August 7, 2020

IHDS Quick Start Guide

I recently used data from the India Human Development Survey (IHDS) for a paper on learning outcomes data in India. IHDS is a panel survey led by Sonalde Desai and others at NCAER/UofMd which collected on a wide range of topics from ~42,000 households across India. The survey includes two rounds of data collection, the first round in 2004/5 and the second round in 2011/12. (According to the website, a third round is in the works. Read more

July 12, 2020

School versus household based surveys for collecting learning outcomes data

I recently wrote a working paper in which I look at the reliability of learning outcomes data in India. The main findings of the paper are a) the government-run survey of learning outcomes (called the NAS) likely contains a lot of noise and b) the main independent survey of learning outcomes (ASER) is a tad bit noisier than the survey’s sample size would lead one to believe. In the process of working on the paper, I spent a bit of time looking at learning outcomes surveys across the world. Read more

May 26, 2020

Why India Should Focus on Educational TV rather than EdTech

A couple of weeks ago, Rob Sampson and I published an op-ed in Quartz India arguing that the central and state governments in India should use Educational TV rather than smartphone-based EdTech to reach students out of school due to the covid crisis. We use survey data to show that smartphone penetration in India is much lower than sales figures would lead you to believe. By contrast, TV ownership is relatively high even among the poor. Read more

May 6, 2020

Estimating seroprevalence with data from an imperfect test on a convenience sample

Several recent studies have used data from antibody tests performed on a convenience sample to estimate seroprevalence of covid 19 in a population. Estimating seroprevalence from this data presents two challenges. First, the analyst must take steps, through weighting or other measures, to deal with likely sample selection bias. Second, the analyst must take into account imperfections in the test itself. Addressing either of these challenges on their own is relatively straightforward to do using existing tools. Read more

April 28, 2020

An Alternative Approach to Power Calculations

The typical approach to power calculations goes something like this: first, the evaluator estimates the smallest MDE for which the intervention would be cost-effective. Second, the evaluator calculates the sample required to detect that MDE. Third, the evaluator throws out the calculations from step two after realizing the evaluation would be way too expensive and instead estimates the MDE she can detect given her budget constraints. The problem with this approach is that it doesn’t take into account the cost of the evaluation itself. Read more

© Doug Johnson 2020. Site design by Emir Ribic

Powered by Hugo & Kiss.