· How to make a GIF of a Stata graph on Stata Technical Tips
· A new SIEF policy note reports on an experiment testing different outreach strategies for TB testing in India. It finds incentivizing patients was more successful in bringing in people for TB testing in India than were health workers.
· On VoxDev, Erica Field and Rohini Pande provide an overview of their different experiments on making credit work better for women in India.
· A low-cost way to add a question or two to a survey in Kenya: I received the following email from Melissa Baker of Kantar Public, a survey firm I have used in Nigeria, and (with her permission), thought it might be of interest for many development researchers:
“We feel that development programmes in Kenya can benefit from more information from the population, but many organizations cannot afford the full cost of research. As such we have launched the partly Kantar-funded “Usemi” survey, on to which NGOs, government and foundations can add questions at low cost.
The price per question for the first wave is $200 / Kshs 20,000 for a single coded question (two answers) and $250 / Kshs 25,000 for a multiple coded pre-coded question, subsequent waves will be charged at $100 more per question type. Your questions will be asked by telephone to 1,000 Kenyan adults, and you will be provided with data in excel, which will be weighted for non-response and coverage providing representative results. Your data will belong only to you and will remain confidential.
Please let me know (melissa.baker@kantar.com) asap if you may like to add questions to this wave or let me have any questions or if you would like more information. Deadline for PO is 15th October.”
Note that this is a new service, and I do not know any other details, but thought it would be worth noting. We did something similar back when I worked on the 2007 World Development Report, adding a few questions about to surveys being done by the company Intermedia in 7 countries to get some descriptives on issues where we otherwise did not have data (e.g. who has most influence on schooling, work, and marriage decisions; internet usage; desire to migrate; whether people communicate with people in other countries). I blogged about some ideas on when adding a few questions to someone else’s survey could be useful back in 2011.
Join the Conversation