How publishing works in this field
Replication is institutionalised on the quantitative side. Leading journals require a replication package — data and code sufficient to reproduce every table and figure — deposited at acceptance in a named archive, and some run verification before publication. Building the package as you go is far easier than reconstructing it afterwards.
Qualitative and interpretive work is held to a transparency standard too, though a contested one: what sources exist, how access was obtained, and how interpretations were reached. Journals vary on how much of this they require, and the debate over applying data-sharing norms to fieldwork is live in the field.
What slows papers down here
Fieldwork in conflict and authoritarian settings raises ethics that a standard committee form does not cover — participant risk from being identified, researcher safety, data security against seizure, and whether informed consent can be meaningful. Journals increasingly want these addressed explicitly.
Causal identification draws the same scrutiny as in economics on the quantitative side: observational results framed causally without a design to support them are rejected by referees trained to look for exactly that.
Worth reading next
Questions we are asked in this field
The analysis data, the code that produces every table and figure from it, a readme naming software versions, and documentation of any restricted data. Many journals require deposit in a specific archive at acceptance, and some verify it before publication.
Address it in the method, not just the ethics form: how identities were protected in storage and in the text, how consent was obtained where recording was unsafe, what was not written down, and why. Reviewers in this field expect that account.
Usually, with an exemption. Journals have processes for restricted data — supply the code, document the access conditions, and raise it in the cover letter rather than at acceptance when it causes delay.