· Tim Harford on why experiments matter and why we hate them.
· I listened to the Center for Open Science seminar “Pre-results Review in Economics: Lessons Learned from Setting Up Registered Reports” on Tuesday, that had Andrew Foster discussing lessons from the JDE and Irenaeus Wolff lessons from a symposium of registered reports done for the journal Experimental Economics (EE). A few notes that may be of interest to our readers:
o The JDE has had 70+ submissions so far, and has accepted 15 stage 1 (pre-results) submissions, 2 of which have now been accepted as stage 2 final papers (with results). They have a steady flow about 2-4 submissions per month, and expect to end up with around 4 or so papers per year accepted through this route. Submissions may fall at the moment because a lot of fieldwork is suspended, while the fast turnaround of many Covid-19 related interventions does not allow time for the reviewing process needed for a registered report.
o The EE symposium had 21 submissions, and 3 of them are in R&R stage and the rest rejected.
o There is still some confusion among some authors about the difference between a pre-analysis plan and writing up a full registered report. So make sure you look to see what a registered report needs to contain before submitting.
o On that note, the JDE has set up a website http://jde-preresultsreview.org/ which has the accepted Stage 1 and Stage 2 papers listed, and in most cases downloadable, so you can look and see what is required.
o Registered reports have been harder than expected to judge because of the issue of power – the amount of guesswork involved in power calculations makes it hard to know how precise results will be ex post. At least having baseline data helps a bit, which is why timing-wise, it is best to have baseline data, and at least 4 months before follow-up data is to be collected when submitting a registered report.
· On the IDInsight blog, Ruth Levine offers guidelines for how to safely return to in-person surveying.
· Stata trick I learned this week: often I create graphs in Stata, and then want to pretty them up using Stata’s graph editor. But while you can record what you do in the graph editor as a .grec file, this is a pain for replicability since you need to go in and edit each time. I learned you should do the following steps:
1) Open the graph editor, start recording, and do the editing you want, and save the editing file as a .grec file.
2) Then open the .grec file, and you will see a bunch of commands. Take these commands and add gr_edit in front of them in your main do file.
E.g. now my do file would read something like
graph use "output\myfigure.gph"
gr_edit style.editstyle boxstyle(shadestyle(color(white))) editcopy
gr_edit legend.style.editstyle boxstyle(linestyle(color(white))) editcopy
gr_edit yaxis1.title.DragBy -.1931240126534509 -2.317488151841824
· Online course on impact evaluation: Billy Jack of Georgetown University, Alaka Holla of the World Bank’s SIEF program, and our former colleague Owen Ozier have put together an edX course on “Impact evaluation methods with applications in low- and middle-income countries” – it starts August 10, is 9 weeks, is free to do, with an option to pay for a verified certificate.
· Job opportunity: “The World Bank’s Poverty and Equity GP is looking to hire a consultant to work on a wide range of analytical activities across the Nigeria team’s work program. The team is looking for someone who is currently undertaking or who has recently finished a PhD in economics, with strong econometric and writing skills, and a solid grounding in Stata and other statistical software. See further details here.”
Development Impact will now be on break for August, and return after Labor day.
Join the Conversation