Turning a swamp into fish farm
21 November 2015
Credit shield insurance – Piloting microinsurance products in Bangladesh
7 January 2016
November 16, 2015
A field officer visits a household of a BRAC client to collect data.

A field officer collecting data on her smartphone.

We use our smartphones for numerous quotidian purposes: taking photos, accessing social media, browsing the web, and of course, making phone calls. But BRAC has been employing these devices for an entirely different purpose, and it is extremely innovative.

On August 2014, BRAC’s integrated development programme (IDP), which operates in the hard-to-reach areas of Bangladesh, commenced a pilot project in the remote Shemarchar and Chamakpur villages in the Sylhet district. Frontline field officers were provided with smartphones to collect information on IDP clients digitally. Since then, once every two months field officers have been collecting detailed data on the condition of each household. The data is electronically sent to the BRAC head office in Dhaka for analysis. The result is a rich longitudinal dataset with over 70 indicators and 9,000 households.

Here are five main reasons why this ‘smart’ method of data collection should replace the traditional means of acquiring data on paper.

Time saving and cost-effective. In the traditional system where data is collected on paper manually, the process can be very cumbersome. On the contrary, the  ‘smart’ method is faster as it uploads digitised, real-time data on a central server and merges everything automatically.

Immediate access. The real-time data from smartphones is stored in a structured central database. This enables researchers to instantly locate relevant data and add further analysis or comments.

Aggregation of data.The smartphone platform automatically generates a unique household ID for each household. The unique ID enables BRAC researchers to observe how the ten programmes that IDP comprises of, actually affect the lives of the clients.

Research potential. A routine data collection system like this opens the door to numerous research questions. Just with a handful of other indicators, the current dataset is capable of answering numerous research questions like what factors affect the probability of a household sending its school-aged children to school or if households receiving support of the TUP programme do significantly better economically than other low-income households in the long run, etc.

Ease of impact analysis.During the time of data collection I found that the rich panel datasets that were being generated are great tools to answer impact analysis questions.In order to understand the impact of IDP on the lives of those it serves, I analysed a sub-set of this dataset, which includes the households linked with IDP’s targeting the ultra poor (TUP) programme. TUP uses a ‘graduation model’ to uplift households from cycles of poverty.veryhigh

The result?

According to the analysis, extremely poor households that received support from the IDP-TUP programme 16 to 18 months ago, have better income the month they were surveyed than the previous month. There was also an increased usage of contraception and hygienic latrines. Compared to households that received the support 0 to 3 months ago, their families were also perceived to be healthier.

This agrees with eminent economists who argue that sustainable poverty alleviation techniques should be holistic. The rich dataset also helped answer some important questions regarding the programme: it showed that the TUP programme and government safety net programmes do not reach out to the same population, indicating that the two institutions must have different selection criteria. The promising findings of this research add to the tremendous potential of the programme.

It is an exciting time at BRAC. The new avenues that smartphones have opened up indicate there is a lot to learn from this immensely cost-effective and innovative approach for collecting data.


Inara Sunan Tareque is an intern at BRAC’s social innovation lab.

Skip to content