Surprising Trends from Wealth Data // How Data are Differentiated by Gender and Not Just by Geograph

Sometimes we might be surprised by what a thorough analysis of our data shows us. For example, if you had asked me a couple of weeks ago what the most influential factor on wealth definitions was, I would probably have told you geography. Geographical variations, including things like climate, altitude, the types of crops people are able to grow, and how these factors affect market and subsequent levels of wealth, are what I would have said first. In different geographical areas, modes of livelihoods will be different and this is why you get different descriptions of what wealth means to people.

I recently started writing a paper about the different definitions of wealth across our study sites to explore these differences, and to try and understand why there are such differences in the way people conceptualise wealth.

The data for this came from focus groups, conducted across Tanzania, for the wider Livelihoods Project. Focus group discussions were conducted with men and women separately, in Kiswahili, and then translated into English. I have conducted a thematic analysis of these data.

Analysing the data for this paper has been an interesting and exciting process. And it has led me to take another look at the way we make assumptions about data that we *think* we know well.

Initially I had been confident that much of the differences we were finding were due to the geographical variations I discussed above. As I began examining these data more closely I found that differences were in fact more pronounced between genders, even within the same study site, rather than geographically. There are geographical differences, but the difference in what men said, and what women said, is far more pronounced.

At this stage in the analysis and writing process, I don't want to draw too many conclusions, but from my initial analysis it certainly looks like gender is playing a much more important and prominent role than I originally thought it would.

My first example is women's and men's discussions of the attributes of a 'wealthy' home. There are indeed geographical variations - for example, all the houses in our Arumeru study sites are very modern. They all have sheet metal or tiled roofs, glass windows, furniture, and well kept gardens. In Iringa, houses tend to be more commonly built of mud bricks, many are still grass thatched, and very few homes have glass windows, like in the photograph below of the mud house with the metal roof. However, when you ask men and women about houses, their answers are different. Men give far less detail, mainly talking about attributes like building materials, whereas women talked about the furnishings, the type of toilet, and whether a house had an attractive compound with flowers planted.

Without making too many assumptions at this stage, are we seeing women's traditional roles reflected here? Women tend to traditionally spend time at home taking care of domestic matters, and so they might notice such attributes more than men do. Men traditionally build houses. Is this why they are more aware of building materials?

Another example shows how men and women talked about private education and healthcare. It seems that, from these data, women are more focussed on issues of health and education, and the quality of these, than men are. They seem to be more detailed in their discussion of these factors and how they construct their ideas and definitions of wealth.

There are, of course, obvious geographical differences. In some areas, farmers keep pigs. In other areas, there are no pigs. Depending on the geography, some areas have very large farms, with farmers commonly owning more than 10 acres, yet in other areas owning more than a quarter acre is considered a lot. Despite being an unapologetic feminist, I still thought that geographical differences would be more pronounced than gendered variation in wealth definition: I over estimated the importance that environment has on our perspectives. Once again, I’m finding that gender and the proscribed social roles that come with that, are a powerful influence on how humans understand wealth.

Certainly as I continue to write this paper, patterns in responses will emerge, and I will update this post as that happens. But in these initial stages of data analysis and writing, it does seem that there are gendered differences in definitions of wealth which might be more powerful than the geographical variations I expected to see.