- 1 General Information
- 2 Expectations
- 3 The Speakers
- 4 Workflow and Tools
1 General Information
The focus of the course is close reading and re-analysis of emerging research in the political economy of development, broadly construed. The focus is on well identified research whether based on experimental or observational data. It is intended for graduate students who already have strong analytic skills. Auditors are welcome as long as they put in the work—that is, sign up to be on a replication team.
The overall structure is that an external “speaker” comes to discuss new or in-progress research. The speaker does not actually present the work however; instead they share their papers, data and code in advance with the class and a “replication team” puts together a detailed discussion of the work, usually involving a replication of core results and a re-analysis or extension of the main findings.
Note this course will be taught in April 2023 with speaker sessions on 3, 4, 5, 6, 10, 11, 12, 13, 17, 18, 19 April with a final set of sessions to discuss your designs on 20 April. Most days we will try to have a lunch or a dinner with the visiting speaker.
The reading loads are not especially heavy; typically the speaker will provide 1 or 2 readings that give a sense of their research agenda. You should read these carefully. You should also look at the data whether or not you are on the “rep” team. There is no point coming to the class unprepared. My thoughts on reading and discussanting. See also chapter in declare design book
Guest speakers will be asked to share data in advance, and students are encouraged to replicate results and submit the results to robustness checks before each class.
Every registered student will be expected to write a one-page response paper in advance of the talk each day. This is due into the class drive by midnight the day before. If you are presenting in a given day this is not required.
A “rep” team (likely of two students) will be assigned a formal role as discussants and prepare oral and written commentary for the guest speaker. You should expect to be on one rep team each week.
Key elements of this are:
Be in touch with authors and be sure you have the data, papers, and all you need at least a week in advance
Make sure you can make sense of the data and run a basic replication.
When you have a feel of things jot down a brief “pre-replication plan”. What do you plan to look at? What do you expect to find? Post it to Git before engaging in reanalysis (honor code)
Then there are two ways to expand the analysis;
- One is to check for robustness. How much do things depend on the particular models or measurements?
- The second is to go more deeply into the logic of the explanation. This might sometimes require assembling more data, constructing new tests and so on.
Generate a presentation that
- presents the paper in general
- goes through the results and replication and
- goes through robustness and extensions
- does all this in quorto or rmarkdown so that speaker has content and code in a single file (great reference: https://quarto.org/docs/presentations/revealjs/)
(ideally) contribute your replication to a class package (I will share notes on how to do this)
I urge you also to try to use
DeclareDesginto formally characterize the research design in abstract terms and assess its properties
- Note that while we focus a lot on statistical replication and re-analysis there are many sides to a paper. Your presentation should not shy from discussing more fundamental conceptual or interpretational issues as appropriate.
You will be expected to write a research design containing (i) a theoretical argument or motivation, (ii) a proposed empirical test of that argument (iii) a formal design object and (iv) a discussion of policy prescriptions that might result from the argument.
3 The Speakers
We have a fantastic lineup of speakers all from nearby and most recent CU grads. They are all doing inspiring work:
Dates for speakers will be added when we have them.
- Kate Baldwin, Yale
- Guy Grossman, Penn
- Sarah Khan, Yale
- Salma Moussa, Yale
- Summer Lindsay, Rutgers
- Gwyneth McClendon, NYU
- Laura Paler, American
- Cyrus Samii, NYU
- Ken Scheve, Yale
- Tara Slough, NYU
3.1 The Rules
It is a very unusual thing for speakers to come and share data on unpublished work. It makes for terrific feedback and learning, but can also bring some risks to speakers. This cannot be thought of as a public presentation of research in the usual way and different rules apply. In particular:
- If a speaker requests that data not be shared outside the group, or perhaps even outside the replication team, this has to be adhered to strictly on pain of permanent ostracism.
- Any new findings from the analyses do not belong to the class or the students that engaged in the replication. You are working with the data for training purposes not for research purposes; you might see amazing patterns in the data but they don’t belong to you.
- Any public commentary has to be bland at best. If you have to tweet or related after sessions, these should be of no cause for embarrassment for speakers.
4 Workflow and Tools
We are going to be pretty hardcore about the workflow and using a set of very recent research tools to make sure all the work in the class is transparent and replicable.
The main tools that we will employ are:
- GitHub - for collaborating on code, publishing replications and raising issues; make sure you have a
- Drive - for sharing data with one another that cannot be put on git
- R - for conducting statistical analysis and authoring documents in…
- rmarkdown (or quorto)- for authoring replications and pages on GitHub
I plan also to develop an
R package structure so that, if authors give permission, the entire class can be shared as a kind of model replication package.