This article presents a bibliometric analysis aimed at mapping scientific output on gender gaps and women's participation in STEM between 2023 and 2025. The methodology is grounded in the PRISMA model for systematic reviews and the Science Mapping Workflow for bibliometric analysis, employing tools such as Bibliometrix and Biblioshiny within the R environment. The search was conducted exclusively in the Web of Science database, retrieving a total of 2,117 records. After removing duplicates and applying thematic filtering, 1,400 articles were included in the final analysis. Among the main findings, an annual growth rate of 56% was observed between 2023 and 2024. The most frequent keywords were gender equality, STEM education, and inclusion. The United States, Brazil, Mexico, Spain, and China led scientific production. Academic collaboration networks showed high density in Latin America, Western Europe, and East Asia. However, only 26.4% of authorship corresponded to women, highlighting persistent gender inequality. Conceptual, intellectual, and social maps revealed five thematic clusters focused on digital inclusion, higher education, and technological ethics. This study provides empirical evidence to strengthen institutional strategies that promote gender equity in technology-related fields. Furthermore, it proposes a replicable approach for assessing the evolution of gender-focused scientific agendas in Latin American contexts.
The gender gap in Science, Technology, Engineering, and Mathematics (STEM) disciplines remains
a persistent challenge within higher education systems and global scientific production. Despite
numerous institutional and academic eforts to foster the inclusion of women in technology-related fields,
inequalities persist in both access and authorship representation in scientific publications, particularly
in areas such as computer science, engineering, and educational innovation [
In Latin America, this issue is especially relevant given the growing interest in implementing
institutional policies with a gender perspective and in highlighting the contributions of female researchers
to digital transformation processes, inclusive software design, and STEM pedagogies [
This article presents a bibliometric study analyzing the evolution of scientific literature between
2023 and 2025, focusing on gender gaps, women’s participation in STEM, educational inclusion, and
gender-sensitive technologies. The systematic review was based on the PRISMA methodology [
The findings provide quantitative evidence on country-level distribution, authorship patterns, frequent keywords, and collaboration networks, as well as the low participation of women as lead authors (26.4%). Additionally, the study identifies emerging thematic areas and ofers recommendations to strengthen gender equity in academic production. The applied approach is replicable across other regions or disciplines and contributes to building a more inclusive and evidence-based scientific agenda.
Female participation in STEM disciplines remains constrained by structural barriers, afecting access,
retention, and leadership within academia and industry [
In the Latin American context, LAWCC conferences have established a platform for disseminating
research on gender and computing. A bibliometric analysis of papers presented between 2008 and 2023
reveals sustained thematic growth in areas such as accessibility, artificial intelligence, digital diversity,
and inclusive education [
Bibliometric studies have proven to be efective tools for examining the evolution of scientific
communities, collaboration networks, and thematic structures. A recent investigation on process management
in indexed journals employed the Science Mapping Workflow model to represent structural and dynamic
aspects of scientific knowledge [
The use of tools such as Bibliometrix, developed in the R environment, enables automated analysis of
co-occurrences, social networks, conceptual evolution, and co-citation in scientific sources [
This article aligns with that line of work by conducting a bibliometric analysis of academic literature indexed in WoS between 2023 and 2025, focusing on gender gaps, inclusion in higher education, and the design of technologies with an equity-oriented approach.
The systematic review process was structured according to the PRISMA methodology, ensuring
traceability, transparency, and rigor in the selection of analyzed articles [
During the identification phase, a total of 2,117 records were retrieved from the WoS database, considering publications indexed between 2023 and 2025 in Spanish and English, as shown in Table 1. The search was conducted using an advanced strategy with Boolean operators and thematic identifiers related to gender gap, STEM, software design, and access to higher education, aligned with the Sustainable Development Goals (SDGs) 4, 5, 8, 10, and 16 [16]. The temporal delimitation was based on the specific interest of this study in evaluating recent trends in the literature on gender gaps in STEM ifelds, particularly in the post-pandemic context and in response to the growing international focus the topic has received over the past three years. This interval was selected to capture an updated and relevant sample that allows identifying recent changes and initiatives with potential influence on current academic production.
During the initial screening phase, 117 duplicate records were removed, leaving 2,000 unique documents for analysis. Subsequently, in the preliminary selection phase, 500 articles were excluded for not meeting the thematic criteria defined in abstracts and keywords, such as lack of direct relation to gender equity in technological higher education or absence of focus on academic inclusion in STEM environments.
In the eligibility evaluation phase, full-text documents were reviewed. An additional 100 records were excluded for not meeting the established methodological criteria, such as lack of empirical evidence, absence of a solid theoretical framework, or misalignment with the study objectives.
A classification of authors’ gender was performed, revealing that 133 authorships (equivalent to 3.9% of the total) did not provide explicit information on the author’s gender. To ensure the accuracy of the bibliometric analysis and avoid biases derived from uncertainty, the methodological decision was made to exclude these cases from gender-specific analyses, while retaining them for keyword-based analyses. Consequently, quantitative analysis and gender comparisons were conducted exclusively on the set of authors identified as male or female.
Finally, 1,400 scientific articles were included in the bibliometric analysis, forming the documentary base on which conceptual, intellectual, and social mapping techniques were applied using the Bibliometrix tool [15].
This section presents the findings obtained from the bibliometric analysis of the corpus comprising 1,400 publications indexed in WoS during 2023–2025. It is important to note that, since the data collection process was carried out on May 30, 2025, the results for that year reflect only the scientific output from the first five months. A sustained growth in annual production was observed: 366 publications in 2023 (26.1%), 572 in 2024 (40.9%), and 462 in 2025 (33%). This pattern reveals an interannual increase of 56% between 2023 and 2024. Regarding publication type, 73% of the documents correspond to articles in indexed journals, 21% to conference proceedings, and 6% to systematic reviews.
This visualization represents a keyword co-occurrence network structured into two main semantic communities. In Figure 1 , two clusters are identified: the first, in red, is led by the terms gender, women, and gap, and includes concepts related to structural inequality, discrimination, mental health, and unpaid work. The second, in blue, revolves around STEM, science, and education, grouping terms associated with academic performance, self-eficacy, and technical training. The most prominent nodes—gender with more than 350 connections and women with over 340—indicate high thematic centrality.
The analysis in Figure 2 represents a multiple correspondence factorial analysis that groups frequent terms into two semantic dimensions, which explain 73.28% of the total variance (Dim. 1: 61.66%, Dim. 2: 11.62%). The educational cluster (red) includes terms such as “students,” “STEM,” and “achievement,” while the socioeconomic cluster includes “inequality,” “family,” and “segregation.” This thematic segmentation is key to defining priority research agendas.
The educational cluster is linked to SDG 4, while the structural-labor cluster reflects SDGs 5 and 8.
This segmentation demonstrates how the literature addresses inequalities from a gender perspective.
As UNESCO notes, an inclusive digital transformation is essential to closing these gaps [
The treemap visualization shown in Figure 3 hierarchically presents the most frequent occurrences in the analyzed corpus (2023–2025). The terms gender (359), women (348), and STEM (221) stand out, reflecting the focus on gender equity in education and technology. They are followed by science (213), education (211), and gap (188), consolidating a narrative centered on female participation in STEM. This figure complements the semantic analysis in Figure 4 and integrates with the thematic evolution represented in Figure 6.
The hierarchical dendrogram groups 61 key terms into three well-defined macro-clusters, as shown in Figure 4. The structural-labor cluster represents approximately 38% of the total and includes concepts related to inequality, income, and participation gaps. The educational axis comprises 34% of the terms and focuses on STEM training, student performance, and teaching-learning processes. Finally, the psychosocial and identity cluster covers 28% of the terms, addressing topics such as mental health, self-eficacy, and discrimination. This segmentation confirms a complex thematic structure consistent with the results of the factorial analysis in Figure 2.
The chart in Figure 5 illustrates the cumulative evolution in the frequency of the 10 most relevant terms in the analyzed corpus between 2023 and 2025. A sustained increase is evident in concepts such as “impact” (over 340 mentions), “students” (330), and “science” (280), followed closely by “education” (270) and “STEM” (260), reflecting a strong orientation toward educational, scientific, and formative dimensions. Likewise, terms such as “inequality,” “gap,” and “women” show constant growth, reaching between 160 and 200 occurrences.
Figure 6 illustrates the international collaboration network in the field of gender and STEM, showing
geographic nodes and bilateral connections. The United States leads scientific collaborations with more
than 180 foreign institutions, followed by Spain (≈ 150 ) and Brazil (≈ 140 ). Latin America stands
out with Mexico, Chile, Colombia, and Costa Rica as active countries in co-authorship networks. This
collaboration pattern confirms the growing internationalization of the field, particularly in inclusive
agendas and educational policies. In line with [
A stacked bar chart is presented in Figure 7, diferentiating the type of scientific collaboration by country: SCP (Single Country Publications) and MCP (Multiple Country Publications). The United States leads with more than 550 documents and is also the country with the highest proportion of international collaboration publications (around 30% MCP). China and Spain follow, each with more than 250 documents, although with a lower proportion of MCP, indicating a strong capacity for internal production. The prominent presence of Spain, Chile, Mexico, and Brazil among the top 15 countries reflects the growing positioning of Latin America in gender and STEM research.
Figure 8 presents the most cited sources in the corpus analyzed between 2023 and 2025. A high concentration of citations is observed in journals specializing in sociology, psychology, and gender studies. Notable examples include American Sociological Review with 1,139 citations, Journal of Marriage and Family (805), PLOS ONE (760), and Gender Society (738). This pattern confirms the academic robustness of the corpus and its reliance on high-impact indexed publications, reinforcing the scientific validity of the findings obtained in this research.
The Sankey diagram in Figure 9 illustrates the connection between countries of origin (AU_CO), authors (AU), and the most frequent thematic terms (DE) in the analyzed corpus. Dominant flows are identified from China and the United States toward authors such as Li Y., Zhang X., and Wang J., who address topics such as “gender diferences,” “STEM,” and “higher education.” This visualization provides insight into the thematic and geographic distribution of academic production, highlighting the high concentration of authorial leadership in Asia and North America, as well as alignment with topics on gender, equity, and technology. It facilitates the recognition of structural patterns in global academic networks.
This section interprets the main findings of the bibliometric analysis and contrasts them with previous studies to identify implications, limitations, and opportunities for improvement in the research agenda on gender and STEM.
Between 2023 and 2025, a 56% increase was observed in scientific production on topics related to gender equity, educational inclusion, and technological design with a focus on female authorship. This trend reflects growing institutional and academic attention to these issues, particularly in Latin America, where universities such as the University of Chile and UNAM led in publication volume. However, female participation as lead authors remains limited, representing only 26.4%.
The results are consistent with previous studies presented at LAWCC, such as the mapping conducted
up to 2023 of all articles presented at the conference [
Afirmative actions are needed to promote women’s participation in STEM research, including mentoring policies, targeted funding, and visibility in scientific spaces. Experiences such as those of Pacheco and Álvarez [17] and Huston et al. [18], focused on institutional initiatives to reduce the gender gap, show that strategic and financial support can positively influence the retention and advancement of women in academic and professional careers. In Costa Rica, various initiatives aim to reduce the gender gap in STEM and strengthen female participation in research, such as the national campaign “Deciding My Future,” developed by the Ministry of Public Education, the Inter-American Development Bank, and Innovations for Poverty Action [19] , as well as the Women’s Network in Science, Engineering, and Humanities (CIHRED-UCR) at the University of Costa Rica [20].
This analysis was based exclusively on articles indexed in WoS, excluding gray literature and other relevant repositories in the Latin American context. Furthermore, the review was limited to the period 2023–2025, considering publications up to May of the last year. Future research could expand the temporal scope and combine quantitative approaches with qualitative methods for a deeper analysis.
It is recommended to apply this bibliometric approach to other scientific disciplines or geographic regions to identify emerging patterns with a gender perspective. Incorporating altmetric analyses and evaluating social impact indicators would also be pertinent, as well as exploring mentoring networks and female leadership in science and technology. The replicability of the methodology applied in this study constitutes a significant contribution to the systematization of inclusive scientific agendas. Despite the increase in scientific production on gender and STEM topics, significant structural gaps persist in female participation and leadership, even in institutional contexts that promote equity.
Recent studies show that, in Latin America, mentoring strategies, institutional support, and role
models significantly influence the reduction of gaps, although they do not completely overcome historical
barriers [
This bibliometric study enabled a rigorous analysis of the recent evolution of scientific production related to the gender gap in STEM, inclusive software design, and equitable access to higher education between 2023 and 2025. Based on a systematic review using the PRISMA methodology and the Science Mapping Workflow model, 2,117 publications were identified, of which 1,400 met the inclusion criteria and were processed using the Bibliometrix and Biblioshiny tools.
The results show sustained growth in scientific output, with a 56% increase between 2023 (366 articles) and 2024 (572 articles). This rise reflects greater academic attention to the issue of gender inequality in scientific and technological environments. The most frequently addressed topics included gender equality, STEM education, higher education, women in computing, and inclusion, representing more than 70% of the most common keywords. Five thematic clusters were identified, with the densest clusters related to digital inclusion, gender-focused public policies, and educational innovation.
Despite these advances, a low representation of women in authorship was confirmed, with only 26.4% of the total 3,412 identified authorships. In contrast, 69.7% corresponded to men, and 3.9% could not be classified. This finding confirms the persistence of structural gender gaps, even in research addressing equity-related topics.
Regarding geographic distribution, the United States led production with 198 publications, followed by China (156), Brazil (212 total publications), Mexico (187), and Spain (174). Costa Rica, although with a smaller volume, contributed 38 publications in international collaboration, positioning itself as an emerging actor in the region.
Co-authorship networks revealed active participation of Latin American institutions, although collaboration density among female researchers was low. The proportion of publications with international collaboration (MCP) was notable in countries such as the United States (69 MCP publications) and China (44 MCP), evidencing their capacity for global scientific articulation.
This article proposes replicating the applied approach as a strategic tool for monitoring institutional policies and shaping scientific agendas with a gender perspective. Strengthening mentoring programs, targeted funding, and afirmative actions to enhance female participation in academic publications is recommended. Furthermore, future research could integrate qualitative and altmetric analyses to capture broader dimensions of the impact of gender-focused science.
For future research, the aim is to expand bibliographic sources by incorporating databases such as AMC, EBSCO, Google Scholar, IEEE Xplore, RedALyC, ScienceDirect, Scopus, and SciELO, with the objective of improving regional coverage and reducing geographic and disciplinary bias resulting from the exclusive use of WoS.
This article proposes replicating the applied approach as a strategic tool for monitoring institutional policies and shaping scientific agendas with a gender perspective. Strengthening mentoring programs, targeted funding, and afirmative actions to enhance female participation in academic publications is recommended. Furthermore, future studies could integrate qualitative and altmetric analyses to capture broader dimensions of the impact of gender-focused science.
Based on this study, several future research lines are proposed, aimed at evaluating the impact of digital systems on institutional eficiency, their replicability in other university campuses, and the integration of these developments into training processes with an applied learning approach. Additionally, it is suggested to deepen strategies that promote women’s participation in technological projects and to develop gender-sensitive metrics to monitor their academic contribution. Comparative studies between campuses or universities are also recommended to validate the results obtained and enrich the proposed intervention models. Finally, these future works could serve as input for designing university policies aligned with the Sustainable Development Goals, particularly regarding gender equality (SDG 5), quality education (SDG 4), institutional innovation (SDG 9), and capacity building in public management (SDG 16).
During the preparation of this work, the authors used ChatGPT for grammar and spelling correction, and Copilot for translation. After using these tools, the authors reviewed and edited the content as necessary and assume full responsibility for the content of the publication. doi/10.1002/asi.21525. [15] M. Aria, C. Cuccurullo, bibliometrix: An R-tool for comprehensive science mapping analysis, Journal of informetrics 11 (2017) 959–975. URL: https://www.sciencedirect.com/science/article/ abs/pii/S1751157717300500. [16] D. o. E. United Nations, S. Afairs, The 17 goals | sustainable development, https://sdgs.un.org/goals, 2025. Accedido: 2025-05-15. [17] C. J. Pacheco Figueroa, M. A. Alvarez Lemus, Mentoring women in STEM: empowering through social technologies for enhanced inclusivity and professional growth. A case study, in: Frontiers in Education, volume 10, Frontiers Media SA, 2025, p. 1512143. URL: https://www.frontiersin.org/ journals/education/articles/10.3389/feduc.2025.1512143/full?utm_source=chatgpt.com. [18] W. M. Huston, C. G. Cranfield, S. L. Forbes, A. Leigh, A sponsorship action plan for increasing diversity in STEMM, Ecology and Evolution 9 (2019) 2340–2345. URL: https://pmc.ncbi.nlm.nih. gov/articles/PMC6405484/?utm_source=chatgpt.com. [19] Innovations for Poverty Action, Deciding my future: Evaluating the impact of text messaging on female STEM enrollment in Costa Rica, https://poverty-action.org/ deciding-my-future-evaluating-impact-text-messaging-female-stem-enrollment-costa-rica, 2023.
Accedido: 2025-08-09. [20] Universidad de Costa Rica, La Red de Mujeres en Ciencias, Ingenierías y Humanidades (CIHRED-UCR) ya es una realidad, https://www.ucr.ac.cr/noticias/2025/2/18/ la-red-de-mujeres-en-ciencias-ingenierias-y-humanidades-cihred-ucr-ya-es-una-realidad.html, 2025. Accedido: 2025-08-09.