Gender-responsive social science research has, as a starting point, the acceptance of two basic assumptions One, sex differences are not gender differences. Two, whatever differences and similarities exist between genders, social, political, and economic subordination is a cultural rather than a biological phenomenon and has to be examined on that basis. Gender-responsive research must be prepared to view gender (the social -cultural construct/definition of sex differentiation) as a critical variable in social organisation. Such research should then have on its agenda the production and presentation of data that tell her story as well as they tell his story. Gender-responsive research must go beyond providing evidence of gender equality or inequality to providing solutions and directions towards a less hierarchical society. Gender-responsive social science research must be seen in its ability to provide a better understanding of the persistence of underdevelopment in the region, whose major characteristic, among others, is the feminisation of poverty. Gender-responsive research must be seen to ask the right questions. What, for example, keeps women marginal when they play a most central role in the production and reproduction of social life? Do we really understand the marginality of the central actors? What are the linkages between gender hierarchies, patriarchal ideologies, the "marginality of the centre", and ultimately underdevelopment and backwardness?
The main point of this paper is that we do not have data that can explain these phenomena because traditional research (whether by men or women) has tended to ignore the fact that gender is a critical variable in social life. Because social science research has tended to be biased, its research findings have not reflected the empirical realities of gender.
It is my view that not only do we need gender-responsive social research but researchers themselves will have to go through a conceptual, ideological, and ethical reorientation in order to utilise these tools effectively. An analysis of the history of research processes clearly reveals that the ideology and background of researchers has affected not only the search process but also the data themselves. The existing sexism in social science research is directly related to the cultural ideologies of gender hierarchies which have greatly influenced the researchers' perceptions, conceptualisation, collection, and interpretation of the data.
In the paper some of the biases and related distortions of traditional research in the presentation of gender experiences will be discussed. The achievements of the research activities responding to earlier biases will be outlined. Finally, specific areas of concern for a gender-responsive research approach are identified.
One of the major achievements of the feminist movement in the academic world is that it has succeeded in casting doubts on the possibility of total objectivity in the social sciences. More especially, feminist researchers have shown that "complete freedom of research from its socio-political environment" is impossible (Margrit Eichler 1988). An analysis of early social science research objectives and methodologies shows these influences in different ways. One of the main objectives of early 19th social science was to provide data that could account for and explain the existing gender hierarchies, especially why women were and must remain inferior to men. Available scientific data at the time did not assist the social scientist, as that too tended to emphasise the naturalness of gender inequalities. A few examples will illustrate the point.
First, in the late 19th century, it was argued that because of menstruation, women should not and could not pursue certain careers. It was argued, for example, that university education "would deplete the energy necessary" for women's menstruation and reproductive functions, thus preventing them from becoming mothers. Second, another theory was the brain-size argument, which suggested that women's small brains made it impossible for them to equal men intellectually. Third, it was argued that women's hormones make them more caring, whereas men's make them more aggressive. Because of their natural role as mothers, women can never hope to achieve equality with men in public life.
The first two arguments were eventually overtaken by events, as more and more women joined higher education and performed as well as men. But the third argument is still used to try and prove that biology is destiny. It is obvious that underlying all these arguments is a basic self-defensive bias. As more women joined the formal industrial employment sector, what society saw as ideals of sexual differences have diminished, thus threatening the patriarchal ideology arrangement. Twentieth century research does not differ much from nineteenth century research. It too continued to produce data that tended to confirm theories of gender hierarchies which are incompatible with the real situation of men and women. Today, convictions about these gender hierarchies still persist. A typical characteristic of social science research data from the mid 40s to the 60s is the "invisibility" of women's work (or what one social scientist has tamed GYNOPIA).
The African societies are highly characterised by a distinct sex division of labour, directly related to existing gender hierarchies; yet, the nature of this division of labour is hardly understood. This is because researchers in this area have emphasised research on paid tasks and activities which left "women's work" uncounted, unpriced and unpaid. If we trace the development of economic research, its concepts, and priorities, the gender biases become more obvious. The definitions of work, production, and income as well as the distinctions between what is economic and social, private and public, illustrate biases that are worthy of a research focus themselves. It may be coincidence that we talk about "man hours", but it is neither coincidence, nor of small consequence, when the GNP statistics leave out housework with all it implies. Clearly there is a specific and powerful ideology that justifies classification of what is economic work to be paid and what is not. What needs to be recognised is that there are definite negative socio-economic consequences affecting the men and women involved in this kind of classification. But of immediate concern to us is the fact that given these biased concepts, it is not and it has not been possible to analyse the empirical realities of gender adequately. A gender-focused research approach is a response to this situation.
A gender-responsive research approach advocated by feminist researchers has cast light on the nature and issues of gender inequality. But feminist research of the early 1970s also had its own biases related to the urge to make women statistically visible. At that particular point, researchers were interested in proving that women have all along been involved in production. The debate on "integration" of women in development intensified and different views of the concept of integration produced data that were suitable for the purpose. There were those who were willing to accept the economic definition of "work" and production. From their perspective, it would appear that women "needed" to be "given" an economic activity as a process in integration. Consequently, projects, especially of the dress-making and bakery nature, mushroomed. The other perspective emphasised the need for a process that will give women more decision-making powers in order to enable them participate more fully in the economy.
It is this perspective which emphsises the need for research models, concepts, and approaches that explain the universality of women's economic and political subordination and illuminate reasons for and consequences of the choice that individuals make in life, e.g. is child rearing a question of individual taste and preference or work that is socially imposed and manifests a gender hierarchy? Whatever the shortcoming, the strengths of this research approach can be summarised as follows:-
clarifying sexism in the research process as well as existing gaps in available data;
development of appropriate statistical measurements and indicators of traditionally uncounted work;
refining research techniques to take the lives of subordinate and disadvantaged men and women into account;
refining and defining the concept of gender as a critical factor in social organisation; and
refining the concept of development to include issues and mechanisms of redistribution, and empowering the subordinate groups.
A few lessons have also been learnt from all the research experiences accumulated over the years. These include, among others:-
the knowledge that neither men nor women can be seen as a single separate unit of analysis; the appreciation that an understanding of men's experiences does not necessarily tell us everything about women's experiences; and the realization that we cannot learn fully about one gender through the eyes of another gender.
However, social science researchers in general have yet to realise that the misconceptions about where and what men and women are, what they can and cannot do, is directly related to the inadequacies and biases of our research methodologies. It is this realisation that will in future direct African social scientists towards a gender-responsive research process.
The future direction and strategies of social science research will, to a great extent, be determined by the role social scientists want to play in their respective societies. A major question we have to ask ourselves is why or even whether we want to understand gender dynamics. Do we want to understand gender dynamics in order to change the gender status quo or simply to accumulate knowledge about gender relations, differences and similarities? My own observations and suggestions about a future, more gender-responsive research approach is related to how I define the role of or what I expect from a social science researcher. I believe that her/his role is to provide data that can be utilised ultimately to reduce the existing gender hierarchies both at the ideological and empirical level. Some people have argued that sexual inequality is inevitable because women are innately less capable than men, but as other researchers have put it, this is "like saying we must accept near-sightedness because it is genetic and forget about developing eye glasses." (Janet Sayers: Analysing Gender) If the role of the researcher is to provide data necessary for reducing gender inequalities, then a gender-responsive research approach is inevitable. In order to achieve this task, I see three areas of immediate concern; two of them research concerns which must be given priority in the African setting, and one, a policy concern.
Over the years, attempts have been made to give visibility to women's roles in the production and reproduction sector. However, in the East African region very little research has been done on the relationship between cultural definitions of gender, gender roles and underdevelopment (underdevelopment as used here refers to resource under- utilisation and misuse). The gender belief system here refers to the socially constructed male/female categories. A society's gender belief system becomes very important because it identifies, expresses, maintains and justifies what is male and female as well as the appropriate gender relations.
Several reasons make gender belief systems a priority research concern, in a truly gender-responsive African research agenda. These have been outlined in general elsewhere (Kay Dean X, Mary Kite: in Analysing Gender, 1987; H. Moore 1988) but they are particularly pertinent in the African setting.
First, in spite of the fact that gender - related behaviour changes in time, place and context, beliefs in specific sex differences seem to persist.
Second, gender beliefs are important because they not only describe what is, but also prescribe what should be male/female behaviour.
Third, in the African setting, an understanding of the gender belief system becomes particularly imperative, since it may shed light on how women, who are the centre of reproduction, continue to be politically and economically marginalised.
Since the belief systems include stereotypes of men and women, attitudes of appropriate gender roles and behaviours, it becomes important in understanding why individuals make the choices they make,the possible degrees of freedom of diversion from descriptive and prescriptive behaviour,and the likely consequences. If research focuses on cultural construction of gender then it is able to address certain basic questions.
How do gender hierarchies shape lives of men and women? What is the role of cultural institutions (e.g. family/marriage) in maintaining the hierarchies. How do the concepts we hold about women and men affect our behaviour towards them and, in turn, how do individuals' perceptions of who they are (as men/women) affect their behaviour?
Research in other cultures has shown that beliefs individuals hold about each other affect the way they act towards each other. How does the perception of the African woman by the African man influence her behaviour in the management organisation and home setting? If, for example, an employer believes that all women are emotional and not fit for a management job, will he hire a woman for such a position? How does the social construction of domestic responsibility in which child-rearing (rather than just child-bearing) is seen as solely a women's role,affect the performance of women managers and ultimately their employment opportunities? Is the social construction of gender compatible with empirical realities of gender roles, relations, and positions?
These questions posed above are indication that a study of the gender belief system (of the way sex-differences are culturally interpreted, justified, and maintained) is important. It is important because its influences on gender relations both at the personal and social level are not yet clearly understood. It is also important because it may provide a clue to the universality of female subordination. In most of the African rural setting, where women are responsible for 80% of daily subsistence, a clue to their marginality may be part of the solution to Africa's food crisis and equity issues. The need for such a solution brings us to the second concern for gender-responsive social research. How do we measure or get indicators of the empirical reality of gender inequality and differentiation?
The role of gender-responsive social science research is to provide data that are useful in restructuring gender hierarchies. This basically implies equalising opportunities for men and women and, in many cases, opening up new roles for them. While equalising has not raised much protest (everybody seems to understand the need for social, political and economic equality) there is always a tendency to resist role change.
This is because in most cultures, role change challenges the traditional/cultural construction of gender roles and relations. Yet, any policy aiming at equalsing opportunities and increasing equity must identify and address the necessary changes in sex-roles. (Bonepath and Stjoper 1988). If changes in sex-roles are to be considered, then, the exact nature and impact of existing gender roles must be clearly understood.
For example, while the role of child bearing is specific to the female gender, child-rearing can be undertaken by either man or woman. What then is the basis for allocating the child rearing role to women alone? What is the actual impact of this role structure (ideological/empirical) on availability, performance and choices of men and women in the labour force? What, in fact, is the actual/empirical gender role structure?
Anxious to make women statistically visible, feminist researchers have focused mainly on women's productive and reproductive roles. A more gender-responsive research approach is now needed to balance the information on the actual roles and relations of men and women. A gender role analysis, i.e. assessment and documentation of who does what, when and where as well as the nature of differential access to and control of resources is being advocated as an effective means of collecting such information, especially where issues of equity and empowerment are major development goals. The approach is gender responsive in that it highlights the existing gender structures in productive and reproductive activities, in resource allocation, in control and use, and the related gender relations.
All of this is important information in development policy planning processes. If the goal of a gender responsive researcher is to provide data for equitable development policy then we have to consider the researcher's role in the process of policy development as the third area of concern for gender-responsive social science research.
The last decade is referred to as the lost decade for Africa by economists and developmentalists, while among researchers it is termed the decade of discovery, characterised by greater refinement in conceptualisation research methodologies and documentation of gender issues. Yet, it is also true that "if we look at women themselves at the end of the decade, research findings increasingly suggest that development strategies failed to improve their position and programmes often have worsened the position of women." Indeed, researchers have failed in their role "if achievement is measured in terms of sensitising and requiring that those responsible for the development strategies take account of women's needs and interests and of their capabilities to major development concerns." (Elisabeth Croll. 1985)
If one looks at the relationship between policy development and researchers, it is not difficult to explain this failure. Traditionally, researchers have concentrated mainly on identifying, defining, and investigating gender-related issues of development. Sometimes they have interpreted and disseminated their findings. But researchers in general have hardly followed up to ensure effective utilisation of their research findings (At the time, this was for the better, especially since the research methodologies were so biased that it would have been more calamitous to utilise the results, than to leave them dormant).
Given a more action-oriented gender-responsive research objective, researchers now need to become more politic and aggressive in making sure that their findings are utilised by the policy makers in the development process. To achieve this, social science research as a policy must proscribe unreliable data and develop a mechanism for data quality control. Researchers, as a "group", must adopt a policy of operation that gets them closer to the corridors of power, and gets them a more positive image. It may take another research agenda to find out how to penetrate the corridors of power.
The evidence for ineffectiveness and negative image of research in Africa is on the surface. The suspicion with which researchers are still regarded is one indicator.
Others are lack of national research budgets and the existence of development policies which do not reflect the empirical realities of men and women in different sectors. The fundamental reason is that because research data have not reflected reality they have failed to be impressive. Researchers need to study themselves to establish the fundamental reasons for their ineffectiveness. Whatever questions, explanations, and reasons we pose it is clear that social science research must ultimately be concerned with issues of socio-economic equity as well as efficient and effective utilisation of human resources. The data provided must illustrate this concern if they are to be treated seriously by policy planners. A gender-focused social science research approach is critical. It is this approach which will enhance our understanding of the social construction of gender and how this construction defines and determines the social allocation of gender roles, activities, opportunities, relations, and ultimately, the nature of gender inequality in African societies.
Finally, we have to understand that gender-responsive social science research which focuses on the three concerns identified above is only possible if the researcher is gender-responsive, mentally and practically.
This means that first and foremost, the researchers must change their conceptual, theoretical, ideological, ethical, and practical research orientation. Then and only then can we say we are on the road towards gender-responsive research for the year 2000.
1. Bonepath, E. and Stoper, E. 1988. "Introduction A Framework for Policy Analysis", in Women, Power and Policy.
2. Eichler, M. 1988. Non-Sexist Research Methods: A Practical Guide.
3. Goll, Elizabeth, 1985. "Lessons from the Decade: A Retrospective of Research in Women Creating Wealth".
4. Himmelstrand, K and Birkham, N. 1985. The Peripheral Centre: Swedish Assistance to Africa in Relation to Women. ALLF Publishing, Falths Printing House.
5. Beth B. Hess and Myra Mar Ferree: (eds.). 1987. Analysing Gender: A Handbook of Social Science Research. Sage Publications.
6. Hogvelt, A.M. 1980. The Sociology of Developing Societies, Second Ed. The Macmillan Press Ltd.
7. Moore, H.L. 1988. Feminism and Anthropology. The University of Minnesota Press.
8. Overhott, C., Cloud, K, Anderson, M.B. and Austen, J.E. 1986. "Women in Development: A Framework for Project Analysis," in Gender Roles in Development Project.
9. Schaffer, D.M. 1988. "The Feminisation of Poverty: Prospects for an International Feminst Agenda," in Women, Power and Policy. Toward the Year 2000. Boneparth, E. and Stoper, E. (eds).
10. Spivak, G.C. 1987. In Other Worlds: Essays in Cultural Politics. Methuen Inc.
11. Tadria, H.M. 1985. Changing Economic and Gender Patterns Among The Peasants of Ndejje and Ssegulu in Uganda. Ph.D. Thesis, University of Minnesota.