Syndicate content

Development Impact turns 6: six questions for our sixth birthday

David McKenzie's picture
We are proud to have kept the blog going for another year, and would like to note its 6th birthday. In lieu of presents, we’d love your thoughts on what things you would like to see more or less of going forward. In particular, any comments or feedback on the following would be great:
  • One of the things we’ve kicked around is trying to put together an e-book to bring together (and do a bit of updating to) many of the posts around techniques and measurement into one place – which could then be a reader or reference for people working on impact evaluations. Does this seem useful, or do you think our curated links on technical topics and measurement topics are enough?
  • Was our learning from failure series useful or itself a failure? Anyone have additional failures they would like to share? If so, please email them to [email protected] and see the details at the end of this post for what we are looking for.
  • Is there anyone you would love us to get as a guest blogger, and if so, what would you like to hear from them on? Our rule for guest bloggers is that we aim to get people not just summarizing their most recent research paper (with the exception of our job market series), but rather have them talk about either one particular technical or measurement problem that comes up in their work, or their views on recent research in their field, or discussion of interesting experiences during fieldwork, etc. For example, I particularly like this post by Bruce Wydick on measuring hope, and this one by Abhijeet Singh on the issues in using standard deviations to measure effect sizes.
  • We ran our job market series again this year, running 17 posts in total over a month, which was just under half the number of submissions we received. We like being able to feature a lot of young researchers and get a nice variety of topics. However, we are not sure what the right quantity-quality trade-off is here, and whether this seems like too many posts, or is ok (I don’t think we want to be running more). Thoughts from both the supply and demand sides of the market, as well as from regular readers most welcome.
  • We are always delighted to visit universities and hear of the blog being assigned reading in some development courses, or of particular posts being included on syllabi. For those of you less delighted because your professor has assigned you to read this, anything we can do to make it more useful?
  • Anything you want to see less of?


Submitted by Stuart on

An e-book would be a useful resource, perhaps with a bit more framing of the posts within an overall approach. How about using an open review process like

Submitted by Asif Dowla on

First, congratulations. I read this blog religiously and learn a lot from them. I don't read the ones about the food fight about methodology--not my cup of tea.

I would recommend publishing an e-book based on blogs geared towards non-specialist and undergraduate students. You will probably need an outside editor for this. I have a hard time assigning readings from World Development Report. No disrespect, because WDRs are a compromise documents, they don't satisfy anyone and include too many angles and jargon that an undergraduate find hard to follow.

Submitted by Sam Manning on

As a current student with aspirations of pursuing a PhD someday, the job market series is really valuable and fun to read. Just one person's opinion, but I would love to see the number of posts stay around 15-20, and not less! Thanks

Submitted by Dan Stein on

Hi team, congrats on your six years and for maintaining such a useful resource for the community.

I think an e-book would be a really valuable resource. Here at IDinsight we recently had a bi-weekly "book club" discussion based around your curated links of technical topics. It was a big hit, especially among young staff who were looking to build their technical skills. I think converting to an e-book would allow us to use this material even more effectively as a resource for self-directed study.

As demand side of the job market, we also found the job market posts very useful. Apart from the interesting content, we used them to reach out to candidates we thought might be a good fit for IDinsight. As a newer player in this space, we couldn't assume good candidates would automatically apply, so this was a great resource for us. From our perspective, the more of these posts the better, as it exposes us to more potential candidates.


Submitted by Berk Ozler on
Hi Dan,

Really nice input - thanks a lot.


Submitted by Peter on

I'll contrast with others and say that, unless it adds content, an e-book won't be more useful than organizing links around methods and measurement, which are great posts.

For additional content to make an e-book more worthwhile, perhaps providing more code or detailed examples (e.g. for random assignment techniques, MHT adjustments, survey questions etc.). That level of detail may be too much for a blog post but ideal for an e-book.

Other aspects I like: summaries of papers and conference highlights and tips about conducting field work (e.g. implementation, measurement, random assignment)--particularly around overcoming challenges. These are great insights to draw on.

Great blog. Thanks and congratulations!

Submitted by Danae Roumis on

An e-book would be a useful resource, but perhaps it might also be worth considering a wiki, that can be updated periodically while housing all the great resources from past posts as well?

Submitted by John Quattrochi on

It'd be great if the blog had a search box separate from the search box for the whole WB website.

Submitted by Anna D'Souza on

Happy 6th birthday!

This blog is one of my favorites; it is invaluable for those of us working on applied microeconomic development topics. Thank you!

Here are my reactions to your list above:

1. I don't see many added benefits to an e-book over the current page of curated links. The real issue is the search functionality. I may be doing something wrong, but when I try to search by tag, only a small fraction of your tags are included in the drop down menu. And if I try to use the search feature at the top of the blog page, it provides many results that are not related to your blog. Google searches help, but you have to know what you are looking for.

2. The learning from failure series was very interesting, even for those of us who do not do fieldwork. Part of what makes this blog special is the fact that you seem to be honest about the ups and downs of fieldwork and research (including Berk's post on the impostor syndrome from a while back...).

3. I would like to see a guest post from someone working on post-conflict and conflict countries (e.g., Eli Berman, UCSD, or Jacob Shapiro, Princeton), given the specific challenges related to survey work and measurement in such environments.

4. I like the job market series post, but 17 did seem like too much this year. I think 10-15 would be ideal.

5. I look forward to assigning your blog in my MIA development course next year... let's see how the students like it!

6. The paper summaries after conferences are extremely valuable for those of us who don't get to go to many conferences. Thank you!

This blog is a great public service... thank you and please keep it up!

Submitted by Berk Ozler on
More feedback from a reader (via email):

"Absolutely love your blog, even though I'm not an economist but a semi-retired public health epidemiologist and teacher.  It's interesting, well written, and I often share articles with my university colleagues working in areas such as child marriage.  Love your technical blogs, which I share with some of my methodologist-friends and use myself--would be great to have these all put together in an e-book.

Many, many thanks for the work that goes into this!

Submitted by Amanda on

I agree with previous posts that an ebook with detailed examples of how to apply discussed methods would be very useful.
As a PhD student, I really liked your job market series and didn't find that 17 posts were too many.

Add new comment