We conducted an online search for large-scale funding institutions (herein referred to as “funders”) across the EU-27 member states, Schengen area and the United Kingdom, that offer neuroscience research grants. Data was initially collected between 2020 and 2025, and all data was double-checked between September and December 2025. We started our search from public national funding agencies for each country (e.g., Deutsche Forschungsgemeinschaft-DFG, for Germany). In case we did not know the national funder of a country, we asked colleagues/collaborators from that country. We looked for any data reporting funding awarded over any time period, on the funders website or using Google search engine. Keywords in the search included, but were not limited to, “Neuroscience,” “Funding,” “Diversity,” and “Report.” Our goal was not to perform a comprehensive search but to identify at least one funder per country. We included all national, public funders as well as large private or umbrella funders widely used in the neuroscience field (e.g., the European Research Council-ERC). Where possible, we focused on data specific to neuroscience funding. If neuroscience-specific funding results were not available, we looked for and summarized data provided as a whole. We documented and recorded links to all data found, and data that was available to download was added to a publicly available OSF repository². From each data source found, we extracted the data type (report, list, tabular or available in a portal) and the years covered by the reporting. We then rated each data source according to the TRS, outlined below.

We developed the TRS, a new scoring tool to evaluate funders’ ability to assess and transparently report gender and other characteristics, inspired by FAIR principles (Wilkinson et al., 2016). To build the scale, we established 15 meaningful criteria, grouped into five categories (see below). These criteria enable analysis, interpretation, and tracking of potential biases in funding applications and awards.

A: Accessibility and FAIR principles

B: Gender/DEI and demographics

C: Temporal information

D: Funding

E: Field

Each criterion can score 1 or 0. Scores for each point in the scale are summed up with equal weight. Therefore, the maximum score on the TRS is 15. If a funder had data available in multiple formats (e.g., a report on their website and downloadable tabular data), they were first rated separately, and in further analysis, we took their highest possible score. Therefore, scores reflect the entirety of the data published by the funder rather than any specific format, since lower-scoring data sources would provide less information. Based on the total score, each funder received a transparency rating: bronze (1–8 points), silver (9–12 points), or gold (13–15 points). We also noted whether ethnicity or disability was reported; however, these criteria were not included in the scale due to differing rules across countries for collecting such sensitive data.

The scoring was performed by 5 of the authors (Natasha Clarke, Abigail E. Licata, Paola Mengotti, Antonietta Gabriella Liuzzi, Doris Pischedda), independently. We adopted an objective scoring method, therefore, each author scored different funders and it was not necessary to calculate inter-rater reliability. For each scale item, we assigned a score of 1 if the answer to the relative question (see previous paragraph) was positive and 0 otherwise. For example, for the “downloadable” criterion, if the answer to the question “Can data be downloaded?” was positive, we assigned a score of 1. In case of doubt, the item was scored by at least another person, until agreement was reached.

We tested whether there was an association between TRS scores and the (a) Gender Equality Index (GEI), which measures progress in gender equality in different domains across EU countries, developed by the European Institute for Gender Equality (2025), and (b) Gross Domestic Product (GDP) per capita, by calculating the Spearman’s rank correlation coefficient. Confidence intervals (CI) for Spearman’s ρ were estimated via Fisher z-transformation. We obtained the most recent data available: GEI data for 2025 were obtained from the European Institute for Gender Equality website³, and GDP data for 2024 were obtained from the World Bank using the world-bank-data Python library. For both analyses, for countries with multiple funders, we took only the maximum TRS score, to highlight what is possible in reporting as opposed to penalizing lower TRS scores, however we also tested the correlations using the minimum and mean TRS scores per country.

Data was visualized using Python. All code is available at https://github.com/WomenInNeuroscience/funders_project. We report descriptive statistics for the TRS and its subcategories. We also provide code to easily calculate scores on the TRS, using Python.

Across 32 countries and the EU as a whole, we identified 39 funders that offer neuroscience research grants. Of these, 90% (35/39) shared publicly available data on funding results, and 75% (29/39) shared downloadable data. Only 26% (10/39) shared data in tabular format. The majority of funders (75%, 29/39) shared data in English. Across all 53 data sources, the majority was provided in an annual or one-off report (47%, 25/53).

Examining DEI metrics, 59% reported data on gender (23/39), and 67% (26/39) reported age or career stage (65% (25/39) reported career stage while 26% (10/39) reported age). Information on parental leave could be found for 47% (18/39) of funders, while 59% (23/39) had a webpage or information on their DEI policies. Although not part of the TRS scoring, we note that 8% (3/39) of funders reported ethnicity or disability.

The year of award was reported by 80% (31/39) of funders, while 34% (13/39) reported the grant duration and 85% (33/39) published data over three or more years. The earliest data available was from 1963 (Novo Nordisk Foundation, Denmark). After 1993, the number of funders reporting data steadily increased (see Figure 2) until 2020, after which a drop in reporting in the recent years (2021–2025) was observed.

The total amount of funding awarded was reported by 72% (28/39) of funders, while 57% (22/39) reported the amount given per individual awardee. Data for all applicants, as opposed to only those who were successful, was provided by 57% (22/39) of funders.

Neuroscience-specific data was retrievable or could be disaggregated in a minority of funders (29%; 11/39), whereas in the majority of cases neuroscience projects were embedded into macro-categories (e.g., medicine, psychology, biology).

We applied the TRS to the 35 funders with available data (see Figure 3). A gold rating was awarded to five funders: the ERC, Swedish Research Council, Swiss National Science Foundation, The Wellcome Trust (United Kingdom) and the Dutch Research Council. Silver was awarded to 18 funders, and bronze to 13 funders. No data could be found for three funders, receiving a rating of zero, and one country (Liechtenstein) had no national funding programs. A full breakdown of TRS scores is available as a .csv file at both the GitHub repository⁴ and the OSF project page (see Footnote 2).

The percentage of funders in the bronze, silver and gold scoring categories fulfilling each TRS criteria can be seen in Figure 4. As with other reported results, this analysis takes the highest score awarded to a funder if multiple data sources were found. Gold funders met all criteria of the “Gender/DEI and demographics” domain (in yellow), and most of the “Accessibility and FAIR principles” domain (in blue), with the exception of publishing tabular data. Four criteria were well-met across bronze, silver and gold funders: having findable data, reporting in English, reporting the year of an award, and reporting data longitudinally (over at least 3 years).

Since Liechtenstein did not have a national funder, and therefore had no data to report, it was excluded from correlation analyses. EU-wide funders were also excluded, since their funding schemes are open to the United Kingdom and Schengen area countries, and therefore do not match EU-level GEI and GDP indicators. Spearman’s rank correlation was used to examine the association between TRS scores and the GEI for each country in the EU with available data (N = 25), and each country’s GDP (N = 31), as the data were non-normally distributed. There was a significant positive correlation between TRS scores and both the GEI [ρ = 0.64, 95% CI (0.33, 0.83), p = 0.001] and GDP per capita [ρ = 0.51, 95% CI (0.20, 0.74), p = 0.003] of the countries where funders are based, indicating that funders from wealthier countries with higher GEI scores tended to score higher on the TRS (Figure 5). Results using minimum TRS score, and mean TRS score, per country, were similar for both GEI [minimum TRS: ρ = 0.52, 95% CI (0.162, 0.761), p = 0.007; mean TRS: ρ = 0.605, 95% CI (0.28, 0.81), p = 0.001] and GDP [minimum TRS: ρ = 0.454, 95% CI (0.118, 0.696), p = 0.010; mean TRS: ρ = 0.50, 95% CI (0.17, 0.72), p = 0.004] correlations.

The OSF repository contains 43 data records from 29 funders, spanning 25 countries or the whole of Europe, from 1974 to 2025. It is publicly accessible at https://osf.io/9cy6a/, and organized according to the format “Country-Funder.”

We aimed to assess the transparency of European funders (EU member states, Schengen area, and the United Kingdom) in sharing demographic and other data important for the evaluation of equitable research funding, motivated by the FAIR principles (Wilkinson et al., 2016). To meet this goal, we developed the TRS, offering a structured, reproducible framework for assessing reporting practices and rated funders, acting as a benchmark. By aggregating publicly available funding data into a single repository, we provide researchers with an unprecedented opportunity to evaluate and compare potential biases and progress over time. We aimed to evaluate transparency in neuroscience funding; however, only data from 29% of funders could be disaggregated specifically for the neuroscience category. Others included neuroscience data pooled with other fields, for example falling under Life Sciences, Health or Psychology. This lack of specific data limits the conclusions that can be drawn about transparency and equity in funding within neuroscience and supports the notion that neuroscience poses a particular challenge in evaluating and addressing gender disparities (Fulvio et al., 2021; Morgan et al., 2021).

Overall, five funders received a “gold” TRS rating, which provide funding across Europe or in Sweden, Switzerland, the United Kingdom or the Netherlands, demonstrating that comprehensive and accessible reporting is feasible. The predominance of “silver” ratings indicates widespread efforts to be transparent with funding data. Our rating should not be taken as discouraging or as punishing funders who showed a less structured, efficient or complete sharing of their data, but rather show a way forward, by following the practices of the funders with the higher ratings. However, the lack of findable data from a small number of funders suggests that structural barriers to monitoring equity in research funding persist, consistent with evidence that gender and other biases remain challenging to track systematically (Boyle et al., 2015; van der Lee and Ellemers, 2015; Schmaling and Gallo, 2023). We found a correlation between transparency and a country’s GDP, suggesting that data collection and/or sharing may be partially explained by financial or organizational resources, though this correlation should not be interpreted as causal. We also observed a correlation between transparency and the GEI, which measures a country’s progress in gender equality. While also only an association, this result provides evidence of convergent validity for the TRS, as it suggests that transparency in funding practices co-occurs with broader structural commitments to equity.

Our analysis revealed that the availability of data has improved over time, with a rise in public reporting since the 1990s (see Figure 2), however with a decrease in reporting from 2020 on. We can only speculate on the reasons behind this decrease. Missing reporting in the most recent years (2020–2023) involve mostly funders who publish the data in the form of an annual report. The delay may be due to a delay in writing such reports, as data are often presented in a processed form. Other reasons may include the fact that not all programs are offered annually, or the impact of COVID-19, if grants were not provided or funders did not have the infrastructure to publish data. Missing data from 2023 to 2025 may simply reflect delays in data sharing. The majority of funders shared some data, often in English and downloadable, supporting findability and accessibility of data. However, the scope, quality and structure varied substantially. Data was provided in different formats, with the majority embedded in annual or one-off reports, limiting interoperability (Wilkinson et al., 2016). Almost 60% of funders reported data on all applicants (not only awardees) and provided the amount awarded per grantee. These data are essential for assessing potential biases, as prior work shows that women apply for funding less often and, when successful, tend to receive smaller awards than men (Boyle et al., 2015; Hechtman et al., 2018; Schmaling and Gallo, 2023). These metrics are also critical for evaluating the effects of COVID-19, which disproportionately affected women in the workplace, including in the rate of grant submissions (Berge et al., 2023).

Gender remains the most widely studied source of bias in science (Lewis et al., 2021; European Commission: Directorate-General for Research and Innovation, 2025). A number of factors, including citation practices (Dworkin et al., 2020; Rosenblatt et al., 2025), teaching loads (Gibney, 2017; O’Meara et al., 2017), lower representation in senior and decision-making positions (Palser et al., 2022; Joyce et al., 2024; European Commission: Directorate-General for Research and Innovation, 2025), and birthing or adopting a child disproportionately impact women in STEM careers compared to men (Cech and Blair-Loy, 2019; Morgan et al., 2021). A particularly strong form of gender bias is the maternity bias, which affects women with children or who are pregnant. In fact, mothers are less likely to be hired or to get a promotion (Wolfinger et al., 2008), they receive lower salaries and fewer job offers (Kelly and Grant, 2012), and are regarded as less competent than fathers and people without children (Correll et al., 2014). These factors can accumulate across women’s careers and create a more challenging environment, making them less favorable recipients of funding (Sato et al., 2021). However, intersectional factors such as age, race, ethnicity and disability, can compound barriers to funding (Berge et al., 2023; Boyle et al., 2015; Ginther et al., 2018; UK Research and Innovation [UKRI], 2023). The reporting of career stage can serve as a useful proxy for age-related inequities, as research shows that women in later career stages or older age receive less funding than men at equivalent stages (Boyle et al., 2015). Our TRS results for DEI reporting revealed a lack of transparency across funders: 67% reported age or career stage, and 59% reported gender. While not included in the TRS specifically, only 8% reported ethnicity or disability. This limited reporting hinders the identification and remediation of potential biases, particularly intersecting ones, in funding allocation.

The underreporting of DEI data may reflect limited awareness of its importance, resource constraints for data collection and analysis (supported by the finding of an association with GDP), reluctance to disclose potentially unflattering findings, or national regulations that restrict the collection of sensitive data. However, the consequences of this lack of transparency are significant, as it perpetuates a system in which biases can persist, limiting opportunities for underrepresented groups in neuroscience. In our data, very few funders reported ethnicity or disability, with United Kingdom funders being the exception. Although collection of such sensitive demographic data is generally permitted in EU member states, national laws may limit what is collected, with some countries focusing instead on nationality or citizenship. Socioeconomic status, a key factor in academic pathways (Morgan et al., 2022), may also impact funding equity, and intersect with gender and other demographics (Ruediger et al., 2025); however, to our knowledge, funders do not collect or publish this data. Additionally, only 44% of the funders mention parental leave in their grant guidelines, providing information about what would happen to the funding (e.g., grant time extension) during the leave time. Some funders, such as the ERC, show best practices to this regard, taking into account time dedicated to care for children in correcting academic seniority for eligibility, with an 18-month flat-rate extension per child.

Some policies have already been proposed to improve inclusivity in the research funding system. Torres et al. (2024) discuss some of these measures aimed at promoting inclusion of caregivers (particularly women), that could help address gender bias (especially maternity bias) and inequalities in research funding. For example, extending grant eligibility criteria or track-record evaluation by at least 18 months per child would support continuity in research and reduce the impact of career gaps (e.g., missed grant calls, delayed manuscript submissions, lost data collection time). Moreover, providing the same paid parental leave options to both parents, could help mitigate gender disparity in workforce participation and parents’ income, and foster a more equal sharing of caregiving duties by mothers and fathers, while, at the same time, being more inclusive for same-gender couples. Funders could play an important role in this respect by providing additional paid paternity and maternity leaves to make parental leave opportunities more equitable. Funders should also provide flexible funding to allow for hiring support figures to perform lab work during parental leave, in order to ensure continuity in research activities. Additional measures could be implemented to reduce gender bias. For example, funders could mention actions to counter this bias in the reviewer guidelines, reporting clear examples and citing relevant laws against discrimination, or even provide specific training with proven effectiveness to reviewers.

Furthermore, Gender-Net Plus⁵, a European research project funded under the Horizon 2020 program, provides practical recommendations for funders to promote gender equality. Importantly, they highlight six concrete actions: (1) National authorities should support and monitor funding agencies’ efforts to promote gender equality. (2) Funding agencies should have a permanent structure (department or task force) to monitor a clear Gender Equality Plan. (3) Evaluation panels and decision-making bodies should aim for at least 40% representation of each gender. (4) Funding agencies should encourage women to apply, design calls inclusively and account for career breaks. (5) They should also make funding processes, evaluation criteria and feedback publicly available. Finally, and completely in line with our work on developing the TRS, they recommend that (6) funding agencies should collect, analyze and publish gender-disaggregated data on applicants, grantees and reviewers to access trends and gaps. This mirrors our approach and reinforces our view that funding agencies should record application and award data comprehensively and in a systematic way, and make it easily accessible to the public to improve transparency in resource allocation. Recording data from all applicants (including gender, etc.) is particularly important, as gender imbalance is often present already at this stage. This would also allow further analysis to identify the main sources of biases and the most effective measures to prevent them.

A limitation of this work is that our analysis was constrained by the authors’ geographical and language knowledge, which may introduce bias. Where possible, data in other languages was evaluated if an author spoke the language or it could be translated online; however, this was not always feasible (for example, data embedded in files that online translation tools could not recognize). The TRS criterion of reporting in English is also biased, but necessary, as a common language used across scientific publishing that enables comparisons across countries, and benefits researchers seeking funding in other countries. We also acknowledge that our list of funders is not exhaustive, and other agencies may demonstrate good practices in transparent reporting that are not captured here. The goal of this paper is to present the TRS and demonstrate its feasibility with preliminary data; future work should expand its use to other countries to provide a more comprehensive view of funding transparency and potential biases worldwide. We also note that high transparency does not imply a lack of bias, but is rather a critical step for evaluation and insight. Finally, we recognize that gender is non-binary. Current reporting typically uses a binary classification (women/men); identifying non-binary individuals can pose a risk to privacy, but we hope that future work will enable more gender inclusive analyses of funding equity.