Published on Development Impact

Most popular Development Impact posts of the past year

This page in:

While we ease our way into writing new posts for the new year, I thought I’d take a look back, using our blog analytics platform to see who was reading Development Impact in 2023, and which posts were most read. One note here is that many of our readers just receive the blog posts in their email, and these email reads don’t show up in the overall statistics here. We also miss people reading posts through Feedly or other RSS feeds.

We had 416,715 unique visitors to the blog in 2023, with one-third of readers in the United States, and then the UK, India, Philippines, and Canada rounding out our top five countries. There were 56 countries from which we got at least 1,000 unique visitors, which is pleasing to us as we try to provide (niche) content of interest to readers around the world.

One of the benefits of writing blog posts is having people still finding posts useful several years later. In fact 10 out of the top 12 posts that generated the most traffic in 2023 were written before 2023 (which again might in part reflect people reading the recent posts in their email, and going to the website when searching for older posts):

1.       Jason Kerwin’s 2014 job market post on whether people think it is easy to contract HIV, which seems to generate a lot of traffic from people who would leave comments offering fake remedies for curing HIV until we turned off comments on this. So feel free to take off 80,000 unique visitors from our traffic numbers to get at what we think are more genuine readers.

2.       The topic of when to cluster standard errors remains popular, with this 2017 post summarizing Athey et al’s work still remaining very popular.

3.       Making visually appealing maps in Stata, a guest post by Asjad Naqvi from 2022.

4.       I’m not sure what is the search term that still leads so many to this 2014 post on resilience and recovery ten years after the Indian Ocean tsunami.

5.       Revisiting the difference-in-differences parallel trends assumption part II is the 2020 sequel post that was more popular than…

6.       Revisiting the difference-in-differences parallel trends assumption part I (also from 2020) 

7.       A 2014 post on Likert scales and how to ask whether you agree or disagree

8.       A summary from 2017 of 10 journals for publishing short economics papers – Dave Evans has now updated this with this Spreadsheet of 47 journals where you might send a short paper including many that have taken on the AER insights short paper format.

9.       A 2016 post on 11 tips for making a short presentation based on your research

10.   The only post from 2023 on our top-10 traffic list was this guest post by Owen Ozier on whether AI can write your Stata code – something that would be good to revisit again sometime soon given the rapid advances in AI.

11.   An overview of multiple testing commands in Stata from 2020. Note more people still go to the 2020 post, than to the updated post published in 2021.

12.   Remembering Martin Ravallion, written at the start of 2023.

Rounding out the top 5 posts that were published in 2023 were:

·         The state of development journals 2023 – my annual round-up of journal stats.

·         Interpreting treatment effects on an inverse hyperbolic sine outcome variable

·         Seven ways to improve statistical power in your experiment without increasing n

So apparently our readers love reading about impact evaluation methods and about how to publish research. Remember we’ve curated a lot of our posts into three curated lists on our webpage:

·         Postings on technical topics and methods

·         Postings on survey methods and measurement

·         Miscellanea, which includes interviews and advice

Thanks for reading, and hopefully this can serve as a spur and challenge for us to write some helpful and interesting posts in 2024.


David McKenzie

Lead Economist, Development Research Group, World Bank

Join the Conversation

The content of this field is kept private and will not be shown publicly
Remaining characters: 1000