Skip to content

Latest commit

 

History

History

considering-projects

README.md

Project submission, selection and prioritization

Submission/evaluation funnel

As we are paying evaluators and have limited funding, we cannot evaluate every paper and project. Papers enter our database through:

  1. submission by authors;
  2. our own searches (e.g., searching syllabi, forums, working paper archives, and white papers); and
  3. suggestions1 from other researchers, practitioners, and members of the public, and recommendations from high-impact research-users2. We have posted more detailed instructions for how to suggest research for evaluation.

Our management team rates the suitability of each paper according to the criteria discussed below and in the aforementioned linked post.

Our procedures for identification and prioritization

We have followed a few procedures for finding and prioritizing papers and projects. In all cases, we require more than one member of our research-involved team (field specialist, managers, etc.) to support a paper before prioritizing it.

We are building a grounded systematic procedure with criteria and benchmarks. We also aim to give managers and field specialists some autonomy in prioritizing key papers and projects. As noted elsewhere, we are considering targets for particular research areas and sources.

See our basic process (as of Dec. 2023) for prioritizing work: process-prioritizing-research

Authors' permission: sometimes required

Through October 2022: For the papers or projects at the top of our list, we contacted the authors and asked if they wanted to engage, only pursuing evaluation if agreed.

In our Direct evaluation track under certain conditions3, we inform authors but do not request permission. For this track, we have largely focused on NBER working papers.

Communicating: "editors'" process

In deciding which papers or projects to send out to paid evaluators, we have considered the following issues. We aim to communicate the team's answers4 for each paper or project to evaluators before they write their evaluations.

Summary: why is it relevant and worth engaging with?

Consider: global priority importance, field relevance, open science, authors’ engagement, data and reasoning transparency. In gauging this relevance, the team may consider the ITN framework, but not too rigidly.

Why does it need (more) review? What are some key issues or claims to vet?

What are (some of) the authors’ main claims that are worth carefully evaluating? What aspects of the evidence, argumentation, methods, interpretation, etc., is the team unsure about? What particular data, code, proof, etc., would they like to see vetted? If it has already been peer-reviewed in some way, why do they think more review is needed?

To what extent is there author engagement?

How well has the author engaged with the process? Do they need particular convincing? Do they need help making their engagement with The Unjournal successful?

See what-research-to-target.md for further discussion of prioritization, scope, and strategic and sustainability concerns.

Footnotes

  1. E.g., following the 'bounty' we posted here and advertised in other places. ↩

  2. E.g., GiveWell, Open Philanthropy, or government agencies and NGOs. ↩

  3. "We will choose the most relevant papers in this prestigious series (in certain categories, and with certain restrictions). We will inform these papers' authors when we have decided to commission their work for evaluation. We will give the authors the opportunity to respond and engage, including the opportunity to ask for a temporary embargo for sensitive career issues, and the opportunity to respond to the evaluation before it is published. We discuss this further below, explaining the reasoning behind it." ↩

  4. However, we may hold back some of this discussion to avoid biasing the reviewers or in the name of discretion. ↩