Thursday 17 November 2011

Research on Gender Relations

There has been an encouraging flurry of research projects, "baseline" studies and evaluations on gender relations.  It seems that in recent years, hundreds of thousands - maybe millions - of people have been interviewed. They have been asked what they think and do about the roles of men and women, girls and boys (and maybe people who fit into an in-between category) - in the household, at school, at work, in politics and other aspects of life in society. I suspect that quite some people may have undergone several interviews by different research teams, especially in those parts of the world that receive extra donor attention - my mental map shows big blotches of donor interest around Bukavu, Goma, Kabul, Phnom Penh, to name a few places.

Does anyone care as to what that means to the people who are the objects of such studies? People with a professional research background normally do, as they have learnt about the importance of ethics in research. Basically, you don't want to inflict harm on the people you study. The risk of causing potentially lethal mental health problems is particularly high when you investigate gender-based violence - that is why PATH and UNICEF have produced a most commendable guide on Researching Violence Against Women. If you plan to do any research - even just a few interviews - on the subject, do read this guide. It can save lives.

Now, I feel that we should think beyond the recommendations of the PATH/ UNICEF guide if we want to avoid deepening the attitudes and social norms that underpin unequal gender relations and gender-based violence.

Research on social cognition - i.e., on the ways in which people and societies think - shows that even the tiniest gesture or conversation contributes to reinforcing - or weakening - individual attitudes and social norms. Hence, it is important you design and phrase your questions very carefully. For example, the question "it is important for a man to show his wife who is the boss" implicitly transports at least two gender-unequal notions: that (i) the man is the boss (note that he is not referred to as "husband", but as the generic "man"), (ii) he owns "his" wife. Now, if your questionnaire includes the symmetrical question "it is important for a woman to show her husband who is the boss", you may cause some hilarity - but it may be a good way to ensure you do not deepen those stereotypes about men being true men only if they hold power over women. The same applies to questions of the type "A woman should be able to choose her own friends even if her husband disapproves". Again, gender bias could be offset by asking the same question with reversed roles ("a man should be able..."). 

Imagine a woman being taken through, say, 30 questions, most of which implicitly portray women as having to ask permission from men, as being beaten by men, and so forth. The questions are administered by people who have come from outside, who take notes and who record the interview - a very impressive setting. In my view, such a situation is harmful and very likely to deepen women's sense (the interviewee's and the interviewer's!) of powerlessness. That in turn will prevent them from developing their own strength and power. Likewise, men who are interviewed with similar stereotypical questions (without the "neutralising" questions that would reverse roles) may leave the interview with a deepened feeling it's a true man's duty to control his wife.

Any research is an intervention. That applies to natural science (remember Heisenberg's uncertainty principle?) and even mores so in social science, where the researcher is part of what he studies. Interviews manipulate people's minds. If they reflect the dominant bias and stereotypes about gender relations, they will reinforce such bias and stereotypes, making gender equality and violence-free lives even more distant goals. And they are unfair to both women and men. They ignore the power women (can) have, and the burden "male supremacy" may mean to men.
Do you know anyone who works on these issues? I'd love to hear about that. Please share such information as a comment (below).

No comments:

Post a Comment