Archive for November 2nd, 2009

I recently finished and highly recommend Portfolios of the Poor, a book examining the financial lives of dozens of impoverished families in Bangladesh, India and South Africa. The understanding of how the world’s poor survive on less than $2 per day was captured through an intensive research process that spent a year chronicling their financial activity through “financial diaries.” Researchers visited the surveyed families every two weeks to discuss their uses of money and in doing so came to have intimate insights into how they made and spent money.

The study sought to fill in the gap in between small ethnographic studies and large-scale surveys. The former cannot capture how lives change over time and the latter fail to capture the nuances of individuals. Portfolios is filled with revelations that came about solely due to their method.

While reading Portfolios, I couldn’t help but feel that ICT4D suffers from the same gap as microfinance. On the one hand, there are plenty of small, short-term studies of people’s interaction and relationship with technology (e.g. Chipchase), and on the other are large-scale surveys of national ICT usage (e.g. RIA).

So, what would an ICT diary project look like? Luckily, the authors of Portfolios provided an appendix detailing their methodology and processes. Some key points:

• Conceptualizing as diaries is useful because tracking in (near) real-time avoided memory loss (who really remembers how they used a cell phone two months ago?). It also helps to capture the relationship between money (or ICTs) and time. Looking at the flows was important for the Portfolios project because if they had simply surveyed irregularly they would have thought that large assets, which were the majority of the wealth for the subjects, were important; instead, it was the margin – small loans, daily wages – that provided most of the activity about their welfare.
• Given the deeply personal nature of both finances and communication, a long-term relationship between researcher and subject is necessary to develop trust. The Portfolios researchers were close in socioeconomic characteristics to their subjects to minimize the barriers. Even then, subjects were not always forthcoming and the researchers needed to really probe to discover all the financial instruments used. The trust issue also meant that the bimonthly visits needed to be conversational, not interviews. However, they write that “the strength of the diaries approach is that it can, over time, break down much of this reticence and confusion.”
• The effort wasn’t meant to be statistically representative, but they did choose families in low, middle and high levels of poverty.
• The information they gathered did not match their preconceived categories, so the researchers had to iterate towards perfection.
• There is the possibility that merely observing the financial lives of the subjects changed their behavior (a Heisenberg effect).

For an ICT diary project, there would obviously be differences. I imagine that ICTs are used more frequently and with less thought than major financial transactions, so there will definitely be uses that are forgotten or not mentioned to the interviewer. This problem could be minimized by using software to track usage, but capturing this quantitative data would need to be done very carefully to avoid privacy implications. A solid framework for usage would need to be utilized to make sense of various uses of ICTs and the behavioral evolution that I would expect to see in a year’s time.

Discussing this at Mobile Tech for Social Change Cape Town on Saturday, it was pointed out that the ISPs and telcos have much of this data, but getting them to share it would be next to impossible. Certainly looking at that would be useful, but the ICT diaries would be another way, and perhaps a better one, because the discussions with the participants would provide insights into their relationship (perceived and real) with technology.

If development practitioners are going to build ICT solutions for the poor, they must understand how they use technology. That understanding is certainly aided by the existing studies, but the gap in the middle seems to be where much of the insights will be found. Following ICT usage for a year would allow researchers to explore the evolving behavior that comes from both internal and external changes.

[Image from Kiwanja's mobile gallery.]