Random assignment is intended to create comparable treatment and control groups, reducing the need for dubious statistical models. Nevertheless, researchers often use linear regression models to "adjust" for random treatment-control differences in baseline characteristics.
Last week David linked to a virtual discussion involving Dave Giles and Steffen Pischke on the merits or demerits of the Linear Probability Model (LPM).
Recently I was spending some time with a survey firm in Tanzania, pre-testing a survey. I got to talking with one of the folks working at the firm about how they compensated their enumerators. He made it clear that they follow a fixed efficiency wage (i.e.
In a recent post, I described a randomized experiment in Jordan that I (along with Matt Groh, Nandini Krishnan and Tara Vishwanath) have been working on.
The Yale Savings and Payments Research Fund (YASPR), under the Global Financial Inclusion Initiative at Innovations for Poverty Action (IPA), is planning a training event for PhD (and PhD-track) researchers who are interested in conducting randomized control trials (RCTs) on micro-savings and payments.
The previous post in this blog discussed the positive dynamic effects of conditional cash transfer (CCT) programs in Mexico and Nicaragua – in particular on asset accumulation and the incidence of entrepreneurship by the rural poor.
The short-term benefits of certain social support programs such as CCTs have been well documented –CCT programs tend to raise household consumption as well as the utilization of schools and health clinics. It is a natural question, and one of great interest, to think more dynamically and ask whether these programs also enable households to invest in productive assets.
As part of a new series looking how institutions are approaching impact evaluation, DI virtually sat down with Nick York, Head of Evaluation and Gail Marzetti, Deputy Head, Research and Evidence Division. For Part I of this series, see yesterday’s post. Today we focus on DFID’s funding for research and impact evaluation.