Methods

The following information provides more details about the structure of the weekly reports and how the analyses are performed.

Weekly Terminations vs Reinstatements

To determine the number of terminations and reinstatements in a given week, we use the variables termination_date and reinstatement_date. For each week starting from February 3rd, 2025, we count how many times a termination_date and/or reinstatement_date for that respective week is listed. We then calculate the “Weekly Change” as the number of reinstatements minus the number of terminations for that week, with the “Cumulative Net Terminated Grants” acting as a running total of the change in terminated grants across the weeks.

It’s important to note that the cumulative reinstatements and terminations calculated through this longitudinal method will not always match up with the values in the summary boxes that display the current number of terminations and reinstatements on the data pages, especially if a grant is missing information about their termination/reinstatement dates (read this related post) or a grant has gone through multiple rounds of terminations/reinstatements.

Consequently, by relying on the termination_date and reinstatement_date to determine when these disruptions occurred, this section focuses on specific snapshots of disruptions in terminated and reinstated grants from week to week. However, all grants are still included in calculations about the total number of disrupted grants, state-level analyses, etc. and the weekly reports will continuously update information about missing and changing dates as they become available.

Grants by State

To look at disruptions at a state level, we group grants by the state in which the recipient resides (org_state). We calculate the “total disruption” to include the number of grants that have been ever terminated. Consequently, even if a grant is currently reinstated, it is included in this total, since the grant had been previously terminated. However, “currently terminated grants” exclude any grants that have been reinstated and “currently reinstated grants” similarly exclude any grants that have been terminated. “Lost funding” is the amount of funding awarded to grants that was not paid out because of terminations, excluding reinstated grants; this is calculated by summing over the grants’ estimated remaining funds (i.e. the amount obligated to a grant minus the amount already outlaid)

(For NIH Reports) Grants by Type

NIH grants are unique as they can be distinguished by the following grant types:

  • R Series: These grants fund independent research projects led by a principal investigator. These are the most common types of NIH grants. Funding goes towards research aims rather than training or career development.
  • T Series: These grants go to institutions to support training programs for undergraduate, graduate, and postdoctoral researchers. A single T grant will support multiple trainees. These grants fund stipends, tuition, and training activities (e.g. courses, workshops, conferences).
  • Individual Training / Fellowships (F Series): These are research training grants for individuals, rather than institutions. The goal of these grants is to help predoctoral and postdoctoral trainees gain skills needed for a successful research career.
  • Early Career (K Series): These are grants for individual researchers (usually postdocs or early-career faculty) to help them become independent scientists. They provide salary support and research funding.

In this section, we group grants based on state the recipient organization is based, and then calculate the number of confirmed terminated grants across grant types as well as the amount of funding lost (calculated by summing over the grants’ estimated remaining funds, which is the amount obligated to a grant minus the amount already outlaid).

The states with the largest disruptions are determined by the total number of grants that were ever terminated in a given state, which includes grants that are currently reinstated since at one point those grants had been terminated.

Grants by Recipient

To determine disruptions at a recipient level, we group the grants based on their recipient organization and count the total number of reinstatements and terminations that occurred at a given institution. We created separate tables to demonstrate which institutions had the most number of disrupted grants (terminations + reinstatements), the most number of current terminations for that report’s week, and the most number of reinstatements for that report’s week, so that the full range of grant disruption is characterized.

Data and Agency Trackers

For more information about the agency trackers and what information is collected for the respective agencies, read these relevant posts in the “Updates” page of the Grant Witness website: