Gender Disparities in Grant Submissions and Allocations: Examining Patterns Across STEM and Non-STEM Fields

Abstract

Gender inequality remains a persistent challenge in academia, shaping career trajectories and access to critical resources. Research funding plays a central role in academic advancement, yet studies of gender disparities often overlook how patterns vary across disciplines and stages of the grant process. This study analyzes gender differences in both the submission and allocation phases using administrative data from four national Israeli funding agencies, covering nearly 5000 applications submitted over two years (2017–2018). It examines two key dimensions within both stages—PI gender and grant amount—and disaggregates results across 12 academic fields. The findings indicate isolated, rather than systematic, gender inequalities, with disparities appearing more frequently in STEM fields but remaining limited overall. The results underscore the importance of field-specific analysis for understanding gender inequality and suggest that targeted interventions, rather than uniform policies, are needed to promote gender equity in research funding.

1. Introduction

Gender inequality remains a defining characteristic of academic institutions globally (Barnard et al. 2022; Forman-Rabinovici et al. 2024). Women are persistently underrepresented in faculty positions, with disparities becoming even more pronounced at higher academic ranks (Greska 2023). The imbalance is particularly severe in the areas of science, technology, engineering, and mathematics (STEM) (Casad et al. 2021; Colwell et al. 2020; Liu et al. 2019).

Among the structural forces that may sustain gender inequality in academia, access to research funding stands out as potentially highly consequential. Grants are not only essential for conducting scholarly work, but also serve as a key marker of academic prestige and a gateway to professional advancement (Bloch et al. 2014; Dubosh et al. 2020; Eloy et al. 2013). Because of this dual role, disparities in funding access may not only affect individual careers but also contribute to the persistence of gendered hierarchies, particularly in fields where women remain underrepresented.

Empirical research supports this concern. Studies have shown that women often face disadvantages in securing research grants, suggesting that funding processes themselves may reinforce existing structural inequalities in academia. Funding appears to disproportionately favor male researchers (Yip et al. 2020), and these gender gaps may emerge at several points along the funding process. Lower rates of grant applications among women, even relative to their already smaller representation on the faculty, may contribute to this disparity (Ley and Hamilton 2008; Pohlhaus et al. 2011). In addition, women may be less successful in securing grants (Bornmann et al. 2007; Bowman and Ulm 2009), may request smaller amounts of funding (Gordon et al. 2009), and may receive a smaller proportion of the amounts they request compared to their male counterparts (Bedi et al. 2012; Eloy et al. 2013).

However, the existing literature presents an incomplete picture. Many studies examine only a single stage of the funding process—most commonly, the award stage—and report mixed results regarding the existence of gender disparities (e.g., Gordon et al. 2009; Waisbren et al. 2008). Moreover, much of this research focuses on specific disciplines such as pediatric medicine, archaeology, biomedicine, or isolated institutions, limiting the generalizability of the findings (Bowman and Ulm 2009; Yip et al. 2020). Given that gender norms and power structures vary substantially across academic fields, insights drawn from one discipline may not necessarily translate to others.

While prior research has often focused on individual stages of the grant process or has examined disparities within specific fields, few studies adopt a multi-stage, cross-disciplinary approach. This study addresses that gap by analyzing gender disparities in research funding across multiple stages using multiple units of analysis. Specifically, it considers both the gender of the principal investigator (PI) and the size of the grant as key dimensions, measured during both the submission and allocation phases. This framework captures four critical elements in which gender inequality may manifest: submission rates, amount requested, success rates, and amount awarded. Each of these is analyzed across 12 distinct academic disciplines. This multi-dimensional approach enables the study not only to assess whether disparities exist, but also to pinpoint where in the process they emerge and which disciplines are most affected.

To explore gender disparities in academic funding, this study analyzes two years of data, from 2017 and 2018, from four leading national research funding bodies in Israel. The dataset includes 4998 applications and provides detailed information on the applicants’ gender, requested funding amounts, awarded amounts, and disciplinary affiliations. It employs statistical methods to detect gender differences in submission behaviors, funding requests, success rates, and grant outcomes, with comparisons made across 12 academic fields.

By adopting a multi-stage, multi-unit, field-specific perspective, this study offers a more nuanced understanding of gender disparities in academic funding. Moving beyond aggregate assessments allows for the development of targeted, evidence-based interventions tailored to overcoming the distinct barriers that women in different academic disciplines face. Ultimately, such efforts are essential for promoting greater equity and inclusivity within the academic research environment.

1.1. Gender Inequality in the Submission Stage

Gender disparities in research funding may originate at the earliest stage of the grant process: submission. At this stage, researchers submit proposals—typically accompanied by detailed budgets—for peer review. The underrepresentation of women in academia may partly account for the lower overall submission rates among female faculty. However, even after adjusting for women’s representation on the faculty, studies consistently find that women submit fewer grant applications than men (Burns et al. 2019; Ley and Hamilton 2008; Rissler et al. 2020; Waisbren et al. 2008). This gap persists even when controlling for academic rank. Indeed, this gap may be further exacerbated in higher levels of the faculty ranks. Senior women submit proportionally fewer applications relative to senior men than the gap observed among lower-ranked faculty members (Nguyen et al. 2023). Based on this existing knowledge, this study hypothesizes that women will be underrepresented among grant applicants relative to their proportion within the academic faculty.

Beyond differences in submission rates, gender disparities may also emerge in the amounts requested in the grant proposals. Even when men and women submit applications at comparable rates, systematic differences in the requested funding amounts can reinforce inequities in research resources. Studies conducted in medicine and other fields have found that women often request smaller sums than their male colleagues (Gordon et al. 2009; Waisbren et al. 2008), although exceptions have been documented (Yip et al. 2020). While findings vary somewhat by discipline, the majority of field-specific studies report patterns of gender inequality that disadvantage women. Building on these findings, this study hypothesizes that women will request smaller average amounts of funding than men.

One possible explanation for these disparities lies in the horizontal gender segregation of academic fields. Male-dominated disciplines, particularly in the exact sciences, tend to attract larger grants and more substantial funding opportunities (Rusu et al. 2022). To disentangle the effects of gender from those of funding patterns based on the discipline, it is important to analyze submission behaviors and the requested amounts using field-disaggregated data. Beyond differences in field-level funding patterns, internal dynamics within disciplines may further amplify gender disparities.

Patterns of inequality may be particularly pronounced in STEM fields, where a combination of psychological, institutional, and cultural factors exacerbates gender gaps. In these fields, where women constitute a small minority, additional psychological and structural barriers may further deter women from seeking research funding (Casad et al. 2021; Sidelil et al. 2023). For instance, women in STEM areas are more likely to experience imposter syndrome—a psychological pattern characterized by feelings of self-doubt and perceptions of personal success as illegitimate or happenstance (Beesley et al. 2024; Heslop et al. 2023). These feelings may be intensified by exclusionary environmental cues and limited access to mentorship and informational networks, further reducing women’s opportunities to apply for grants (Casad et al. 2021). Therefore, this study hypothesizes that gender disparities in both submission rates and requested funding amounts will be more pronounced in STEM fields than in disciplines with more gender parity, such as the social sciences and humanities.

1.2. Gender Inequality in the Funding Allocation Stage

The second stage where gender disparities may arise is during the allocation of research funding. Funding outcomes are typically determined through peer review and committee deliberations. While not all differences in funding outcomes necessarily result from bias, evaluation processes are susceptible to the influence of both explicit and implicit gender stereotypes (Marsh et al. 2009; Helmer et al. 2017). Reviewers’ assessments may be shaped by subjective perceptions of a researcher’s competence or the perceived value of their proposed work (Ellemers 2018; Heilman et al. 2024), creating potential pathways through which disparities could emerge.

Studies examining success rates have produced mixed findings. Some research suggests that men are more likely than women to be awarded grants (Bornmann et al. 2007; Bowman and Ulm 2009; Severin et al. 2020; Tamblyn et al. 2018). In contrast, other studies detect no gender-based difference (Gordon et al. 2009; Waisbren et al. 2008) or even higher success rates for women in certain institutional contexts (Yip et al. 2020). Disparities in success rates are often attributed to differences in the reviewers’ evaluations of the “quality of the researcher” rather than the merit of the proposal itself, with men frequently regarded as more capable researchers (Van Der Lee and Ellemers 2015; Witteman et al. 2019). These perceptions imply a form of gender bias, whereby evaluations are influenced not solely by the content of the proposal, but by gendered assumptions about competence, ambition, and scientific potential.

Structural factors—particularly academic rank—further complicate these patterns. Women’s concentration in the lower academic ranks, which are associated with lower rates of funding success, may partly explain the observed disparities (Steinþórsdóttir et al. 2020). When academic rank is accounted for, gender gaps often narrow or become statistically insignificant (Waisbren et al. 2008; Warner et al. 2017). Nevertheless, because women remain underrepresented in senior positions, this study hypothesizes that women will have a lower overall grant success rate than men.

In addition to success rates, gender disparities may emerge in the size of the grants awarded. Some studies find that men not only receive larger grants (Schmaling and Gallo 2023) but also are awarded a greater proportion of the amount they request (Bedi et al. 2012; Eloy et al. 2013). However, some of these differences may reflect initial disparities in the amounts requested rather than biased evaluation processes (Waisbren et al. 2008). To account for this possibility, this analysis examines both the absolute sizes of the awards and the proportions awarded relative to the amounts requested.

Given that prior research has yielded mixed findings, it is possible that gender disparities in the allocation of funding vary across academic disciplines. Therefore, rather than positing a general hypothesis, this study tests field-specific variations. Thus, this study hypothesizes that gender gaps in both success rates and awarded amounts will be more pronounced in STEM disciplines than in fields with greater gender parity.

Several mechanisms likely contribute to these disparities. First, women in STEM disciplines are disproportionately concentrated in the lower academic ranks compared to other disciplines (Howe-Walsh and Turnbull 2016; Van Miegroet et al. 2019). This fact may reduce their funding success, given the strong association between academic rank and grant outcomes. Second, evaluative bias may be particularly acute in STEM fields. Studies show that stereotypes portraying men as more competent scientists systematically disadvantage equally qualified women during evaluations (Ellemers 2018; Heilman et al. 2024). Such biases are especially pronounced in disciplines regarded as masculine, such as the STEM fields of engineering and physics (Bird and Rhoton 2021). Gender biases in perceptions of individuals are well-documented in hiring and promotion processes (Howe-Walsh and Turnbull 2016), and these patterns could potentially be mirrored in the grant review process. Finally, the composition of review panels may further exacerbate disparities based on gender. Evidence suggests that both men and women exhibit same-gender preferences in evaluations (Helmer et al. 2017). In male-dominated STEM fields, the underrepresentation of women among reviewers may limit advocacy for female applicants.

Taken together, these structural and socio-cognitive factors suggest that gender gaps in grant success rates and awarded amounts will be greater in STEM disciplines. Therefore, this study hypothesizes that, while women will have lower grant success rates than men overall, gender disparities in both success rates and proportions awarded will be more pronounced in STEM fields than in fields with more gender equity, such as the social sciences and humanities.

2. Method and Materials

2.1. Data

Data were collected through an initiative led by The Council for the Advancement of Women in Science and Technology (2019), within the Israeli Ministry of Science and Technology (MOST). The Council compiled data from four major research funding bodies in Israel: the Israel Science Foundation (ISF), the German–Israeli Foundation for Scientific Research and Development (GIF), the U.S.–Israel Binational Science Foundation (BSF), and MOST itself covering the years 2017 and 2018. Each funder provided information on grant submissions, including the gender of the principal investigator (PI), research field, amount requested, and amount awarded. All budget requests and grants are reported in Israeli New Shekels (NIS), each equal to about 28 American cents. These grants are primarily intended for tenured or tenure-track faculty members to support basic research.

The grant application review was not blind. The PI’s identity was known to both the funding organization and the reviewers. That said, in the data made available by the organizations, only the PI’s gender was provided. Other identifying details such as name or academic rank were not included.

Although the classification of research fields varied slightly across funders, the data were standardized into 12 categories using the taxonomy of the Israel Council for Higher Education (CHE). These categories included agriculture, biological life sciences, chemistry life sciences, computer sciences, engineering, environmental sciences, humanities, interdisciplinary research, mathematics, medicine, physical life sciences, and social sciences. All data were made publicly available in a report published by The Council for the Advancement of Women in Science and Technology (2019). Among the four organizations, there was a total of 4998 submissions, of which 1485 (29.7%) were awarded funding.

Data were disaggregated by field for several key reasons. First, field-level analysis is critical for disentangling gender disparities in grant outcomes from patterns that reflect horizontal gender segregation across disciplines (Macarie and Moldovan 2015). Both the average grant size and gender representation vary substantially between fields, potentially confounding aggregate comparisons. For example, the average grant size in engineering was 261,749 NIS, compared to 183,759 NIS in the social sciences. Similarly, only 11.9% of engineering submissions came from women-only principal investigators (PIs), compared to 36.4% in the social sciences. Without accounting for these disciplinary differences, observed gender gaps risk conflating structural field-based funding patterns with gender-specific dynamics.

Second, field-level analysis improves the ability to detect the sources of inequality. Much of the existing literature is field-specific and has produced inconclusive or conflicting results. These outcomes may reflect genuine variation across disciplines rather than inconsistencies in the findings. Disaggregating the data allows for a more nuanced analysis of where the disparities emerge most sharply, providing a clearer understanding of the mechanisms underlying gendered outcomes in research funding.

Each submission was coded based on the gender composition of the principal investigator(s) (PIs) as female-only, male-only, or mixed-gender teams. Most submissions involved a single PI, with an average of 1.12 PIs per application.

Table 1. Submissions by field and the gender of the PI.

Field	Women Only Projects	Men Only Projects	Mixed Team Projects	Total Projects
Agriculture	3 (27.3%)	7 (63.6%)	1 (9.1%)	11
Computer Science	46 (15.5%)	237 (80.1%)	13 (4.4%)	296
Engineering	38 (11.9%)	261 (81.8%)	20 (6.3%)	319
Environmental Sciences	39 (17.6%)	160 (72.1%)	23 (10.4%)	222
Humanities	161 (30.6%)	333 (63.2%)	33 (6.3%)	527
Biological Life Sciences	216 (24.1%)	572 (63.8%)	109 (12.2%)	897
Chemistry Life Sciences	16 (11.3%)	125 (88.0%)	1 (0.7%)	142
Physical Life Sciences	146 (20.5%)	519 (73.0%)	46 (6.5%)	711
Interdisciplinary	28 (16.7%)	133 (79.2%)	7 (4.2%)	168
Mathematics	15 (7.8%)	171 (89.1%)	6 (3.1%)	192
Medicine	152 (20.9%)	434 (59.5%)	143 (19.6%)	729
Social Sciences	285 (36.4%)	367 (46.8%)	132 (16.8%)	784
Total	1145 (22.9%)	3319 (66.4%)	534 (10.7%)	4998

presents the distribution of submissions by the PI’s gender across fields. The highest proportion of female-only submissions occurred in the social sciences (36.4%), while mathematics had the lowest (7.8%). To isolate the relationship between the PI’s gender, submission behavior, and funding outcomes, submissions from mixed-gender teams (534 in total) were excluded from the primary analysis. Mixed-gender teams were excluded because their funding patterns may reflect collaborative dynamics rather than the independent effect of the PI’s gender. The final analytic sample consisted of 4464 applications.

Data on women’s share of the academic faculty by field came from the same report as the grant data. Note that, while some grant submissions fall under the interdisciplinary category, no women faculty members are categorized as “interdisciplinary.” Therefore, analyses that compare the rate of women’s submissions to the proportion of women on the faculty do not include the “interdisciplinary” category.

As the results in

Table 2. Women as a proportion of the academic staff by field.

Field	Women on the Academic Faculty (%)
Agriculture	34%
Computer Science	20%
Engineering	22%
Environmental Sciences	42%
Humanities	37%
Biological Life Sciences	25%
Chemistry Life Sciences	13%
Physical Life Sciences	13%
Mathematics	14%
Medicine	20%
Social Sciences	44%

indicate, the representation of women varies considerably by discipline. Fields such as the social sciences (44%) and environmental sciences (42%) exhibit relatively high proportions of female faculty members. Conversely, women are notably underrepresented in several STEM fields, including chemistry life sciences (13%), physical life sciences (13%), mathematics (14%), and computer science (20%). Medicine also has a relatively small proportion of women (20%), despite being outside the traditional STEM cluster. These data provide the baseline context for evaluating gender disparities in grant submissions within each field.

2.2. Method

To test the hypotheses related to submission rates, requested amounts, win rates, and the proportion of funding awarded, statistical methods were selected based on the type of variable, distributional assumptions, and sample-size considerations. Analyses were performed separately for each academic field to account for discipline-specific variation in gender representation and grant dynamics. All statistical analyses were conducted using SPSS 29 software.

To evaluate whether men submitted more grant applications than women relative to their representation on the faculty, a binomial test was conducted. This test evaluates whether the observed proportion of a binary outcome (here, applications submitted by women) deviates significantly from the expected proportion. The expected proportion was based on the share of women on the academic faculty for each field. This approach enabled the identification of the under- or over-representation of women among the applicants, relative to their presence on the faculty. The binomial test is particularly appropriate for this hypothesis, as it is designed for binary outcomes and incorporates baseline expectations (in this case, gender representation), providing a direct measure of over- or under-submission rates (Abdi 2007).

To compare the mean amounts requested by male and female PIs, an independent samples t-test was conducted. This test assesses whether the means of two independent groups differ significantly, assuming approximate normality and the homogeneity of the variance (Kim 2015). Cohen’s d was also used to quantify the effect size of any statistically significant differences (Cohen 2013).

In fields with small sample sizes for one gender, the one-sample Kolmogorov–Smirnov (K-S) test was used to assess the normality of the distribution. The K–S test determines whether a sample’s distribution deviates significantly from the normal distribution (Ghasemi and Zahediasl 2012). If the data violated normality assumptions, the Wilcoxon rank-sum test (Mann–Whitney U test) was employed. This non-parametric alternative to the t-test compares medians and rank distributions and is robust to outliers and skewness (Conover 1999).

To assess whether grant success rates varied by gender, a chi-squared test of independence was conducted. This test evaluates whether an association exists between two categorical variables. In this study, the variables were the PI’s gender and the grant award outcome. This test is commonly used to examine group differences in proportions (Agresti 2007).

Finally, an independent samples t-test was also used to compare the proportion of the requested funding that was ultimately awarded. This method is appropriate, as it assesses differences in the mean proportions of two independent groups. The corresponding Cohen’s d value provided an estimate of the effect size (Cohen 2013). One-tailed significance tests were used for all relevant tests because the hypotheses specifically predicted a difference in favor of men, rather than a general difference in either direction.

3. Results

The following section presents the results of the statistical analyses, organized according to the study’s four primary hypotheses. It begins by examining differences in the submission rates of men and women across academic fields. Regarding the hypothesis that women will submit grant applications at lower rates than men, the results in

Table 3. Binomial test of proportionality between the share of women’s grant submissions and their share on the academic faculty.

Field	Women (N)	Men (N)	Observed Proportion	Expected Proportion	p-Value	Summary
Agriculture	3	7	0.30	0.34	0.710	Not significant
Computer Sciences	46	237	0.16	0.20	0.064	Not significant
Engineering	38	261	0.13	0.22	0.000 ***	Significant (fewer women)
Environmental Sciences	39	160	0.20	0.42	0.000 ***	Significant (fewer women)
Humanities	161	333	0.33	0.37	0.023 *	Significant (fewer women)
Biological Life Sciences	216	572	0.27	0.25	0.065	Not significant
Chemistry Life Sciences	16	125	0.11	0.13	0.332	Not significant
Physical Life Sciences	146	519	0.22	0.13	0.000 ***	Significant, reversed (more women)
Mathematics	15	171	0.08	0.14	0.009 **	Significant (fewer women)
Medicine	152	434	0.26	0.20	0.000 ***	Significant, reversed (more women)
Social Sciences	285	367	0.44	0.44	0.457	Not significant

* p < 0.05; ** p < 0.01; *** p < 0.001 (one-tailed test).

indicate that women submitted significantly fewer applications than expected in 4 of the 11 fields analyzed: engineering, mathematics, humanities, and environmental sciences. In contrast, women submitted proposals at higher-than-expected rates in two fields: physical life sciences and medicine. In the remaining fields—social sciences, biological life sciences, chemistry life sciences, computer sciences, and agriculture—there were no statistically significant differences between men and women in submission rates.

The second hypothesis proposed that women would request smaller grant amounts than men.

Table 4. Average grant size requested (NIS) by gender and field.

Field	Women	Men	Mixed Teams
Agriculture	166,094	147,949	240,000
Computer Science	235,082	241,718	302,098
Engineering	272,875	261,211	247,626
Environmental Sciences	215,543	205,807	218,618
Humanities	163,216	177,329	213,196
Interdisciplinary	278,253	246,546	238,655
Biological Life Sciences	277,644	275,631	267,780
Chemistry Life Sciences	260,497	267,111	423,730
Physical Life Sciences	260,453	263,444	274,810
Mathematics	186,667	218,575	217,690
Medicine	297,167	295,962	270,880
Social Sciences	177,548	182,795	199,852

summarizes the average requested grant sizes across fields. The largest average requests were in medicine and biological life sciences, while the smallest were in agriculture, humanities, and social sciences. Men requested larger grants on average in 6 of the 12 fields analyzed.

Table 5. Test of the difference between the average grant size (NIS) requested by women and men.

Field	Average Female (SD)	Average Male (SD)	t	df	p-Value	Cohen’s d
Agriculture	166,094.33 (126,521.86)	147,949.57 (113,329.93)	0.225	8	0.414	0.155
Computer Science	235,082.17 (97,755.27)	241,717.79 (98,671.21)	−0.418	281	0.338	−0.067
Engineering	272,874.53 (97,785.31)	261,211.16 (101,359.18)	0.666	297	0.506	0.116
Environmental Sciences	215,543.44 (57,740.34)	205,806.83 (82,399.24)	0.697	197	0.487	0.124
Humanities	163,215.69 (82,558.24)	177,329.50 (98,274.08)	−1.573	492	0.058	−0.151
Interdisciplinary	278,252.61 (131,346.55)	246,545.74 (95,963.92)	1.483	159	0.070	0.308
Biological Life Sciences	277,644.29 (143,234.44)	275,630.54 (119,922.66)	0.199	786	0.842	0.016
Chemistry Life Sciences	260,496.69 (76,043.80)	267,110.62 (135,775.96)	−0.191	139	0.425	−0.051
Physical Life Sciences	260,452.76 (119,430.56)	263,443.82 (116,025.72)	−0.273	663	0.392	−0.026
Mathematics	186,667.00 (55,222.46)	218,753.63 (89,794.01)	−2.027	21.153	0.028 *	−0.366
Medicine	297,167.45 (148,141.45)	295,961.80 (137,143.26)	0.091	584	0.927	0.009
Social Science	177,547.56 (80,755.22)	182,794.89 (90,714.74)	−0.768	650	0.222	−0.061

* p < 0.05; **; p < 0.01; *** p < 0.001 (one-tailed test).

presents the results of statistical tests comparing the mean requested amounts between men and women. A statistically significant difference was evident in only one field: mathematics. In mathematics, women requested an average of NIS 186,667, while men requested an average of NIS 218,575. The Cohen’s d value of −0.366 indicates a moderate effect size. It is important to note, however, that the mathematics sample included only 15 submissions from women out of 186 total submissions, warranting caution in interpreting these findings.

The third hypothesis proposed that women would win grants at lower rates than men.

Table 6. Success rate by field and the PI’s gender.

Field	Female Success Rate	Male Success Rate	Mixed Team Success Rate
Agriculture	0%	0%	0%
Computer Science	39%	38%	31%
Engineering	39%	26%	15%
Environment Sciences	38%	25%	9%
Humanities	37%	37%	27%
Interdisciplinary	43%	29%	14%
Biological Life Sciences	34%	28%	27%
Chemistry Life Sciences	44%	34%	100%
Physical Life Sciences	27%	36%	24%
Mathematics	47%	46%	33%
Medicine	24%	24%	20%
Social Sciences	24%	22%	25%

presents the gender breakdown of success rates across fields. Although men secured a higher share of awarded grants in every field, the findings of the table demonstrate that men’s success rates were not necessarily higher. Men had higher success rates in the field of physical life sciences only. Women achieved higher success rates in eight fields. In three fields men and women achieved identical success rates.

Table 7. Difference between men’s and women’s rates of grant wins relative to their rate of submission.

Field	Chi-Squared Value	df	Exact Sig. (One-Sided)	Valid Cases
Agriculture		1		10
Computer Science	0.041	1	0.482	283
Engineering	2.791	1	0.072	299
Environmental Sciences	2.841	1	0.071	199
Humanities	0.004	1	0.516	494
Interdisciplinary	1.958	1	0.121	161
Biological Life Sciences	3.325	1	0.042 *	788
Chemistry Life Sciences	0.542	1	0.318	141
Physical Life Sciences	4.239	1	0.024 *	665
Mathematics	0.006	1	0.573	186
Medicine	0.120	1	0.410	586
Social Sciences	0.291	1	0.327	652

* p < 0.05; ** p < 0.01; *** p < 0.001 (one-tailed test).

summarizes the results of statistical tests comparing grant success rates relative to submission rates. Of the 12 fields analyzed, only physical life sciences and biological life sciences exhibited statistically significant differences. In the physical life sciences men won grants at a disproportionately higher rate relative to their share of submissions, while in biological life sciences women won grants at a disproportionately higher rate relative to their share of submissions.

The fourth hypothesis posited that women would receive a smaller proportion of the funding they requested compared to men.

Table 8. Average size of grants requested and grants received (NIS) by grant recipients, field, and the PI’s gender.

Field	Female Requested	Female Won	Female % Won	Male Requested	Male Won	Male % Won	Mixed Requested	Mixed Won	Mixed % Won
Agriculture	0	0		0	0		0	0
Computer Science	245,172	210,422	85.8	257,644	211,090	81.9	397,836	231,700	58.2
Engineering	318,902	261,614	82.0	263,839	232,182	88	243,971	207,000	84.8
Environmental Sciences	214,557	189,345	88.2	205,803	178,574	86.8	223,389	205,000	91.8
Humanities	172,080	129,424	75.2	175,049	134,738	77.0	246,622	185,889	75.4
Interdisciplinary	255,467	214,667	84.0	264,457	218,597	82.7	238,500	220,000	92.2
Biological Life Sciences	296,011	228,953	77.3	304,564	245,145	80.5	285,720	242,403	84.8
Chemistry Life Sciences	276,803	235,050	84.9	282,276	229,297	81.2	423,730	330,000	77.9
Physical Life Sciences	265,298	200,913	75.7	281,896	231,305	82.1	300,198	237,350	79.1
Mathematics	220,222	192,857	87.6	242,503	189,646	78.2	199,820	134,200	67.2
Medicine	320,562	237,023	73.9	332,287	243,565	73.3	340,708	253,839	74.5
Social Sciences	169,813	144,816	85.3	179,790	147,772	82.2	197,266	164,909	83.6

presents the average size of the grants requested, the average amount awarded, and the proportion of requested funding awarded, disaggregated by gender. Men received larger total grant awards on average in 7 of the 12 fields analyzed—biological life sciences, computer science, humanities, interdisciplinary studies, medicine, physical life sciences, and social sciences. However, these differences largely reflected higher requested amounts at the submission stage. In only four fields—biological life sciences, humanities, engineering, and physical life sciences—did men receive a greater proportion of their requested funding compared to women. As

Table 9. Average proportion of amount won relative to amount requested.

Field	Women (SD)	Men (SD)	t	df	p-Value (One-Sided)	Cohen’s d
Computer Sciences	0.911 (0.230)	0.857 (0.208)	0.990	105	0.162	0.256
Engineering	0.859 (0.174)	0.888 (0.116)	−0.790	82	0.216	−0.225
Environmental Life Sciences	0.903 (0.137)	0.881 (0.143)	0.519	53	0.303	0.157
Humanities	0.802 (0.173)	0.831 (0.167)	−1.080	180	0.141	−0.171
Interdisciplinary	0.870 (0.216)	0.862 (0.240)	0.099	49	0.461	0.033
Biological Life Sciences	0.842 (0.196)	0.852 (0.199)	−0.342	230	0.366	−0.048
Chemistry Life Sciences	0.848 (0.124)	0.854 (0.178)	−0.081	48	0.469	−0.033
Physical Life Sciences	0.823 (0.186)	0.847 (0.171)	−0.766	223	0.223	−0.135
Mathematics	0.877 (0.108)	0.816 (0.170)	0.929	83	0.178	0.367
Medicine	0.800 (0.182)	0.799 (0.195)	0.024	139	0.491	0.005
Social Sciences	0.856 (0.117)	0.845 (0.144)	0.477	147	0.317	0.078

* p < 0.05; ** p < 0.01; *** p < 0.001 (one-tailed test).

indicates, statistical tests revealed no statistically significant differences between men and women in the proportion of requested funds ultimately awarded in any field.

4. Discussion

By disaggregating gender differences by field, across multiple stages of the grant allocation process and using multiple units of analysis, this study provides a more detailed understanding of where and when inequality in research funding emerges. The results reveal little systematic evidence of widespread gender inequality, with no field exhibiting consistent disparities across all stages and measures.

Some instances of inequality in grant allocation can be attributed to lower submission rates among female academics. However, the patterns of under-submission do not align with a clear division between STEM and non-STEM fields. While women were underrepresented among applicants in engineering and mathematics, they submitted proposals at higher-than-expected rates in the physical life sciences. Moreover, in fields with greater gender balance, inconsistencies remained: women submitted proposals at lower rates in the humanities but showed no significant difference in submission rates in the social sciences. Regarding the size of the grant requests, the only statistically significant gender difference was in mathematics, where women requested moderately smaller amounts than men.

Mathematics was also the only field that displayed consistent gender disparities both in submission rates and requested amounts. Note that mathematics is a STEM field and also has one of the smallest proportions of female faculty members (14%). This finding suggests that gender inequality might be more prevalent in STEM disciplines that are heavily dominated by men. However, the absence of consistent disparities in other STEM fields indicates that inequality is not inherent across STEM fields as a whole.

In terms of grant success rates, only in the physical life sciences did men secure funding at disproportionately higher rates relative to their submission rates. However, there were no statistically significant differences between men and women in the proportion of requested funding awarded in any field. Where men did receive larger overall grant amounts, this result appeared to be primarily a function of requesting larger sums rather than evidence of bias in the allocation process. Again, the presence of disparities, mainly in traditionally male-dominated STEM fields, may suggest that gender stereotypes and structural barriers are more persistent in these contexts.

Although initial expectations posited greater gender inequality in STEM fields, the findings suggest that disparities do not align neatly along a STEM vs. non-STEM divide. Instead, the results reveal substantial variation within STEM fields themselves, with only a few disciplines, such as physical life sciences and mathematics, exhibiting significant disparities. This underscores the importance of attending to disciplinary institutions and norms rather than treating STEM as a uniform category. Gender stereotypes, perceptions of scientific excellence, and field-specific institutional practices may all shape application behaviors and review outcomes in ways that vary across disciplines.

Although the data do not allow for causal inferences regarding the mechanisms underlying these disparities, several potential explanations emerge from the literature. First, academic rank is a strong predictor of grant success, and women remain disproportionately concentrated in the lower academic ranks, particularly in STEM fields (Howe-Walsh and Turnbull 2016; Van Miegroet et al. 2019). Second, gender stereotypes are more deeply entrenched in STEM disciplines, where women are often regarded less favorably in evaluative contexts such as hiring, promotion, and peer review (Ellemers 2018; Heilman et al. 2024). Third, the research has established that both male and female reviewers exhibit same-gender preferences during peer review (Helmer et al. 2017). Given that women are underrepresented on review panels in STEM, there may be fewer advocates for female applicants in these fields. Together, these factors may contribute to greater susceptibility to bias in certain STEM disciplines. Nevertheless, the findings do not suggest widespread or uniform bias across the broader academic landscape.

Overall, the results indicate that gender inequality in research funding is not pervasive but occurs at isolated points in the grant pipeline. There is evidence of a generally reasonable level of gender equity in grant application behavior and funding outcomes. Nonetheless, targeted interventions could help close the remaining gaps. Fields such as engineering, mathematics, humanities, and environmental sciences might benefit from initiatives designed to encourage greater submission rates among women. Lessons could also be drawn from fields such as medicine and physical life sciences, where women submitted proposals at rates exceeding expectations.

In the area of grant allocations specifically, the only field that exhibited disparities—physical life sciences—also has relatively limited female representation (13%). Future efforts to promote gender equity in this field may therefore require a dual focus: supporting women’s career advancement and addressing any subtle barriers in the grant evaluation process. If funding inequities persist, they may contribute to the long-term underrepresentation of women within these disciplines, reinforcing existing structural imbalances.

5. Conclusions

This study examined gender inequality in academic grant submissions and allocations in 12 academic fields in Israel, using data from nearly 5000 grant applications over two years. By analyzing both submission behavior and funding outcomes disaggregated by field, the study aimed to identify when and where gender-based disparities occur.

The findings suggest that there is no systematic or universal pattern of gender inequality across the grant process. Rather, isolated instances of inequality appear at specific stages of the process in particular fields. Women were underrepresented among grant applicants in four fields and overrepresented in two others. Mathematics stood out as the only field showing consistent gender disparities in both submission rates and requested amounts. Regarding grant allocations, only physical life sciences showed statistically significant gender gaps in win rates. However, no field exhibited significant differences in the proportion of funding awarded relative to the amount requested.

These findings challenge assumptions about widespread or uniform gender bias in grant funding. They also emphasize the importance of field-specific analysis. The few areas where gender disparities in allocation were observed were primarily in STEM fields. In these disciplines, women are particularly underrepresented, and structural barriers may play a more prominent role. Such barriers include gender stereotyping, the concentration of women in the lower academic ranks, and their lack of representation on peer-review panels.

Despite its strengths, this study has several limitations. First, the data come from only one country, underscoring the need to proceed with caution when attempting to generalize the results to other academic systems. Second, some fields—such as mathematics and agriculture—had small sample sizes, limiting the study’s statistical power and the ability to draw robust conclusions.

Although this study examines funding dynamics within a specific national context, recent political shifts in countries such as the United States, marked by challenges to DEI frameworks and changing support for academic fields, underscore the need for comparative research on how political climates shape gendered patterns in grant allocation. Exploring these dynamics across diverse contexts can shed light on the structural resilience, or vulnerability, of equity in academic funding.

Future research could expand this work by including more years of data, additional research funding organizations, and comparative analyses across a variety of countries. Further studies should also seek to uncover the mechanisms behind the observed inequalities, particularly why disparities emerge in certain fields or countries but not others. Qualitative or mixed-methods approaches might help illuminate how factors such as academic rank, cultural norms, peer-review dynamics, and institutional practices contribute to gendered outcomes in grant processes.

Understanding how structural factors shape outcomes and patterns of inequality would significantly deepen our knowledge of disparities in research-funding allocation. Two factors, in particular, warrant further investigation: academic rank and gender bias in the review process. Academic rank has consistently been shown to predict grant success; thus, examining how the gender distribution across ranks in different fields correlates with funding disparities could provide a more comprehensive picture. In addition, gender perceptions of researchers may play a critical role. In the Israeli context, as in many other funding systems, reviewers are asked to assess the competence and excellence of the PI. Because gender stereotypes vary across disciplines, it is important to understand whether and how these perceptions influence outcomes and contribute to inequality. Although both of these questions lie beyond the scope of this study, they are essential for future research aiming to explain when and why gender disparities emerge in the grant-allocation process.

Ultimately, understanding the field-specific dynamics of gender inequality in research funding is essential for developing targeted, evidence-based interventions that promote fairness and inclusion in academic careers. Future research should further investigate how the social and structural dynamics of different academic fields contribute to gendered outcomes in research funding.