Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Food policy

Men in the food chain: Reasons for similarities and dissimilarities between Asian and African households

Margareta Wandel


The nutritional well-being of households of developing nations depends on the work and decisions around numerous food-related activities that are carried out by household members. It often relies on the cooperation among the members, particularly between spouses. Numerous review articles have been written about women's role in food-related activities. It is also important to shed more light on men's role with regard to food and nutrition in Asian and African households. This study is based on the existing literature on gender relations in activities concerning food.



During the last ten to fifteen years, development research has given us greater knowledge and understanding of and insight into changes in women's life and working status in developing nations. The research was initiated out of concern that development efforts often had a gender bias favouring men for example, with regard to agricultural services, credit, and education. This led to marginalization of women, with detrimental effects not only for themselves but also for the food and nutrition status of their households.

Men have often been viewed as winners and women as losers in the process of socio-economic change in developing countries. However, men's changing circumstances have not received attention similar to that given those of women. Virtually all studies on gender roles with regard to food in developing countries have focused heavily on women; data pertaining to men are few and scattered.

In most societies the husband (if he is present) is seen as the head of the household and the one with responsibility for the welfare of the family. What this means in terms of food and nutrition for household members may vary a great deal in different societies. To clarify men's role in different food-related operations, the concept of the food chain has been employed. This concept denotes the sequences of events that take place around food, from the time it enters the household by production, purchase, and other forms of procurement until it is consumed by the household members (fig. 1). The concept of the food chain has so far been used as an organizing principle in studies of women's competing responsibilities and work [1, 2] - for example, to identify time constraints affecting child care. However, it is also important to understand the role men play in the food chain and the possibilities for cooperation between men and women.

In societies where women have a heavy work load, if men took on some of the work it could smooth out seasonal work peaks and time shortages, giving women more time for other activities and thereby a potential for better child care and feeding. It could also contribute to more timely cultivation schedules and a more diverse cropping pattern, with positive effects for food and nutrition status.


Women's and men's expectations of a "good spouse"

As a part of field studies in Sri Lanka and Tanzania, women and men were questioned as to what they expected of a "good" husband or wife. Food-related behaviour naturally played an important part in such expectations among poor peasants, a large part of whose daily activities and worries were centred around food.

In Sri Lanka the women saw providing food as men's foremost role (table 1). They also expected men to help with child care. The Tanzanian women were concerned with the threat of being beaten by their husbands. They expected men to work hard in the maize fields, which contributes to both food and cash in the household, but they did not mention providing food, whereas Sri Lankan women saw this as the main job for men.

The Tanzanian men were mostly concerned that the women should work hard in the field, whereas the Sri Lankan men were concerned with women's ability to give good child care and to cook. Cooking ability was also mentioned by the men in Tanzania but with lower priority.

These are just two examples of role expectations regarding food among a few men and women in two small communities in Africa and Asia. They illustrate some important aspects of men's and women's relationship to the work and responsibilities the food chain.

FIG. 1. The food chain (Source: Ref. 2)

TABLE 1. Qualities expected of a "good" husband, as expressed by Sri Lankan and Tanzanian wives

Sri Lanka (27 respondents) Tanzania (24 respondents)
Provides food for family Does not beat wife
Provides for children's Works hard in the field
education Looks after family, buys
Gives love and affection, clothes and shoes for
honesty wife and children
Gives help in child care Does not drink too much
Is able to put up a new Is not a thief
house Gives wife money when
Does not dunk and she needs it
gamble Helps look after children
Does not fight  

The first three qualities mentioned by each respondent are included in the compilation. They are arranged according to the frequency with which they were mentioned.


Men's allocation of time

Many studies from Africa have shown that men work in the field less or about the same amount of time as women, but the total time men devote to food-related work is usually substantially less than for women because they usually contribute very little to food processing, cooking, and feeding.

In a detailed study from Zambia [3], men's and women's time use was recorded five days a week over a whole year. The total time the men spent in work and other duties was around five hours per day, but the women spent substantially more (table 2). It is of interest that there was quite a large difference in working time between men of different ages; those under 40 years old contributed most to agricultural work. Men both over and under age 40 spent very little time on food preparation and other food-related activities.

In places where the time devoted to food crops can be separated from that for cash crops, it is often found that men spend relatively little time on the former and more on the latter (table 3) [4]. Thus it is important to obtain information on what men contribute to the actual food supply. Data from the Ivory Coast showed men contributing to the kitchen food supply mainly with respect to sugar, meat, milk, eggs, and drinks, but they also made a contribution to staple foods [4].

In Tanzania [5], where the main food crop, maize, is also a cash crop, men had a shorter working day in the field than women on the average and worked fewer days. This was also the case in households where the main economic activity was farming. In this area it was not land shortage but labour shortage that was the major constraint in food production. There was a large variation among the men with regard to the time devoted to work.

We tried to find out whether, when men worked hard, women could work less in the field and therefore have more time for other activities in the food chain. Our findings in Tanzania indicated that the relationship between men's and women's work was rather complex (table 4): In the households where men put in a substantial amount of labour in the field, the women also had longer working days. However, in households where men did not contribute anything to agricultural work, the women had a higher work load than in the households where men did at least do some work.

TABLE 2. Average time spent per day on different activities, by sex and age group (under or over 40 years), Zambia



Time (hours)








Work and other duties        
food preparation and related activities










child care





agricultural activities





animal husbandry





model house construction





maintenance and repair work





meetings and discussions





funeral attendance





medical care










Other activities        





drinking beer





personal hygiene





going out of the village


0 33








Recorded leisure










Source: Ref. 3.

TABLE 3. Daily average time devoted to cotton, food crops, and other crops over the year, Central African Republic



























Food crops









Other crops


















Source: Ref. 4.

In a study in Sri Lanka we found that men did most of the work in paddy cultivation with the help of agricultural labourers, using tractors or cattle [6]. Women were busy preparing food for the labourers and carrying food and water to the fields [7]. In addition, they did some vegetable cultivation.

A similar pattern of work was found in an area with rain-fed rice in the Philippines [8]. Men spent on the average three times as much time in cultivation and farm wage work as their wives, who cultivated crops other than rice, raised pigs, and did household chores. Much of paddy cultivation was done by hired labourers. This study indicated that men's work changed according to the life cycle of the family, in the same direction as that indicated in Zambia. Men's labour input in agriculture peaked when the family was in the child-bearing phase (fig. 2). When the children grew up, it was especially the men who relaxed while the children took over some of the duties. The households were also evaluated according to social status. Men of higher status had more land and spent time organizing the hired labourers. Their work was less heavy, but their total working time was about the same as that of less well-to-do farmers.

Rice farming is predominantly men's work in Asia. Irrigated rice cultivation is an ancient farming system in some parts of Asia, where the so-called hydraulic societies were built around the technology for distribution of water for farming. As described by anthropologist E. R. Leach [9], these were highly hierarchical and rigid systems. Farmers who cultivate with water from large irrigation systems still have to obey rules and strictly follow the agricultural calendar, which allows a relatively short time for each operation. Even in this predominantly male work of cultivation, women perform a great deal of the labour, mostly in the tasks of transplanting, weeding, and harvesting, but much more of the female labour in this system is hired labour than in Africa [10].

TABLE 4. Relationship between time spent on field work by men and women, Tanzania


Time (minutes)b


No. of house holds



















Source: Ref. 5.

a. Households grouped on the basis of the amount of time spent on field work by men.
b. Average time per day for only those days on which some field work was done by members of either sex.

Analysis of variance main effect after controlling for the effect of village: F = 8.1, F probability c too. Group 1 different from group 2 ( p < .05). Group 4 different from groups 2 and 3 (p < .05 by test of least-significant differences).

FIG. 2. Work allocation (averge hours per household per year, over three years) among husband. wife, children, and hired labour, by phase of family life cycle Philippines (Source: Ref. 8)

The literature gives examples also of Asian societies where women contribute as much as or more than men to agriculture and to household income. A study from Nepal [4], conducted in eight villages representing all ecological zones, showed that men and women spent about equal time on agricultural activities, but when all the food-chain activities were taken into consideration, the time spent by men was considerably less, much as in the African examples. Men contributed more in outside cash earnings than women, but that money made up only a small part of the total income. When outside earnings and home production were considered together, men contributed 44%, women 50%, and children 6%. In many Asian societies the contribution of different household members to the income is related to the economic standing of the household. In lower-income households women's income is more important absolutely and proportionately than men's [11].


Factors governing men's participation in food production

A number of researchers have discussed sources of variability in male labour inputs to agriculture. For example, although female farming is more common in the so-called extensive farming systems, where the hoe is the principal tool, agricultural intensification with the use of a plough or cattle increases male farming [12]. An alternative hypothesis concerns the effect of the climate, such as the number of dry months [13]. In line with this is the postulation that a narrow "seasonal window" puts a premium on the labour of young men; that is, the time pressure of soil preparation is so great that young men's physical strength advantage should make them the best candidates for farm labour.

According to ILO estimates [14], the male share of the total agricultural labour force is somewhat higher in Asia than in sub-Saharan Africa. However, within these regions there is substantial diversity, as was also shown by the cited studies. The male share in sub-Saharan Africa ranges from around 90% in Angola and Sudan to under 50% in Botswana, the Central African Republic, and Zaire [14]. In Asia, the male share is estimated at 74% in the south, 63% in the south-east, and 54% in the east. The highest figure was from Pakistan (about 90%).

The ILO data from 95 less-developed countries [14] were used in multivariate analyses of the sources of variation in male and female participation in the agricultural labour force [15]. It was postulated that men's role in farming would increase if there were unequal distribution of land with a high share of large land holdings, more wage labourers were employed in cultivation, and more cash crops were cultivated, whereas women's role would increase with male out-migration. These factors explained 57% of the variation. However, regional categories (North Africa, sub-Saharan Africa, Asia, Central and South America, the Caribbean) were more powerful predictors of the division of labour, explaining 71% of the total variation. When the regional variables were included in the model, the structural variables added virtually nothing to the prediction.

FIG. 3. Relationship between men's and women's work in specific farm operations-Nigeria. (1) Clearing land, (2) preparing seed-beds, (3) sowing/transplanting, (4) weeding, (5) staking, (6) fencing, (7) scaring off birds, (8) applying fertilizer, (9) harvesting (Adapted from ref. 16)


Interchange of food-chain work and responsibilities by sex

Studies from many countries report that activities in food production often are highly rigidly allocated by sex. This was illustrated very clearly in Nigeria, in an area where cassava, yams, and rice were the main crops [16]. Such operations as sowing, transplanting, harvesting, and threshing were considered preferably the work of women, but they could be performed by men. However, for certain tasks the possibility of such substitution was constrained by culturally ascribed roles in the farming system. Women rarely replaced men in preparing seed beds, and men rarely replaced women in weeding (fig. 3). According to the men, weeding was regarded as women's work because it requires diligence and a "soft touch" that they did not have. Since weeding is very time-consuming, this division of labour had great implications for women's time use.

Time-allocation studies from African countries have shown that men's participation in parts of the food chain other than food production was very low [3,17]. One exception was milling, which most often was taken over by men if it was done with a machine-driven mill. Men on the average spent only small amounts of time in such activities as food preparation and feeding.

In the Philippines [8], although there was a clear division of tasks by sex, it was not generally very rigid. Women could handle a plough if necessary, and a man could do the cooking if his wife was off marketing her products. In Sri Lanka, however, activities around rice cultivation were rigid [7]. Men were the farm managers. A few women were rice cultivators, but they did not have a man in the household. In other parts of the food chain the division of labour was not as strict. Women expected help from men with child care, and most of them got it. When asked if they knew how to cook and if they ever were involved in this activity, men living in nuclear families commonly answered they did know how to cook and this allowed them to manage in times of pregnancy and sickness. Men were never in charge of the cooking when women were at home and well, but they often helped prepare the food for cooking and helped serve food when guests were present.

Men's behaviour in relation to food during women's pregnancy is particularly interesting. In Sri Lanka it is culturally accepted and expected that a woman may crave different kinds of foods during the first part of pregnancy, demands that must be attended to by her husband and kinsfolk. This is called cola duka, two words meaning "craving" or "perverse appetite" and "suffering" respectively. The anthropologist Obeysekera, who studied this phenomenon thoroughly, wrote:

During dola-duka the domestic situation is such that the husband is forced to draw water, bring firewood, look after the children and even cook on occasion. In normal living, women are expected to obey and serve their husbands; but here the roles are inverted and the husband is compelled by custom to serve the wife. He has to yield unto her "perverse" demands, often walking long miles and incurring great expenditures and inconvenience in order to bring her the required foods. In normal life it would be demeaning for the husband to act in this fashion, but in dola-duka the otherwise adhered to division of labour between the sexes is reversed. Dola-duka gives the village women a much needed holiday from life [18].

Obeysekara's study was carried out about 30 years ago, and much has changed since that time. However, during my field work in 1982 this cultural tradition was still alive: men engaged in household work and did their utmost to acquire the food their wives demanded during pregnancy. Even if these foods were not always the most nutritious, it can be assumed that this practice, with a focus on food and rest, was beneficial for the expectant mother and her child.

The question of the interchangeability of men's and women's roles in the food chain is connected to prestige or loss of prestige, which may occur if men engage in activities that are usually considered womanly. In many cases it seems easier for women to engage in men's work than vice versa, which may entail more loss of prestige. Substitution is typically lowest at the tail end of the food chain, household work. It may be higher in some parts of Asia (table 5), although not enough data are available to generalize for Asia as a whole on this issue.

TABLE 5. Time spent on household work by women and men


Hours per day






Upper Volta (Burkina Faso)















Source: Ref. 17.


Men's role in decisions about food

An important issue that can have large consequences for a household's food and nutrition is that of decisions concerning the sale of food crops. The decision may involve the sale of one type of food to be able to buy other foods or non-food goods. Men and women may have different opinions in line with their responsibilities for the household. Studies of decision making within the household are very difficult and cumbersome, and few researchers have gone into this issue.

A study from Nepal [19] recorded large differences with regard to who decided what should be sold. On an aggregated basis, women had this responsibility more than men. Even in some communities where men usually made most other decisions regarding cultivation, they did not typically make this one.

In Tanzania, according to both women and men, women seldom made this decision alone [20]. In about half the households the decision was made by the husband alone, and in the other half it was made jointly. When disagreement arose, it was usually because men wanted to sell the food and women wanted to keep it for household consumption. However, in many African societies women have more say about the sale of foods. Those who have their own fields and cultivate their own crops often decide whether and when to sell.

Some studies from Asia as well as Africa have observed that the income earned by men is less likely to be used to buy food for the family than that earned by women. The distribution of food among family members may also be different depending on whether the father or the mother earns the income. Data from three rural areas in the Philippines were used to analyse the intrahousehold distribution of food in relation to individuals' recommended daily allowances for energy [21]. In households in which the wife's estimated wage was higher, both she and her children received a relatively greater share of the available energy. When the husband's wage rose, the intrahousehold allocation of food energy to him and his wife increased but not to the children.

Mothers' income has been found to have a greater positive effect on child nutrition status than fathers' in both Asia (India) [22] and Africa (Ghana) [23]. The nutrition status of Kenyan children was better in households headed by women [24]. This was not the case in Zambia, where the nutrition status of preschool children in male-headed households, in which both husband and wife were active in agricultural cultivation, was better than in female-headed households, where only the wife was cultivating (S. K. Kumar, personal communication, 1985). However, the female-headed households had a significantly higher level of child nutrition status at any given income level.


Men's status in intrahousehold food allocation

Factors that govern intrahousehold food allocation in rural households can he summarized into three principles that may operate either alone or in concert. According to the first principle, those who contribute to the household income have preferential access to the food, while the second principle is to give preference according to status, and the third according to needs. Men may be at an advantage in many ways. In societies where they are considered the main breadwinners, they may get preference according to the first principle. The first and second principles are closely linked, since men often get preference according to both. But, whereas the first gives preference to those of income-earning age, the second also may give preference to boys over girls. These principles of food distribution are most important in poor families, where preferential access to food may imply life or death.

A number of studies from Asia give evidence of sex bias in intrahousehold food distribution. According to a large sample surveyed in Bangladesh, energy intake as a proportion of the requirement was higher for males, especially those 04 years old and those over 45 years [25]. As a result, girls from birth to 5 years of age were more likely to be moderately or severely malnourished (14% severely malnourished, versus 5% for boys). Food discrimination favoured males in northern India and the Philippines [26]. However, studies in southern India and Sri Lanka did not find signs of such discrimination [27, 28]. Sex discrimination does not seem to apply for adults in India, but men may receive a greater share of the household food in the Philippines [27].

One must be cautious when interpreting results of studies of food consumption that do not also have data on malnutrition unless one can be sure that the data include all the food eaten between meals. Though children may be discriminated against at regular meals, the apparent bias may disappear when all the foods eaten are considered [29, 30].

A study from India showed that sex differences in child survival were associated with a low education level among fathers as well as with indicators of poverty, such as landlessness, and also a high sex gap in expected lifetime earnings [31]. Favouring of boys in intrahousehold food distribution was not seen in Africa [27, 32]. This may reflect the fact that women's role as procurers of food is more important there than in countries such as Bangladesh and northern India.

The findings from a number of studies on intrahousehold distribution of food were summarized as follows:

It is rare to find food discrimination against adult women in intrafamily food allocation; slightly more common to find it against children, and most common to find it against girls aged zero to four; though even this appears to be typical only in Bangladesh and northern India [27].

It has been suggested that this may often represent a desperately poor family's last resort in trying to pull through by feeding the members who are most likely to earn incomes, either currently or when they are older [27]. However, this focus on the direction of causality from the market to the household has been criticized, because it tends to ignore the underlying factors that may be important for both the sexual division of labour and the allocation of food. Thus, it is necessary to focus attention on the cultural factors and power relations of the household that permit this sexual bias to occur.


Marginalization of the role of men and consequences for food availability

Closely related to work and prestige are social changes that lead to the marginalization of men's role in the food chain.

For example, according to the Department of Community Health and the Department of Psychiatry in Nairobi, the Kisii district of Kenya has one of the highest levels of serious alcohol abuse in the country and also a high frequency of mental disease among men [33]. Men from the district began labour migration very early, leaving the women behind to do almost all the cultivation, from breaking the ground with hoes to harvesting [34]. Today, jobs outside the district are scarce and there is less migration, but it is no longer possible for the men to go back to the activities they had before migration-keeping cattle and milking-because the land is now used for other purposes. Most men do not see food cultivation as a way to contribute to the household's welfare because this is not considered a man's job. The repeated arguments were, "We were never shamba [food-crop field] people," and "If a man goes to the shamba, he is not a real man." The men who ventured to break this principle were excluded from men's circles. But, even though women were viewed as the food producers, it was men's responsibility to see to the welfare of the family. This was difficult with the sporadic incomes many of them had. Their real dilemma could therefore be summed up as two questions: How could they be respected in their society if they did not fulfil their male obligations toward the food needs of the household? and, Could they be respected if they did shamba work? This dilemma was seen as one of the factors contributing to increasing alcoholism and mental diseases [33].

In the Rukwa area in Tanzania there are also indications that men have been marginalized and that some of the activities to which their identity and status were tied are no longer valid [20]. At one time men cut trees for slash-and-burn cultivation and went hunting. Nowadays, due to deforestation, few trees are left to cut and the game has disappeared. Some men have managed to adapt by being employed in new areas, such as weeding, but others have had more difficulty in doing so. In this society the consumption of alcohol (beer) has increased. Traditionally, beer was used for celebrations; now visits to the beer halls are a regular form of social gathering.

Men's increasing alcohol consumption in East African countries has also been linked to changes in men's role in other ways [35]. It is almost always women who brew the local beer, and this is often an important source of income for them. Also, the children of beer brewers may be better off nutritionally, because of both their mothers' extra income and the energy-rich malt porridge that is often made in connection with brewing. However, the large amounts of money men spend on beer could be used to buy food for their families. Thus, while men's excessive drinking may solve a food problem in the households where the beer is brewed, it creates a problem in the households whose men indulge in drinking. In times of scarcity, it can also be seen as a system that siphons calories and nutrients to the male members of the community, because food such as grilled meat is eaten with the beer. Thus, visits to the beer halls may have serious implications for the distribution of food and resources both between and within households.

Social changes that may lead to the marginalization of men were vividly described in an area in Sri Lanka where a development project enabled women to take over as breadwinners, which had previously been socially and culturally the role of men [36]. It was suggested that, in the face of a decline in self-image, men were inclined to indulge heavily in behavioural practices that were seen as strongly masculine, such as drinking liquor, with resultant conflicts between husbands and wives. In extreme cases the wives had the economic power to ask their husbands to leave home.



These studies from Africa and Asia show that the contribution of men to the food chain can vary quite a lot, both within and between different countries and regions. However, certain patterns seem to emerge. Let us consider the first step in the food chain, food production. In many societies in sub-Saharan Africa women have the main responsibility for providing the staple food, whereas in Asia this responsibility falls to the men. Even in Asia, however, many women are active in food production, either on the family farm or as hired labourers. Their contribution to food procurement is typically higher in poor households, in which men are not able to fulfil these responsibilities [11].

The notion that men are responsible for the welfare of the family exists in both African and Asian societies. We will discuss this concept only in relation to food. In Asia it most often means that the man is expected to be the main provider or producer of food. This is less so in Africa, where men may be providers of the land and managers of the farms on which the women produce food for the family. Even though the men in this case also are responsible for providing some of the food, such as meat and fish, they are more distant from the day-to-day provision of the most important part of the diet, the staple food.

Factors that are postulated to increase male participation in farming-such as agricultural intensification with the use of technology, unequal distribution of land, more people being landless, more being engaged as wage labourers, and more cultivation of cash crops [12, 13, 15]-are more preponderant in Asia than in Africa [11] and thus may contribute to the differences between these regions. However, evidence that region was a major predictor of the extent of men's and women's contribution to the total agricultural labour force [14] directs attention to other, more culture-specific factors. Important in this regard is the question of what is considered masculine behaviour and what activities men can involve themselves in without loss of image. This has a bearing on the possibilities for the interchange of labour and for cooperation between men and women.

The possibility of interchanging men's and women's roles in food production varies quite a lot according to the tasks involved and in different societies. Agricultural change may require a transformation in the prevailing division of labour between the sexes. A rigid system may lead to a work overload for one sex and marginalization for the other. This is particularly unfortunate when women become overloaded with work while men become marginalized [20, 33. 35, 36], since women have so many other responsibilities in the food chain.

There may also be large differences in food handling, food preparation, child care, and feeding, in which men's participation is usually low. In Sri Lanka and the Philippines men participate in these activities, but not when women are present [7, 8]. In African countries men rarely participate in any activities that can be called household work [3, 17].

It is interesting to relate African women's important role as food providers to the fact that women and girls are not discriminated against in food distribution [27, 32]. Women's income is more likely to be used for food and for greater equity in intrahousehold food distribution, which demonstrates women's important role on behalf of household nutrition status. However, these positive effects of women's work have to be weighed against the potentially negative effects on their own health and on other food-related responsibilities.

Some authors have warned against the overuse of female labour, arguing that, with increased education, young women would be less willing to accept the demands put on them. Therefore, "if the level of farm output is to be sustained and improved, the need to reduce the dominance and inflexibility of female labour for specific farm operations cannot be overemphasized" [16]. The introduction of labor-saving techniques has resulted in some improvements. However, in societies where men's labour potential is underused at the same time as women are overworked, it would seem inexpedient not to pay attention to this resource.

In efforts aimed at improving household food and nutrition status, it is important to pay attention not only to the existing division of labour and responsibilities between men and women but also to the possibilities for increased cooperation between the sexes in food-chain work. The important thing is to make sure that neither women nor men are marginalized, since this may lead to unexpected detrimental effects on the food and nutrition status of numerous individuals in the community.



1. Tinker 1. New technologies for food chain activities: the imperative of equity for women. Washington, DC: Office of Women in Development, AID, 1979.

2. Holmboe-Ottesen G. Mascarenhas O. Wandel M. Women's role in food chain activities and the implications for nutrition. ACC/SCN State-of-the-Art Series. Nutrition Policy Discussion Paper 4. Geneva: WHO, 1989.

3. Skjønsberg E. Change in an African village: Kefa speaks. Hartford, Conn, USA: Kumarian Press, 1989.

4. Berio A-J. The analysis of time allocation and activity patterns in nutrition and rural development planning. Food Nutr Bull 1984;6(1):53-68.

5. Holmboe-Ottesen G. Wandel M. Men's contribution to the food and nutritional situation in the Tanzanian household. Ecol Food Nutr 1991;26(1):83-96.

6. Wandel M, Holmboe-Ottesen G. Household food consumption and seasonal variations in food availability in Sri Lanka. Ecol Food Nutr 1989;22:169-82.

7. Wandel M, Holmboe-Ottesen G. Women as nutrition mediators: a case study from Sri Lanka. Ecol Food Nutr 1988;21:117-30.

8. Res L. Changing labor patterns of women in rice farm households: a rainfed rice village, Iloilo province, Philippines. In: Women in rice farming: proceedings of a conference on women in rice farming systems. Manila: International Rice Research Institute, 1985:91-117.

9. Leach ER. Hydraulic society in Ceylon. Past Pres 1959; 15 :2-26.

10. Safilios-Rotschild C. The role of women in modernizing agricultural systems. Paper presented to WID/AID, USAID, Washington DC, 1981.

11. Safilios-Rotschild C. The role of the family: a neglected aspect of poverty. World Bank Staff Working Paper 403. Washington DC: World Bank, 1980.

12. Boserup E. Women's role in economic development. London: George Allen & Unwin, 1970.

13. White DR, Burton ML, Dow MM. Sexual division of labour in African agriculture: a network autocorrelation analysis. Am Anthropol 1981;83:824-49.

14. International Labour Office. Labour force estimates and projections 1950-2000 2nd ed. Geneva: ILO, 1977.

15. Dixon R. Land, labour, and the sex composition of the agricultural labour force: an international comparison. Dev Change 1983;14:347-72.

16. Okorji EC, Obienchina COB. Implications of overdependence on women's labour for food production in a traditional farming system of Anambra State, Nigeria. Cult Agric 1990;40:7-11.

17. Ellis F. Peasant economics, farm households and agrarian development. Cambridge, UK: Cambridge University Press, 1990.

18. Obeysekera G. Pregnancy cravings (dola duka) in relation to social structure and personality in a Sinhalese village. Am Anthropol 1963;65:323-42.

19. Acharya M, Bennet L. Rural women of Nepal: an aggregate analysis and summary of eight village studies. The status of women in Nepal: Vol 2. Field studies. Kathmandu, Nepal: Centre for Economic Development and Administration, Tribuvan University, 1981.

20. Holmboe-Ottesen G. Wandel M. "Wife, today I only had money for pombe": gender and food-women's bargaining power and agricultural change in a Tanzanian community. In: Stølen KA, Vaa M, eds. Gender and change in developing countries. Oxford, UK: Oxford University Press, 1991;93-119.

21. Senauer B. Garcia M, Jacinto E. Determinants of the intrahousehold allocation of food in the rural Philippines. Am J Agric Econ 1988;70(1):170-80.

22. Kumar SK. The role of the household economy in child nutrition at low incomes: a case study in Kerala. Occasional Paper no. 95. Ithaca, NY, USA: Department of Agricultural Economics, Cornell University, 1977.

23. Tripp R. Farmers and traders: some economic determinants of nutritional status in northern Ghana. Food Nutr 1982;8(1):3-11.

24. Kennedy E, Cogill B. Income and nutritional effects of the commercialization of agriculture in south-western Kenya. Research Report 63. Washington DC: International Food Policy Research Institute, 1987.

25. Chen LC, Huq E, D'Souza S. Sex bias in the family allocation of food and health care in rural Bangladesh. Pop Dev Rev 1981;7(1):55-70.

26. Carloni A. Sex disparities in the distribution of food within rural households. Food Nutr 1981;7(1):3-12.

27. Lipton M. Variable access to food. In: Gittinger JP, Leslie J. Hoisington C, eds. Food policy: integrating supply, distribution and consumption. EDI Series in Economic Development. Baltimore, Md, USA: Johns Hopkins University Press, 1987:385-92.

28. Holmboe-Ottesen G. Wandel M, Oshaug A. Nutritional evaluation of an agricultural development project in southern Sri Lanka. Food Nutr Bull 1989;11(3): 47-56.

29. Harbert L, Scandizzo PL. Food distribution and nutrition intervention: the case of Chile. World Bank Staff Working Paper 512. Washington DC: World Bank, 1982.

30. Nieves I. Maternal food management, intrahousehold food distribution in Guatemala: culture and behaviour. In: Young KW, Cha LY, Yull LK, Soon JJ, He KS, eds. Proceedings of the 14th International Congress of Nutrition, Seoul, Korea, 1989:746-8.

31. Rosenzweig M, Schultz T. Market opportunity, genetic endowments and the intrafamily distribution of resources: child survival in rural India. Center Discussion Paper 347. New Haven, Conn, USA: Yale University Economic Growth Center, 1980.

32. Schofield S. Seasonal factors affecting nutrition in different age groups and especially preschool children. J Dev Stud 1974;11(1):22-40.

33. Silberschmidt M. Rethinking men and gender relations: an investigation of men, their changing roles within the household and the implications for gender relations in Kisii district, Kenya. Copenhagen: Centre for Development Studies, 1991.

34. Le Vine R. Le Vine B. Nyansongo: a Gusii community in Kenya. New York: John Wiley & Sons, 1966.

35. Raikes A. Women's health in East Africa. Soc Sci Med 1989;28(5):447-59.

36. Casinader R. Fernando S. Gamage K. Women's issues and men's roles: Sri Lankan village experience. In: Momsen JH, Townsend J. eds. Geography of gender in the third world. Albany, NY, USA: State University of New York Press, 1987:309-22.

Contents - Previous - Next