) warned that plant-based fish feed lacks essential nutrients, such as methionine, and thus requires supplementation with other amino acids, lipids, and carbohydrates. Moreover, plant-based fish feed can compromise digestibility due to its high fibre content, thereby reducing fish yields.

Ganguly et al. (2021) highlighted various alternative raw ingredients for fish feed formulation, focusing on their nutritional value. These alternatives include soybean meal, sunflower expellers, liquid-dried grain distillers, cassava, cucumber, papaya, white cowpea, green mung beans, and cotton meal. Furthermore, wheat pollard, cottonseed cake, groundfish meal, blood meal, neem seed cake (NSC), and soybean oil cake have been recognised for their high protein content, presenting highly nutritious options (Daniel, 2016). Notably, NSC and soybean are affordable, making them cost-effective alternatives to fish-feed formulations.

Ganguly et al. (2021) highlighted inconsistencies in the collection of sex-disaggregated data in past fish feed studies, noting that knowledge of women’s contributions to aquaculture production was limited and patchy. Few studies have investigated whether households or farms are primarily responsible for purchasing and financing fish feeds, or how feed costs impact women-run operations economically (Adam et al., 2023; Obiero et al., 2019). Limited studies have explored the challenges women face regarding fish feed supply chains, including negotiating power with suppliers, technical knowledge of feed formulations, and feeding practices (Obiero et al., 2019).

Data-driven aquaculture technologies rarely capture or analyse sex-specific information on feeding regimes or feed management (Adam et al., 2023). Studies assessing women’s empowerment in aquaculture often rely on aggregate data on production or overall inputs, without isolating fish feed use, control over procurement, or feed-related decision-making roles (Adam et al., 2023). For example, research using the A-WEFI in Kenyan counties such as Kakamega and Kisumu broadly measured empowerment. However, they did not include fish feed-specific parameters (Adam et al., 2023). There is limited knowledge on whether men and women have differential access to or preferences for specific feed types, such as commercial pellets, homemade feed, or agricultural by-products (Obiero et al., 2019). This gap hinders our understanding of gendered access to quality, affordable food, which, in turn, affects productivity and overall well-being.

Capacity-building and extension services often overlook gender-specific needs related to fish feed management, and there is a lack of assessment of how fisheries extension and advisory services integrate gender considerations into fish feed use, thereby limiting opportunities to tailor support effectively (Adam et al., 2023). This knowledge is critical for designing interventions that empower women financially and reduce gendered constraints. Ganguly et al. (2021) emphasised a risk arising from this lack of knowledge: introducing new ingredients could negatively affect women who already use them in other forms of production, and any negative impact on women would likely go unnoticed by men, underscoring the importance of this study.

Methodology

Research design

This study formed part of a broader investigation conducted from April to June 2023 on the gendered risks associated with novel fish−feed ingredients. We used a mixed−methods design guided by the gender analysis domains in USAID ADS 205 (USAID, 2017). A desk−based literature review was first undertaken to gather secondary data and identify gaps, followed by a structured survey of 213 fish farmers (112 women and 101 men). The survey tools were developed specifically for this study to focus on fish feed ingredients and to ensure that women were well represented among respondents. Questions covered sociodemographic characteristics, types of fish farming and systems, the ingredients farmers used, where they sourced them, and why they chose particular feeds. The list of potential ingredients drew on earlier desk research (Ganguly et al., 2021) and was used to structure multiple−choice and Likert−scale questions, as summarised in Table 1.

In contrast to much of the earlier work, which has relied on case studies, interviews, focus groups, policy analysis, and other qualitative tools, our survey was designed from the outset to collect sex−disaggregated data on feed ingredients. For example, while Weeratunge et al. (2010) used survey data, they drew on information collected for other purposes and had a larger share of men in the sample.

Purposive sampling was used in the survey. Enumerators began by engaging with the district Director of Fisheries, who oversees fisheries management within the district or county, and explained the survey’s objectives. District fisheries staff then helped identify communities engaged in fish farming and locate relevant farmers, using registration lists maintained by the Fisheries Office Department. Enumerators were accompanied by a sub−district fisheries officer, who facilitated introductions and helped build trust with farmers. Preliminary discussions were organised to schedule interviews at times convenient for participants.

Definitions

Definitions of each ingredient category were standardised in the enumerator manual and explained to respondents during data collection, but self−reports reflect farmers’ own classifications and experiences. In the questionnaire, ‘Genetically Modified Plants (GMP)’ referred to genetically modified crop materials used as feed ingredients, such as GM maize or soybean. Enumerators introduced this item using locally known examples of GM crops where relevant, but the survey did not specify which plant parts were used (for example, grain versus leaf). We therefore report GMP use at the ingredient category level and cannot distinguish whether farmers applied whole grain, processed meals, or vegetative material. This limits the nutritional and environmental interpretation of the GMP category, and we treat it as a coarse indicator of any use of genetically modified plant material in fish feed.

In this study, ‘plant leaves’ referred to leafy vegetation that farmers reported collecting and using for fish feed around their farms and ponds. Enumerators did not systematically record botanical species or whether leaves were from weeds, trees, or crop residues, and the survey did not distinguish among these subtypes. As a result, the ‘plant leaves’ category combines multiple sources of leafy biomass, and our analysis cannot specify which plant species or plant types were most important. This limits our ability to draw conclusions about the nutritional quality or ecological implications of specific leaf sources, but it does capture the overall reliance on locally available leafy biomass as a feed ingredient.

Given time and resource constraints, we did not collect detailed information on the exact processing steps, quantities, or the nutritional composition of each ingredient at the farm level; therefore, our analysis is limited to whether each ingredient was reported to be used.

Study locations

The study’s geographical focus was on fishing communities in Western Kenya, specifically targeting the districts/counties of Kisumu, Busia, Kakamega, Vihiga, Homa Bay, and Siaya in the Lake Victoria Catchment area. To analyse the determinants of farmers’ preferences for specific fish feed ingredients, we estimated binary logit models (see Equations 1–6). Figure 1; Table 2 display the sample sizes by location.

Figure 1

Lake Victoria serves as a hub for both subsistence and commercial fishing, offering employment opportunities for men, women, and young people. However, western Kenya is considered rural, with limited infrastructure and access to information, productive resources, technology, extension services, and market opportunities. Additionally, open access has led to the overexploitation of fishery resources and a decline in fish stocks, threatening the livelihoods and cultural identity associated with fishing. A lack of livelihood diversification, combined with various socioeconomic and environmental factors, has led smallholder fish farmers to abject poverty. They are also highly prone to climate change and external shocks (Barclay et al., 2021).

Women largely participate in fish processing, marketing, and value addition, rather than fish farming. However, when they are involved in farming, it is typically through intensive farming because of lower maintenance and entry costs. It is also important to note that approximately 76% of cage culture, and over 80% of pond culture, and the majority of overall production in Kenya is dedicated to Nile Tilapia, with Catfish constituting a much smaller proportion (Munguti et al., 2022; Opiyo et al., 2018). The variation in feed ingredients cannot be attributed to the fish species because both are fed the same ingredients.

Source: Abwao et al, 2021.

Empirical framework

To analyse the determinants of farmers’ preferences for specific fish feed ingredients, we estimated binary logit models. Logit was preferred to probit because the dependent variables are binary and the logit specification converged more reliably. For each ingredient of interest (poultry waste, plant leaves, maize powder), we defined a binary outcome equal to 1 if the farmer reported using the ingredient and 0 otherwise. The explanatory variables included education, age, gender, mobile−phone ownership, group membership, farming system, on−farm income, and county. We derived marginal effects to show how changes in these variables influence the probability of using each ingredient, tested for multicollinearity using variance inflation factors, and assessed model fit using pseudo−R² and likelihood−ratio tests.

Empirically, the dependent variable was specified as follows:

Following Greene (2003), the probability that a fish farmer uses a fish feed ingredient is expressed as:

Where i and j represents farmer and fish feed ingredient preference with 1 indicating preference and 0 otherwise, X is the vector of independent variables for the i farmer.

The empirical estimation of the probability that a farmer uses a fish feed ingredient is expressed as:

Where X represents the vector of independent variables, β is the parameter to be estimated and ϵ is the error term. The factors that are hypothesised to influence fish feed ingredient preference are presented in Table 3.

Marginal effects were computed to determine how changes in the independent variables influence the probability of fish feed ingredient preference. Following Anderson & Newell (2003), the marginal effects for continuous variables and binary variables were computed, respectively as:

A multicollinearity test was done using the variance inflation factor (VIF) for variables included in the model, as:

Ethics

The Includovate Human Research Ethics Committee (HREC) approved this study. All participants provided informed consent through a signature or thumbprint. They were informed of the study’s objective, their right to participate freely, and the confidentiality, anonymity, benefits, and risks associated with participation.

Limitations of the study

Employing quota and snowball sampling can introduce bias, for example, by over−representing more outspoken or well−connected farmers and missing quieter or more isolated individuals. We sought to address this by purposively sampling underrepresented groups such as women, young people, and farmers in remote areas, and by using quotas to improve balance across the sample. Multiple referral sources and screening criteria were used to reduce dependence on any single contact point, although some government lists were outdated and did not always include women fish farmers.

During piloting, each enumerator administered the survey three times. Minor adjustments to the wording were made, mainly to improve translations and to incorporate local phrases where they helped comprehension. We did not, however, formally test the reliability of the Likert scales because of time constraints in the field. Seasonality and self−reporting bias are also possible, as feed choices can change with prices and availability over the year.

The survey did not collect detailed information on the nutritional content of each ingredient or on precise quantities used, as this fell outside the scope of the study. Nonetheless, the responses highlighted a strong demand for better information on nutrition and feed quality and revealed instances of misinformation. These limitations point to the need for future work that integrates on−farm nutritional analysis, tracks seasonal variation, and explores how women and men learn about feed ingredients and evaluate their quality.

Findings

Fish feed ingredients and usage

The analysis of fish feed ingredients used in Kenya shows variations in usage between men and women across different feed types. Rice Bran and Wheat Bran are the most commonly used ingredients, with 60.7% of farmers incorporating them into fish feed, including 66.3% of women and 55.4% of men (Table 4). Plant leaves are also widely used, with 59.2% of farmers reporting this, and 67.3% of men and 50.5% of women utilising them.

Other frequently used ingredients include maize powder (21.4%), which is more commonly used by women (29.5%) than men (13.9%), and poultry waste (19.9%), used by 26.3% of women and 13.9% of men. Kitchen leftovers (13.8%) are also a notable ingredient, with 16.8% of men and 10.5% of women reporting their use. Cassava waste (15.3%) is used more by women (20.0%) than by men (10.9%), while chicken manure and feathers (16.3%) follow a similar pattern, with women (21.1%) more prevalent than men (11.9%).

Less commonly used feed ingredients include earthworms (9.7%), genetically modified plants (6.6%), microalgae (5.6%), JSRSM (1.5%), and ghee residue (0.5%), while ‘Other’ feeds account for 43.9% of total usage, with higher use among men (50.5%) than women (36.8%).

Chi-square tests confirm gendered differences for poultry waste (χ²=4.01, df=1, p=0.045), plant leaves (χ²=5.05, df=1, p=0.025), and maize powder (χ²=6.19, df=1, p=0.013), while other ingredients show no statistically significant gender gaps.

Regarding income contribution, 32.0% of respondents reported that fish farming contributed 21–40% of their household income, and 29.3% reported a contribution of 41–60%. A chi−square test showed no statistically significant relationship between gender and the percentage of household income derived from fish farming (χ²=6.68, df=4, p=0.154). By contrast, farm system type was associated with gender (χ²=8.61, df=2, p=0.013), with men being underrepresented in intensive systems compared to women.

Fish feed ingredients by district/county

The findings from various counties provide a comprehensive understanding of the differences in fish feed ingredient usage between men and women (Table 5). Notable trends include higher reported use of poultry waste among women than among men across most counties, consistent with the pooled results, as well as county−specific variation in how often poultry waste is used. For example, poultry waste use was highest in Busia and Homabay and lowest in Siaya, with women in each county more likely than men to report using it. Across all districts, plant leaves and rice bran are the most commonly used ingredients for both genders. In contrast, ingredients such as jute, ghee residue, genetically modified plants (GMP), microalgae, and non-conventional plant sources were used by only a few individuals.

Kakamega recorded the highest use of potato waste, while plant leaves were most common in Kisumu. Rice Bran and Wheat Bran were used heavily in Homabay and Siaya, reflecting a strong reliance on bran and oil−cake feeds in these counties. Cassava waste was most frequently reported in Siaya among women, whereas GMP use was concentrated in Homabay and Kakamega. Kitchen leftovers were more common in Kakamega and Kisumu than in other counties. Chi−square tests confirm county−level variation for maize powder (p<0.001), plant leaves (p<0.001), the bran and oil−cake group (p<0.001), cassava waste (p=0.010), GMP (p=0.014), and kitchen leftovers (p=0.029).

Fish feed ingredients by farm system

The analysis of fish feed use across earthen−pond farming systems shows clear differences among extensive, intensive, and semi−intensive production. Semi−intensive systems rely on a broad mix of ingredients, with particularly high use of bran, maize powder, poultry waste, chicken manure and feathers, and less common ingredients such as GMP and microalgae. Intensive systems draw more on energy− and carbohydrate−dense ingredients such as cassava waste, maize powder, potato waste, and a wide range of “other” feeds, reflecting higher stocking densities and growth targets. Extensive systems rely more heavily on organic and locally available inputs, including chicken manure and feathers, earthworms, kitchen leftovers, plant leaves, and JSRSM, alongside some bran, microalgae, and non−conventional plant sources. Chi−square tests confirm significant associations between feed use and farming intensity for several ingredients, including maize powder, cassava waste, plant leaves, and poultry waste, whereas cage farming showed no systematic differences in feed use across intensities (see Table 6).

The analysis of cage systems reveals a stark contrast to earthen−pond systems, with only a few ingredients recorded across extensive, intensive, and semi−intensive cage farming (Table 7). Rice Bran and Wheat Bran are the most used ingredients in cage systems, appearing across all three farming methods (extensive, intensive, and semi-intensive), with the highest use in semi-intensive cage farming. Other notable feed ingredients include earthworms (Eisenia fetida) and plant leaves, which are exclusively used in semi-intensive cage systems. In cages, cassava waste appears only in intensive systems and microalgae only in extensive systems, in contrast to the broader ingredient mix in earthen ponds. Meanwhile, the absence of ghee residue, poultry waste, and chicken manure in cages suggests that nutrient-enrichment strategies commonly used in earthen ponds do not apply to cage systems.

Fish feed ingredients by farm intensity and gender

Building on the earthen−pond and cage comparisons above, Table 8 shows how feed use varies by gender across extensive, intensive, and semi−intensive earthen−pond systems. Table 8 provides an overview of fish feed use in earthen pond farming systems, disaggregated by gender (women and men) and farming intensity (extensive, intensive, and semi-intensive).

Men use more poultry waste, rice bran, and GMP in semi-intensive farming. In contrast, women rely more on cassava waste and JSRSM in extensive farming, while men use more plant leaves, chicken manure, and feathers in extensive systems. Men show higher adoption of kitchen leftovers and maize powder, whereas women incorporate a wider variety of organic waste, including ghee residue (100 percent in semi−intensive farming).

Sourcing fish feed ingredients

The analysis reveals no significant relationship between gender and the source of fish feed or farm type. Table 9 presents an overview of the sources of fish feed ingredients, specifically for plant leaves and rice bran, disaggregated by gender. The data indicate that plant leaves are primarily sourced through home processing, with 108 farmers (46 women and 62 men) relying on this method. Bartering with neighbours is the second most common method of sourcing plant leaves, as reported by 32 farmers (11 women and 21 men). Local stores and suppliers are a less frequent source, with only 11 farmers (three women and eight men) sourcing plant leaves from them. Other methods, such as purchasing from big-town suppliers (1 farmer) or collecting from animal waste (1 farmer), are used minimally, while no respondents reported obtaining plant leaves from government stores.

Rice bran is primarily sourced from local stores or suppliers, with 111 farmers (59 women and 52 men) purchasing it from these sources. Many farmers also rely on urban suppliers, with 25 farmers (9 women and 16 men) obtaining rice bran in this manner. Home processing is a less common method of acquiring rice bran (7 farmers: 2 women, 5 men), and bartering is even rarer (3 farmers: 2 women, 1 man). A few farmers also obtain rice bran from animal waste (one farmer) and government stores (one farmer), but these methods are rarely employed.

While plant leaves are predominantly self-processed, rice bran is mainly purchased from local and big-town suppliers. Additionally, men tend to source plant leaves through bartering more than women, whereas women rely slightly more on local stores for rice bran. While there is no relationship between gender, farm type, or feed source, the analysis for plant leaves suggests a statistical association between county and the source of fish feed ingredients. Sourcing of plant leaves varies by county—barter (p<0.001), processing at home (p<0.001), and purchases from local stores (p=0.0027) all show significant county differences, while ‘animal waste’ and ‘bigger-town suppliers’ do not.

Intra-household variance

Table 10 provides an overview of the use of fish feed ingredients (plant leaves and rice bran) by different household members, disaggregated by gender. The data indicate that spouses are the primary users of plant leaves, whereas rice bran is used more evenly across users. Other family members who use plant leaves include male siblings (5 households), sons (4 households), daughters and female siblings (2 households each), mothers/mothers-in-law (2 households), and all family members together (1 household). Fathers/fathers-in-law do not participate in using plant leaves at all.

In contrast, rice bran had greater involvement from male family members: fathers/fathers-in-law (10 households), sons (11 households), and male siblings (8 households) compared to plant leaves. Mothers/mothers-in-law (2 households), daughters (1 household), and female siblings (1 household) have minimal participation in rice bran usage. Unlike plant leaves, no household reported using rice bran as a whole family. The analysis reveals no relationship between gender and the use of ingredients by other household members, nor between the county and the use of ingredients by other household members.

A chi−square test confirms an association between earthen−pond farm system and the family member using rice bran (χ²=59.0, df=24, p<0.001). Father/Father-in-law and male siblings are more involved in intensive and semi-intensive systems. Some groups, such as Daughters and Family (including all family members), exhibit little to no involvement across all categories. Spouses are the most involved across all systems, but their involvement is not evenly distributed across categories. However, this relationship does not exist for plant leaves.

Characteristics of fish farmers by gender

Table 11 shows the Socio-economic characteristics of fish farmers, disaggregated by gender. The average age of fish farmers was 52 years. There was no statistical difference in education across the groups. Forty five per cent of respondents used a semi-intensive farming system and were members of an agricultural group. Ownership of mobile phones was 28% in the pooled sample, with notable gender disparities at the county level. Female farmers accounted for the majority of mobile phone ownership (57%) in Busia County, compared with their male counterparts (17%). In terms of on-farm income, 60% of respondents in the pooled sample derived between 21% and 60% of their income from fish farming. The majority of male farmers in Kisumu County (71%) and Vihiga (77%) generated 21-60% of their income from fish farming, compared with their female counterparts, possibly due to greater male involvement in fishing activities.

Binary logit model results

Table 12 presents the binary logit regression results on factors influencing fish feed ingredient preferences in Western Kenya. Variance inflation factors (VIFs) for all variables were well below the conventional threshold of 10 (mean VIF 2.75), indicating no evidence of problematic multicollinearity (Table 13). The models include farmer gender, age, education, on−farm income, mobile phone ownership, group membership, farming system, and county fixed effects as covariates.

After controlling for these factors, gender remained significantly associated with the use of poultry waste and plant leaves, but not maize powder. Men were 15 percentage points less likely than women to use poultry waste (marginal effect −0.15, SE 0.06, p<0.05) and 8 percentage points more likely to use plant leaves (marginal effect 0.08, SE 0.06, p<0.05), while the marginal effect of gender on maize powder use was smaller and not statistically significant (marginal effect −0.08, SE 0.06, p>0.10). Mobile phone ownership and semi−intensive farming systems also showed significant associations with some ingredients, and several education and income categories were significant at the 10% level, indicating that both socio−economic and institutional factors shape feed ingredient preferences alongside gender. Model fit statistics (pseudo−R² between 0.16 and 0.38; p−values for the likelihood ratio tests ≤0.01) indicate that the covariates jointly explain a meaningful share of variation in ingredient use.

Discussion

The study shows clear gender patterns in how fish feed ingredients are used across farming systems, rooted in labour roles, resource access, cultural norms, and household power relations. Men are more likely to use commercially purchased and unconventional ingredients in semi−intensive systems, which fits with their greater involvement in cash−intensive, physically demanding farm work and their stronger control over external inputs. Women make greater use of organic and locally available ingredients such as cassava waste, poultry waste, chicken manure and feathers, and kitchen leftovers, even though in some extensive pond systems, men report higher use of particular ingredients such as chicken manure and plant leaves. For plant leaves, overall use is higher among men, including in extensive systems, but women remain closely involved in handling and processing leaves within the household. Some ingredients appear to have become gendered niches: women’s exclusive use of JSRSM in extensive systems and men’s higher use of chicken manure in extensive ponds point to specific combinations of labour, knowledge, and responsibility. Both women and men use staple purchased ingredients such as bran, but women use them more in intensive systems, while men’s use is higher in semi−intensive systems, suggesting different positions in decisions about buying inputs and organising production.

Regional differences show how local availability and cultural practices interact with gender. Potato waste is more commonly used in Kakamega, while plant leaves are more widely used in Kisumu. Women usually obtain plant leaves through home processing, whereas men are more involved in buying bran from local suppliers, pointing to gendered access to markets and input supply chains. Within households, male relatives more often buy and use bran, while spouses and female relatives are more involved in using plant leaves, mirroring wider patterns of who does which work and who decides how money and inputs are used. Men also tend to manage the more capital−intensive or technically demanding systems, reinforcing their role in overseeing key resources and investment decisions.

The binary logit results show that these gendered patterns do not disappear once county, farming system, and household characteristics are accounted for. Gender remains a significant predictor of plant−leaf and poultry−waste use, but not maize powder, which means that men’s greater use of plant leaves and women’s greater use of poultry waste cannot be explained solely by where they farm or which system they use. This supports the view that feed−ingredient practices reflect gendered roles and responsibilities, and not just differences in local resources or technology. While women were more likely than men to report using maize powder in descriptive statistics, this association was not statistically significant in the adjusted models, suggesting that maize powder use is better explained by location and production system than by gender alone.

Theories of gendered resource control, mobility, and division of labour help to explain why particular ingredients are associated with men or women. Men’s higher use of plant leaves is consistent with their greater freedom of movement, stronger control over land and common spaces, and ability to spend time outside the home collecting materials from farms, roadsides, and communal land. Plant leaves are, therefore, not just “available biomass” but a resource that depends on who can move, who can decide to use land in that way, and who controls the tools and time needed to collect them. Women’s greater use of poultry waste, cassava waste, chicken manure and feathers, and other household organic materials reflects their responsibility for cooking, managing household waste, looking after backyard poultry, and cleaning. These tasks put women in daily contact with organic by−products that can be diverted into feed, so what appears as individual preference is better understood as a product of how labour is organised.

The farming−system results fit this picture, with semi−intensive earthen−pond systems relying on a broader mix of purchased ingredients, extensive systems drawing more heavily on locally available organic inputs, and cage systems using a narrower set of feed options. Chi−square tests confirm significant associations between feed use and farming intensity for several ingredients in earthen−pond systems, whereas cage farming showed no systematic differences in feed use across intensities.

Seen through a feminist economics lens, these patterns highlight that production decisions are made within households where time, assets, and decision−making power are unevenly shared, and where unpaid domestic labour underpins commercial production. Gender norms shape who owns or controls key assets, who takes on which kinds of work, and who decides how both food and feed ingredients are allocated. When staple foods such as cassava are diverted into feed, there can be trade−offs: women may lose access to low−cost food or have to spend more time finding replacements, with consequences for their workload and their say over household consumption. Women’s more limited access to purchased feeds, extension services, and formal technical knowledge, alongside their reliance on home−processed ingredients, reflects structural barriers and unequal information flows that have been documented in other agricultural sectors. The significant adjusted associations between gender and the use of plant leaves and poultry waste underline that feed practices are socially structured and constrained by what is materially available, rather than arising from neutral or purely technical choice.

These findings point to the need for gender−responsive policy and extension. Improving women’s access to feed inputs, technical advice, and market opportunities is likely to influence both their bargaining position and overall farm productivity. Extension approaches that explicitly recognise gendered roles, mobility constraints, and power relations, and that are delivered in ways that women can realistically attend and use, will be important if feed innovations are to benefit women as well as men. Monitoring frameworks that track changes in women’s time use, decision−making, and control over income can help ensure that interventions reduce, rather than intensify, existing inequalities.

Conclusion

This study shows that feed practices in Western Kenya’s aquaculture are deeply embedded in gendered labour, resource control, and household power relations. Women are more likely to draw on organic by−products from the household and farm, such as poultry waste, cassava waste, chicken manure and feathers, and kitchen leftovers, while men use plant leaves, rice bran and wheat bran, and a wider set of purchased or unconventional ingredients, especially in semi−intensive systems. These patterns vary across counties and farming systems, but they do not disappear once location, farming intensity, and other socio−economic factors are accounted for, as shown by the binary logit results for poultry waste and plant leaves.

Seen through a feminist economics lens, feed ingredients are not neutral technical inputs but part of a wider pattern in which women’s unpaid and often invisible work in cooking, waste management, and small livestock care underpins production, while men’s greater control over land, cash, and mobility shapes access to field−based and market−supplied feeds. When staple foods such as cassava are diverted into feed, there can be trade−offs for women’s access to affordable food and for their workload and influence over household consumption decisions. At the same time, the concentration of men in more capital−intensive and technically demanding systems reinforces gender gaps in control over key assets, information, and decision−making.

These dynamics have direct implications for policy and practice. Efforts to promote feed innovations must consider who currently handles and controls those inputs, whose labour will be affected, and who will gain or lose from changes in feed costs and availability. Gender−responsive extension and advisory services should explicitly target women as feed users and decision−makers, adapt training delivery to their time and mobility constraints, and link them to input suppliers and markets. Monitoring and evaluation frameworks that track changes in women’s time use, decision−making power, and control over income are essential if new feed technologies and value chains are to narrow, rather than widen, existing inequalities.

References

• 1

  AbwaoJ.MusaS.OndibaR.OgariZ. (2021). Socioeconomic dynamics and characterisation of land-based aquaculture in Western Kenya Aquatic Ecosystems. Available online at: https://www.researchgate.net/publication/357223260 (Accessed May 2024).

  • Google Scholar
  • View reference in article

• 2

  AdamP.SubianM.NjoguS. (2023). Abbreviated women’s empowerment in fisheries and aquaculture index (A-WEFI) study, Kenya. doi: 10.14217/9781848599437-en

  • CrossRef
  • Google Scholar
  • View reference in article

• 3

  AwuorF. (2021). The role of women in freshwater aquaculture development in Kenya. Aquat. Ecosyst. Health Manag.24 (1), 73–81. doi: 10.14321/aehm.024.01.11

  • CrossRef
  • Google Scholar
  • View reference in article

• 4

  BarclayK.MangubhaiS.LeducB.Donato-HuntC.MakhoulN.KinchJ.et al. (2021). “ Pacific handbook for gender equity and social inclusion in coastal fisheries and aquaculture,” in Pacific handbook for gender equity and social inclusion in coastal fisheries and aquaculture. Eds. BarclayK.MangubhaiS.LeducB.Donato-HuntC.MakhoulN.KinchJ. (Nouméa, New Caledonia: Pacific Community).

  • Google Scholar
  • View reference in article

• 5

  BarclayK. M.SatapornvanitA. N.SyddallV. M.WilliamsM. J. (2022). Tuna is women’s business too: Applying a gender lens to four cases in the Western and Central Pacific. Fish Fish.Nouméa, New Caledonia: Pacific Community (SPC)23, 584–600. doi: 10.1111/faf.12634

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 6

  BrugereC.BansalT.KruijssenF.WilliamsM. (2023). Humanizing aquaculture development: Putting social and human concerns at the center of future aquaculture development. J. World Aquacult Soc54, 482–526. doi: 10.1111/jwas.12959

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 7

  BrugereC.MsuyaF. E.JiddawiN.NyonjeB.MalyR. (2020). Can innovation empower? Reflections on introducing tubular nets to women seaweed farmers in Zanzibar. Gender Technol. Dev.24, 89–109. doi: 10.1080/09718524.2019.1695307

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  ChepkiruiM.OrinaP. S.OrinaT.GithukiaC.OchingoJ.AchokiJ. K. (2023). Review on the role of women along the aquaculture value chain in Kenya. Ann. Mar. Sci.7, 021–024. doi: 10.17352/ams.000034

  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  CraigS.HelfrichL. (2017). Understanding fish nutrition, feeds, and feeding. Publication 420-256. (Blacksburg, VA: Virginia Cooperative Extension, Virginia Tech). Available online at: https://vtechworks.lib.vt.edu/items/1dac621a-44d6-4f4e-9c5f-1371c05f2fdc (Accessed January 3, 2026).

  • Google Scholar
  • View reference in article

• 10

  DanielN. (2016). Neem seed cake (NSC) as fish feed ingredient: Opportunities and constraints. Int. J. Fisheries Aquat. Stud.4, 20–23.

  • Google Scholar
  • View reference in article

• 11

  EnyidiU. D.PirhonenJ.KettunenJ.VielmaJ. (2017). Effect of feed protein: Lipid ratio on growth parameters of African catfish Clarias gariepinus after fish meal substitution in the diet with bambaranut (Voandzeia subterranea) meal and soybean (Glycine max) meal. Fishes2, 1. doi: 10.3390/fishes2010001

  • CrossRef
  • Google Scholar
  • View reference in article

• 12

  FAO (2018). The State of World Fisheries and Aquaculture. Meeting the sustainable development goals (Rome, Italy: FAO).

  • Google Scholar
  • View reference in article

• 13

  FröcklinS.de la Torre-CastroM.LindströmL.JiddawiN. (2013). Fish traders as key actors in fisheries: Gender and adaptive management. AMBIO42, 951–962. doi: 10.1007/s13280-013-0451-1

  • CrossRef
  • Google Scholar
  • View reference in article

• 14

  GangulyS.DruczaK.EsayasB.BikketiE.YossaR.McDougallC. (2021). Affordable local ingredients for fish feeds in low-income contexts: A social and gender risk and opportunity analysis (Penang, Malaysia: CGIAR Research Program on Fish Agri-Food Systems). Working Paper: FISH-2021-22.

  • Google Scholar
  • View reference in article

• 15

  GhoshS. (2021). Imparting skill on formulated fish feed preparation to women's self-help groups in villages – an experience. Aquaculture Asia25 (3). (Bangkok: Network of Aquaculture Centres in Asia-Pacific (NACA)). Available online at: https://enaca.org/enclosure/?id=1178.

  • Google Scholar
  • View reference in article

• 16

  GithukiaC. M.DrexlerS.ObieroK. O.OtienoB.OdhiamboJ. A.ChesoliJ. W. (2020). Gender roles and constraints in the aquaculture value chain in western Kenya. Afr. J. Agric. Res.16, 732–455. doi: 10.5897/AJAR2020.14783

  • CrossRef
  • Google Scholar
  • View reference in article

• 17

  KantorP.KruijssenF. (2014). Informal fish retailing in rural Egypt: Opportunities to enhance income and work conditions for women and men. 1–25. Available online at: https://books.google.com/books?hl=en&lr=&id=YD6EBQAAQBAJ&oi=fnd&pg=PP3&dq=informal+competition+and+unregistred+firms+informal+OR+economy&ots=DCvIS7Dtox&sig=GACVs2K4cZPIexy9PFOYWNGUZhc.

  • Google Scholar
  • View reference in article

• 18

  Kenya National Bureau of Statistics (KNBS) (2021). “ Economic survey 2019,” in source africa. (Nairobi: Kenya National Bureau of Statistics (KNBS)). Available online at: http://dc.sourceafrica.net/documents/119074-Kenya-National-Bureau-of-Statistics-Economic.html.

  • Google Scholar
  • View reference in article

• 19

  KokB.MalcorpsW.TlustyM. F.EltholthM. M.AuchterlonieN. A.LittleD. C.et al. (2020). Fish as feed: Using economic allocation to quantify the Fish in: Fish-out ratio of major fed aquaculture species. Aquaculture528, 735474. doi: 10.1016/j.aquaculture.2020.735474

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 20

  KruijssenF.McdougallC. L.Van AsseldonkI. J. M. (2018). Gender and aquaculture value chains: A review of key issues and implications for research. Aquaculture493, 328–337. doi: 10.1016/j.aquaculture.2017.12.038

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 21

  KwasekK.Thorne-LymanA. L.PhillipsM. (2020). Can human nutrition be improved through better fish-feeding practices? A review paper. Crit. Rev. Food Sci. Nutr.60, 3822–3835. doi: 10.1080/10408398.2019.1708698

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 22

  LentiscoA.LeeR. (2015). A review of women's access to fish in small-scale fisheries (Rome, Italy: FAO). ISBN: ISBN 978-92-5-108857-9.

  • Google Scholar
  • View reference in article

• 23

  McClenachanL.MoultonA. (2022). Transitions from wild-caught fisheries to shellfish and seaweed aquaculture increase gender equity in Maine. Mar. Policy146, 105312. doi: 10.1016/j.marpol.2022.105312

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 24

  MungutiJ.Charo-KarisaH.OpiyoM. A.OgelloE. O.MarijaniE.NzayisengaL.et al. (2012). Nutritive value and availability of commonly used feed ingredients for farmed Nile tilapia (Oreochromis niloticus L.) and African catfish (Clarias gariepinus, Burchell) in Kenya, Rwanda and Tanzania. Afr. J. Food Agric. Nutr. Dev.12, 6136–6155. doi: 10.18697/ajfand.51.11085

  • CrossRef
  • Google Scholar
  • View reference in article

• 25

  MungutiJ. M.NairutiR.ItebaJ. O.ObieroK. O.KyuleD.OpiyoM. A.et al. (2022). Nile tilapia (Oreochromis niloticus Linnaeus, 1758) culture in Kenya: Emerging production technologies and socio-economic impacts on local livelihoods. Aquaculture, Fish and Fisheries2 (4), 265–276. doi: 10.1002/aff2.58

  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  MungutiJ.KirimiJ.ObieroK.OgelloE.SabwaJ.KyuleD.et al. (2021b). Critical aspects of aquafeed value chain in the Kenyan aquaculture sector- A review. Sustain. Agric. Res.10, 87. doi: 10.5539/sar.v10n2p87

  • CrossRef
  • Google Scholar
  • View reference in article

• 27

  MungutiJ. M.MugiraneaJ. K.OgelloE. O. (2014). An overview of the Kenyan aquaculture sector: Current status, challenges and opportunities for future development. Fish. Aquat. Sci.17, 1–11. doi: 10.5657/fas.2014.0001

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 28

  MungutiJ. M.ObieroK. O.ItebaJ. O.KirimiJ. G.KyuleD. N.OrinaP. S.et al. (2023). Role of multilateral development organisations, public and private investments in the aquaculture subsector in Kenya. Front. Sustain. Food Syst.7, 1208918. doi: 10.3389/fsufs.2023.1208918

  • CrossRef
  • Google Scholar
  • View reference in article

• 29

  MungutiJ. M.ObieroK. O.OrinaP. S.MwalumaJ.MireraD.OchiewoJ.et al. (2021a). “ State of aquaculture in Kenya 2021: towards nutrition-sensitive fish food production systems,” in Techplus media house, nairobi, Kenya. NJAS - wageningen journal of life sciences. Eds MungutiJ. M.ObieroK. O.OrinaP. S.MwalumaJ.MireraD.OchiewoJ. (Nairobi: Techplus Media House), 60 (22), 3822–3835

  • Google Scholar
  • View reference in article

• 30

  MungutiJ.OdameH.KirimiJ.ObieroK.OgelloE.LitiD. (2021c). Fish feeds and feed management practices in the Kenyan aquaculture sector: Challenges and opportunities. Aquat. Ecosystem Health Manage.24, 82–89. doi: 10.14321/aehm.024.01.12

  • CrossRef
  • Google Scholar
  • View reference in article

• 31

  ObieroK. O.WaidbacherH.NyawandaB. O.MungutiJ. M.Julius Otieno ManyalaJ.AraraB. K. (2019). Predicting uptake of aquaculture technologies among smallholder fish farmers in Kenya. Aquacult. Int.27, 1689–1707. doi: 10.1007/s10499-019-00423-0

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  OCDC International Cooperative Research Group (2019). Initial findings: what difference do cooperatives make? Kenya county study. (Washington, DC: U.S. Overseas Cooperative Development Council (OCDC)). Available online at: https://ncbaclusa.coop/content/uploads/2021/05/OCDC_KenyaReport_12.2020_LowRes.pdf (Accessed January 8, 2026).

  • Google Scholar
  • View reference in article

• 33

  OpiyoM. A.MarijaniE.MuendoP.OdedeR.LeschenW.Charo-KarisaH. (2018). A review of aquaculture production and health management practices of farmed fish in Kenya. Int. J. Vet. Sci. Med.6 (2), 141–148. doi: 10.1016/j.ijvsm.2018.07.001

  • CrossRef
  • Google Scholar
  • View reference in article

• 34

  SinghP.PaulB. N.GiriS. S. (2018). Potentiality of new feed ingredients for aquaculture: A review. Agric. Rev.39, 282–291. doi: 10.18805/ag.r-1819

  • CrossRef
  • Google Scholar
  • View reference in article

• 35

  Torre-CastroM. D. L. (2019). Inclusive management through gender consideration in small-scale fisheries: The why and the how. Front. Mar. Sci.6, 156. doi: 10.3389/fmars.2019.00156

  • CrossRef
  • Google Scholar
  • View reference in article

• 36

  USAID (2017). “ Integrating gender equality and female empowerment in USAID’s program cycle (No. ADS chapter 205),” in United states: US agency for international development. (Washington, D.C.: U.S. Agency for International Development (USAID)). Available online at: https://www.usaid.gov/sites/default/files/documents/1870/205.pdf.

  • Google Scholar
  • View reference in article

• 37

  WeeratungeN.SnyderK. A.SzeC. P. (2010). Gleaner, fisher, trader, processor: understanding gendered employment in fisheries and aquaculture. Fish Fish.11, 405–420. doi: 10.1111/j.1467-2979.2010.00368.x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 38

  WeiminM.MengqingL. (2007). “ Analysis of feeds and fertilisers for sustainable aquaculture development in China,” in Study and analysis of feeds and fertilisers for sustainable aquaculture development. FAO Fisheries Technical Paper. No. 497. Eds. HasanM. R.HechtT.De SilvaS. S.TaconA. G. J. ( FAO, Rome), 141–190.

  • Google Scholar
  • View reference in article

• 39

  World Bank (2023). Aquaculture production (metric tons) | Data. Available online at: https://data.worldbank.org/indicator/ER.FSH.AQUA.MT (Accessed May 2024).

  • Google Scholar
  • View reference in article

• 40

  WorldFish (2020). ‘Bringing a change’ in women’s equitable involvement in aquaculture. (Penang, Malaysia: WorldFish / CGIAR Research Program on Fish Agri-Food Systems (FISH)). Available online at: https://worldfishcenter.org/story/bringing-change-womens-equitable-involvement-aquaculture.

  • Google Scholar
  • View reference in article