Introducing “ELLAS Survey Dataset” an open resource about factors that influence career interest and leadership in STEM in Bolivia, Brazil, and Peruo - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Data Brief . 2026 Mar 25;66:112721. doi: 10.1016/j.dib.2026.112721 Search in PMC Search in PubMed View in NLM Catalog Add to search Introducing “ELLAS Survey Dataset” an open resource about factors that influence career interest and leadership in STEM in Bolivia, Brazil, and Peruo Indira R Guzman Indira R Guzman a California State Polytechnic University, Pomona, USA Find articles by Indira R Guzman a, ⁎ , Boris Branisa Boris Branisa b Escuela de la Producción y la Competitividad, Universidad Católica Boliviana “San Pablo”, La Paz, Bolivia Find articles by Boris Branisa b , Guillermo Davila Guillermo Davila c Universidad de Lima, Peru Find articles by Guillermo Davila c , Florencia Sánchez-Guillén Florencia Sánchez-Guillén b Escuela de la Producción y la Competitividad, Universidad Católica Boliviana “San Pablo”, La Paz, Bolivia Find articles by Florencia Sánchez-Guillén b , Leihge Roselle Rondon Pereira Leihge Roselle Rondon Pereira d Federal University of Mato Grosso, Mato Grosso do Sul, Brazil Find articles by Leihge Roselle Rondon Pereira d , Luana Mendes Luana Mendes d Federal University of Mato Grosso, Mato Grosso do Sul, Brazil Find articles by Luana Mendes d , Luciana Bolan Frigo Luciana Bolan Frigo e Federal University of Santa Catarina, Florianópolis, Brazil Find articles by Luciana Bolan Frigo e , Nadia Rodríguez-Rodriguez Nadia Rodríguez-Rodriguez c Universidad de Lima, Peru Find articles by Nadia Rodríguez-Rodriguez c , Silvia Amélia Bim Silvia Amélia Bim f Federal University of Technology, Paraná, Curitiba, Brazil Find articles by Silvia Amélia Bim f , Cristiano Maciel Cristiano Maciel d Federal University of Mato Grosso, Mato Grosso do Sul, Brazil Find articles by Cristiano Maciel d Author information Article notes Copyright and License information a California State Polytechnic University, Pomona, USA b Escuela de la Producción y la Competitividad, Universidad Católica Boliviana “San Pablo”, La Paz, Bolivia c Universidad de Lima, Peru d Federal University of Mato Grosso, Mato Grosso do Sul, Brazil e Federal University of Santa Catarina, Florianópolis, Brazil f Federal University of Technology, Paraná, Curitiba, Brazil ⁎ Corresponding author. [email protected] Received 2026 Jan 27; Accepted 2026 Mar 22; Collection date 2026 Jun. © 2026 The Author(s) This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). PMC Copyright notice PMCID: PMC13090667 PMID: 42007086 Abstract The persistent gender gap in Science, Technology, Engineering, and Mathematics (STEM) fields, particularly in leadership roles, remains a pressing global challenge, including in Latin America. Understanding the factors that influence individuals’ interest in STEM careers and their motivation to assume leadership positions is essential for advancing gender equity and inclusive innovation. Although existing studies have explored these topics, most rely on qualitative methods, small samples, or single-country analyses, limiting regional comparability. In contrast, this manuscript introduces a comprehensive, openly accessible dataset and a detailed multilingual data dictionary that enable systematic comparisons across three Latin American countries, representing an innovative contribution to the study of gender and STEM in the region. We present the methodology and data structure of a large-scale, multi-country survey conducted in Bolivia, Brazil, and Peru, each with >3000 respondents. The study forms part of the Equality in Leadership for Latin America STEM (ELLAS) research network, funded by the International Development Research Centre (IDRC), which promotes the principles of open science and open data. We describe the survey design, ethical protocols, and data curation procedures implemented to ensure transparency and methodological rigor. The resulting ELLAS Survey Dataset offers one of the most extensive quantitative resources on gender and STEM in Latin America. Our goal is to make the ELLAS data accessible to the research community for further analysis. This article aims to guide researchers interested in using the dataset and to support the generation of evidence that can inform policies fostering women’s participation and leadership in STEM fields. Keywords: Women in STEM, Latin America, Career choice, Motivation to lead Specifications Table Subject Social Sciences Specific subject area Gender Studies in STEM; Women Leadership. Type of data CSV data file Raw Data collection Data were collected via a large-scale multilingual (Spanish/Portuguese) survey administered to STEM and non-STEM students and professionals in Bolivia, Brazil and Peru using a purposive sample. Only eligible, consenting respondents were included with ethics approvals in each country. The survey instrument consisted of 136 questions from validated scales from the literature addressing individual, academic, occupational, social, familial, and socioeconomic factors. It incorporated 23 variables, 123 indicators, and demographic variables, including gender, age, occupation, area of work, and country. Responses in Peru and Brazil were collected online, while the responses in Bolivia were collected in person. Data source location Bolivia (16.2902° S Latitude and 63.5887° W. Longitude), Brazil (14.2350° S Latitude and 51.9253° W. Longitude) and Peru (9.189967° S Latitude and 75.015152° W. Longitude). Data accessibility Repository name: ELLAS Platform: https://plataform.ellas.ufmt.br/open-data/1 Data identification number: N/A Direct URL to ELLAS Survey Dataset: https://plataform.ellas.ufmt.br/survey/surveyELLAS_2025–11–21_.csv ; ELLAS Survey Data Dictionary: https://plataform.ellas.ufmt.br/survey/Data_dictionary-2025-11-21.pdf Related research article N/A Open in a new tab 1. Value of the Data • The dataset supports the examination of individual and contextual factors associated with career interests and leadership across STEM and non-STEM fields in urban areas of Bolivia, Brazil, and Peru. With >3000 responses per country, the dataset allows for robust cross-national and demographic analyses that extend beyond the scope of single-country studies. • The dataset is grounded in established best practices in survey research, including the use of validated psychometric instruments and standardized data collection procedures. Surveys were translated into Spanish, Portuguese, and English using back-translation, ensuring cultural equivalence, transparency, and reproducibility for high-quality secondary analyses. • Items apply to both academic and professional contexts, enabling reuse for subgroup analyses by career stage, sector, discipline, gender, or country, as well as focused studies on constructs such as mentorship or leadership attitudes. • The dataset supports evidence-based policies and applied research aimed at addressing gender gaps and structural barriers to women’s advancement in STEM leadership. 2. Background The compilation of this dataset was motivated by the persistent gender disparities in STEM fields and leadership roles across Latin America [ 1 ], and the scarcity of large-scale of large-scale quantitative about underlying contextual mechanisms [ 2 ]. Globally, gender imbalances in STEM are well-documented with women representing 30% of researchers worldwide and 24% of technology leaders [ 3 , 4 ]. These disparities constrain scientific diversity and perpetuate economic inequality because STEM careers drive technological progress and high-income employment [ 1 , 5 , 6 ]. Latin America replicates and intensifies these patterns through structural, cultural, and institutional barriers. Women represent 48% of the regional workforce [ 7 ] yet comprise less than one-third of STEM professionals, with significant underrepresentation in engineering and computing [ 8 , 9 ]. Only 30% of STEM graduates are women [ 10 ], and the topic remains under researched [ 11 ]. Women hold under 20% of senior management positions in public institutions and <10% of corporate board seats [ 12 ]. The ELLAS survey builds on a mixed-methods study that integrated systematic mapping of peer-reviewed literature from the past twelve years with grey literature from UNESCO, OECD, and national research agencies [ 13 ]. This synthesis from diverse countries and methodological traditions informed a comprehensive taxonomy that guided the construction of the survey instrument. 3. Data Description 3.1. Description of the ELLAS dataset The ELLAS survey dataset created as part of the ELLAS Latin American Open Data for gender-equality Policies Focusing on Leadership in STEM project [ 14 ], comprises 10,311 observations and 220 indicators, documented through a comprehensive data dictionary. The dataset integrates two types of information: participant-reported data and survey-company coded data. For example, while participants reported their year of birth, the survey company recoded this information into generational categories. Researchers can download two complementary files: • ELLAS Survey Dataset • ELLAS Survey Data Dictionary To enhance data quality and analytical transparency, an additional indicator “problem_case” was introduced following an initial data review. This variable flags records that may contain unrealistic, missing, or internally inconsistent values, enabling researchers to decide whether to include or exclude such cases based on their research objectives. We recommend removing observations coded as 1 in the problem_case indicator for standard analyses. 3.2. Criteria for identifying problem cases A record was coded as problem_case = 1 if any of the following conditions were met: • The number of children in any single age group (0–3, 4–6, 7–11, or 12–17) was five or more. • The total number of children under 18 was nine or more. • Educational milestones were achieved at unusually early ages: ○ College degree before age 21. ○ Master’s degree before age 23. ○ Doctoral degree before age 26. ○ Postdoctoral qualification before age 30. • The difference between current age and years of work experience was 15 years or less. • A master’s, doctoral, or postdoctoral degree was reported without prior completion of a college degree. All other records were coded as problem_case = 0. The above-described rules were designed to identify potentially inconsistent demographic or educational trajectories. Researchers may apply additional exclusion criteria depending on their specific analytical focus (e.g., excluding cases where degree areas were not specified). Participants’ ages were calculated using their self-reported year of birth and 2024, the year in which the survey closed. Some inconsistencies arose due to language-specific survey versions. For instance, the category “Public Administration and Defense ” appeared in the Spanish surveys for Peru and Bolivia, whereas the Portuguese survey for Brazil listed only “Public Administration.” 3.3. Demographics The dataset includes key demographic variables such as sex at birth, year of birth, marital status, STEM identification, and IT vs. non-IT area of study across Bolivia, Brazil, and Peru. Final survey results are presented in Table 1 . Table 1. Demographic aspects regarding sex at birth, year of birth, and marital status. Demographic Aspects Bolivia Brazil Peru Total Sex at Birth Count Percent Count Percent Count Percent Count Percent Female 1594 50% 1916 51% 1666 50% 5176 50% Male 1595 50% 1872 49% 1668 50% 5135 50% Grand Total 3189 100% 3788 100% 3334 100% 10,311 100% Year of Birth Range Count Percent Count Percent Count Percent Count Percent 1997–2005 1196 38% 1348 36% 1201 36% 3745 36% 1981–1996 1435 45% 1456 38% 1173 35% 4064 39% 1965–1980 478 15% 501 13% 480 14% 1459 14% 1945–1964 80 3% 483 13% 480 14% 1043 10% Grand Total 3189 100% 3788 100% 3334 100% 10,311 100% Marital Status Count Percent Count Percent Count Percent Count Percent Single 2386 75% 1771 47% 2193 66% 6350 62% Married 670 21% 1796 47% 1002 30% 3468 34% Divorced 110 3% 199 5% 110 3% 419 4% Widowed 23 1% 22 1% 29 1% 74 1% Grand Total 3189 100% 3788 100% 3334 100% 10,311 100% STEM career Count Percent Count Percent Count Percent Count Percent STEM career 2420 76% 1750 46% 2426 73% 6596 64% Non STEM career 769 24% 2038 54% 908 27% 3715 36% Grand Total 3189 100% 3788 100% 3334 100% 10,311 100% IT Area of study Count Percent Count Percent Count Percent Count Percent IT 1599 50% 1442 38% 1521 46% 4562 44% Non IT 1590 50% 2346 62% 1813 54% 5749 56% Grand Total 3189 100% 3788 100% 3334 100% 10,311 100% Open in a new tab The variable d13_d_being_in_stem_career captures whether participants self-identified as being in a STEM career (1 = Yes; 0 = No), based on an explicit definition provided in the survey. In contrast, the variable g7_d_it_area was coded by the research team to identify participants in Information Technology (IT) fields. This classification was based on reported academic degrees at the undergraduate, master’s, or doctoral level. For example, respondents indicating Technology Systems Engineering at any level were classified as IT professionals (1 = IT; 0 = Non-IT). 3.4. The data dictionary The data dictionary is available as a separate document that researchers can download from the ELLAS platform website ( link here ). It provides detailed metadata for each of the 220 indicators, including multilingual survey questions and response options. Table 2 describes each column of the data dictionary. Table 2. Data dictionary columns. Column Name Description Factor type Conceptual category to which the variable belongs to group variables according to their content or analytical purpose. Variable Name Full and descriptive name of the variable Variable Detailed Code Long code containing the factor type, variable name and type to uniquely identify the variable. Useful for analysis without the need for a dictionary. Examples given, in Table 8 . Questions in English Question translated into English as presented in the instrument. Questions in Portuguese Question translated into Portuguese as presented in the instrument. Questions in Spanish Question translated into Spanish as presented in the instrument. Possible Answers English Translation of the answer options into English. Possible Answers Portuguese Translation of the answer options into Portuguese. Possible Answers Spanish Translation of the answer options into Spanish. Observations Relevant notes on the variable, such as special coding instructions, warnings on its use, the need for reverse coding, or other. Open in a new tab 3.5. Variable detailed codes The following Table 3 explains the letters used in the variable names, which indicate the type of variable and provide a brief description. To illustrate their application, we also include a second table, Table 4 , with examples showing how these codes were used in naming the variables. The number after the first letter reflects survey item numbering. Table 3. Variable names nomenclature. Code Meaning/ Type t Open text n Numeric d Dummy l Likert scale 1 to 7 c Categorical R Reverse coding suggested X Complexly determined item Open in a new tab Table 4. Variable names nomenclature examples. Variable Detailed Code Example Initial Variable Type Code Variable Type Code d5_t_state_bol d =Demographic t =text bol= Bolivia bra= Brazil per= Peru d1_n_birth_year d =Demographic n =number d2_d_sex d =Demographic d =dummy d9_3_1_c_college_area d =Demographic c =categorical b2_d_mentor b =Academic factor d =dummy b2_1_l_mentorship_quality b =Academic factor l =Likert e3_8_l_R_women_stereotypes e =Social factors l =Likert R = Reverse code suggested f2_5_l_X_parental_expect f =Family factors l =Likert X = Complexly determined item Open in a new tab 4. Experimental Design, Materials and Methods In this section we describe how the ELLAS survey instrument was developed. Most of the survey items were adapted from existing instruments to ensure cultural relevance within the Latin American context and applicability across both academic and professional settings. For example, items that originally referred to a “place of study” were revised to reference either a place of study or a place of work, allowing both students and professionals to respond appropriately. Some items, however, were retained in their original form. For instance, the scale measuring Attitudes Toward Women as Managers was used without modification. The survey was then carefully translated into Spanish and Portuguese by native-speaking researchers and pre-validated with academics and practitioners to ensure semantic clarity and cultural appropriateness. Additionally, variables such as Perceived Gender Inequality were introduced to capture dimensions of women’s STEM experiences that are particularly salient in Bolivia, Brazil, and Peru. Survey items for these variables were developed based on relevant scholarship and refined through expert consultation to ensure contextual relevance and cultural appropriateness. The survey instrument can be classified into five sets of variables: Contextual factors, STEM Career interest, Motivation to lead, Individual Perceptions about Gender Policies, and Control Variables. 4.1. Contextual factors The synthesis identified six interrelated contextual factors: individual, academic, work-related, social, family-related, and socioeconomic , encompassing 196 specific sub-factors [ 13 ]. Table 5 presents the definitions of these six contextual factors based on the ELLAS taxonomy, along with the proxy variables adapted from the literature and operationalized in the ELLAS survey with the respective references from the literature. For each contextual factor , the literature was reviewed to identify previously validated survey instruments. Table 5. Operationalization of contextual factors. Factor Definition Variable Code Number of Items Operational Definition of Variable Source References for Scale Adaptation Individual Factors: internalized beliefs, attitudes, and perceptions that shape women’s engagement with STEM education and careers. A1. Sense of not Belonging 3 Feelings of being excluded and ignored. Zadro, Williams & Richardson (2003) [ 15 ] A2. Self-Efficacy 4 People’s own beliefs in their ability to accomplish given tasks in STEM fields. Withanaarachchi & Vithana (2022) [ 16 ] A3. Self-Confidence 6 Individual’s perceptions of their own competences, intelligence and success relative to others. Lam et al. (2008) [ 17 ] A4. Financial Independence 4 Individual's ability to autonomously manage their financial responsibilities and sustain their livelihood without reliance on others. Bea & Yi (2019) [ 18 ] A5. Female Gender Identity 2 The extent to which an individual identifies with typical female behaviors. Patterson (2012) [ 19 ] A6. Perceived gender inequality 4 Individual's perceptions that women experience fewer opportunities, freedoms, or privileges than men. Kinias & Kim (2012), Liss (1975). [ 20 ] Later aligned with Schwartz-Salazar et al. (2024) [ 21 ] A7. Bullying 5 Individual's perceptions of repeated intimidating and exclusionary behaviors. Ranf et al. (2006) [ 22 ] A8. Work meaningfulness 3 The degree to which individuals perceive their work as meaningful. Spreitzer (1995) [ 23 ] Academic Factors: elements of the educational environment, within and beyond the classroom, that influence women’s participation and progression in STEM fields. B1_1. Stem Career Stereotypes Nerd 6 Participant’s stereotypes about people in STEM careers. Nerd dimension. Starr (2018) [ 24 ] B1_2. Stem Career Stereotypes Genius 4 Participant’s stereotypes about people in STEM careers. Genius dimension. Starr (2018) [ 24 ] B2. Quality of Mentorship 3 Individual's perceptions about the quality of the mentorship received. Xu & Payne (2014) [ 25 ] B3. Perceived Institutional Gender Equality 5 Individual's perception of the existence or application of gender equality. García-Holgado et al. (2020) [ 26 ] B4. Peer-group Interaction 7 Individual's own satisfaction by the development of interpersonal relationships. French & Oakes (2004) [ 27 ] Work Related: structural and cultural characteristics of workplace environments that shape women’s professional experiences and opportunities in STEM fields. C1. Organizational Culture 5 Individual's perception of the set of norms, beliefs, principles, and ways the organization promotes gender equality. Withanaarachchi & Vithana (2022) [ 16 ] C2_1. Empowerment in the workplace - Self-determination 3 Autonomy over the initiation and continuation of work behavior and processes. Spreitzer (1995) [ 23 ] C2_2. Empowerment in the workplace - Impact 3 The degree to which one can influence strategic, administration, or operating outcomes in one's department or organization. Spreitzer (1995) [ 23 ] Social Factors: broader societal and cultural influences that shape women’s attitudes and aspirations regarding STEM careers. E1. Sexism 6 Perceived hostile, offensive, and sexist attitudes towards my gender. Raver & Nishii (2010) [ 28 ] E2. Attitudes towards Women as Managers 7 Individuals’ perception of the prejudiced or stereotyped assumptions that associate effective leadership qualities of women. Terborg et al. (1977) [ 29 ] E3. Stereotypes Related to Women 8 Individuals' perceptions of culturally shared beliefs about the traits and roles typically attributed to women. Castillo-Mayén & Montes-Berges (2014) [ 30 ] Family Factors: expectations, support, and influences provided by family members that can either encourage or discourage women’s participation in STEM fields. F1. Family Support 6 Individual's perception of support and affection received from their relatives. Nava Quiróz et al. (2015) [ 31 ] F2. Parental Expectations 6 Individuals’ perceptions about beliefs, attitudes, and communicated standards that their parents hold regarding their future educational and career paths. Keller & Whiston (2008) [ 32 ] Open in a new tab After identifying the factors and sub-factors in the taxonomy [ 13 ], survey variables were selected based on empirical robustness and demonstrated relevance. For instance, although the taxonomy includes 28 academic sub-factors, the ELLAS survey focuses on four: STEM Career Stereotypes (Nerd and Genius), Quality of Mentorship, Perceived Institutional Gender Equality , and Peer-group Interaction . These were prioritized because the systematic mapping identified them as both highly influential and actionable within institutional settings. Work-related factors , included a total of 17 sub-factors identified in the mapping study, but only three were selected ( Organizational Culture and Empowerment in the workplace and Self-determination and Impact) as focal points for the ELLAS survey. Once again, these factors were selected because they align with key themes found across both the academic and grey literature reviews: namely, the importance of inclusive organizational norms and the availability of empowerment opportunities for women. Our operationalization of these variables builds upon validated scales, but then the survey items were adapted to reflect the specific challenges and realities of the Latin American context, particularly the underrepresentation of women in leadership roles. The ELLAS survey includes social factors , variables such as Sexism, Attitudes Towards Women as Managers , and Stereotypes Related to Women , all of which were repeatedly identified in the systematic mapping study as significant obstacles to equity in STEM. These factors are particularly salient in Latin American societies, where traditional gender norms and “machismo” attitudes often remain deeply rooted and impact both professional and educational trajectories. Finally, family-related factors include two variables, Family Support and Parental Expectations . These factors are crucial in shaping girls’ early interests, academic choices, and professional aspirations. This selection reflects both the empirical importance of family influence identified in the literature and the need to understand how cultural and socio-economic conditions in Bolivia, Brazil, and Peru shape family attitudes toward STEM careers and leadership roles for women. Most of the scales were sourced from literature published in English, with a subset derived from articles written in Spanish. To ensure linguistic and cultural appropriateness across the target populations, all scales were translated into Spanish and Portuguese, followed by a back-translation process to verify the accuracy and clarity of the items in each language. For instance, the scale used to assess Quality of Mentorship was adapted from Xu and Payne [ 25 ]. An additional survey item was included to first determine whether respondents had a mentor; then they were asked to evaluate the quality of that mentorship . This adaptation ensured a more accurate and context-sensitive assessment of mentorship experiences. 4.2. Stem career interest In our study, STEM Career Interest was measured using a set of five items adapted from the instrument developed by Luo et al. [ 33 ], which conceptualizes career interest as students’ general inclination toward pursuing a STEM-related profession ( Table 6 ). The selected items assess participants’ aspirations and preferences related to STEM careers through statements such as “I hope my future job is STEM-related” and “My dream career is STEM-related.” These items emphasize a holistic view of STEM career interest in referencing broad occupational categories and future job characteristics, rather than discrete disciplines. Table 6. Interest in STEM career. Variable Code Number of items Variable Operational Definition Survey Instrument Source of survey items (adapted from) G1. Interest in STEM Career 5 Interest in STEM Career Individuals’ perception to the degree of personal enthusiasm, aspiration, and motivation towards pursuing professional roles within the STEM fields. Luo et al. (2021) [ 33 ] Open in a new tab 4.3. Motivation to lead The Motivation to lead construct was defined by Chan and Drasgow [ 34 ] as the “individual differences construct that affects a leader’s or leader-to-be’s decisions to assume leadership training, roles, and responsibilities, and that affect their intensity of effort at leading and persistence as a leader” (p. 482). In the ELLAS Survey, Leadership Motivation was assessed using the model of motivation to lead developed by Chan and Drasgow [ 34 ], encompassing affective-identity, social-normative , and non-calculative components, Table 7 . Table 7. Variables related to leadership trajectories. Variable Code Number of items Variable Operational Definition Survey Instrument Source of survey items (adapted from) G2_1. Motivation to Lead: Affective Identity 9 Motivation to Lead The extent of which people like to lead and see themselves as having leadership qualities, generally have past leadership experience, and are confident in their own leadership abilities. Chan & Drasgow (2001) [ 34 ] G2_2. Motivation to Lead: Social Normative 9 Motivation to Lead The extent of which people are motivated to lead by a sense of social duty and obligation. Chan & Drasgow (2001) [ 34 ] Open in a new tab 4.4. Individual perceptions about gender policies The survey instrument presented in this study includes a structured set of items designed to measure three interrelated variables essential to understanding organizational dynamics affecting gender equality in STEM and existing gender policies, Table 8 : the perceived importance of gender equality policies, the perceived existence of such policies within organizations, and the experienced consequences of their absence. The first measure captures respondents’ normative evaluations of gender-related initiatives, such as equal pay, mentoring, and leadership opportunities for women. The second assesses respondents’ awareness of formal organizational policies promoting gender equity. The third captures the lived effects of policy absence, including experiences of inequity, harassment, or limited career advancement. Although these measures are novel and lack established reliability metrics, they provide an integrated framework for examining the alignment between policy ideals, organizational practices, and individual experiences in STEM contexts. Table 8. Individual perceptions about gender policies. G3 Policy Importance G4 Policy Existence G5 Policy Barriers I consider the following policies totally relevant: In my organization, there are explicit policies in place to: In my experience, I have suffered the consequences of my organization not: Guarantee equal pay for equal work. Ensure equal pay for equal work. Having guaranteed equal pay for equal work. Preserving work-life balance Preserve work-life balance. Preserving work-life balance. Eradicating sexual harassment and violence Eradicate sexual harassment and violence Having sexual harassment and violence eradicated. Ensuring gender parity for leadership/political positions Ensure gender parity in leadership/political positions Guaranteeing gender parity in leadership/political positions. Ensuring women's participation (gender quota) Ensuring women's participation (gender quota) Guaranteeing women's participation (gender quota) Attracting women's participation in STEM fields Attracting the participation of women in STEM fields Attracting women's participation in STEM fields. Organize mentoring programmes for young women (how to meet female engineers, etc.) Organize mentoring programmes for young women (how to meet female engineers, etc.) Organizing mentoring programmes for young women (such as meeting female engineers, etc.) Demonstrate affirmative action, such as hiring women or minorities from excluded social groups Demonstrate affirmative action, such as hiring women or minorities from excluded social groups Implementing affirmative action, such as hiring women or minorities from excluded social groups. Retain and train women in STEM fields Retain and train women in STEM fields Retaining or empowering women in STEM fields. Support women's professional growth Support women's professional growth Supporting women's professional growth. Providing financial support to female students in STEM fields Provide financial support for female students in STEM fields Providing financial support for female students in STEM fields. Open in a new tab 4.5. Control variables The ELLAS Survey also included some control variables that may have an impact on entry, permanence, and leadership of people in STEM areas [ 35 ]. Socioeconomic factors may influence access to educational resources. Inequalities in access to quality education and science programmes disproportionately impact women from disadvantaged backgrounds [ 18 ]. These variables are shown in Table 9 . Table 9. Control variables. Related Factors Variable Survey Items (yes/no) Socioeconomic Access to a Computer D17. I have a computer/laptop that works well Disadvantaged Economic Class D18. I have stable Internet connectivity 24/7 D19. I have a permanent power supply Academic Mentor B2. I have a mentor Family Parents' Educational Background F3_1. My mum or dad graduated from college F3_2. My mum or dad completed higher education in a STEM career Open in a new tab 4.6. Data collection process Data were collected through a cross-national survey administered in urban areas in Bolivia, Brazil, and Peru. The selection of a professional survey administration company began in September 2022 through UNISELVA, the foundation responsible for the financial management of the ELLAS project. Contracting an external provider was necessary to ensure large-scale data collection, standardized implementation across countries, and access to established survey panels and fieldwork infrastructure, which were beyond the operational capacity of the research team. The initial goal was to engage a single company capable of administering the survey in all three countries; however, after five months without identifying a suitable provider, a two-company solution was adopted. One firm specialized in Bolivia, while a second covered Brazil and Peru, requiring the coordination of bilingual (Portuguese and Spanish) contracts and extending the project timeline. The survey instrument comprised 136 items, primarily multiple-choice and binary questions, spanning individual, academic, occupational, social, family-related, and socioeconomic dimensions. It operationalized 23 variables through 123 indicators and included control variables such as gender, age, occupation, area of work, and country. These specifications were shared with prospective survey companies during the contracting process. Data collection employed purposive sampling, a non-probability approach in which participants were intentionally selected based on predefined demographic and professional characteristics to ensure adequate representation. A minimum target sample of 3189 respondents per country was established, with proportional quotas by age, gender, and area of work (STEM versus non-STEM). Survey administration took place between August 2023 and March 2024. Data were collected primarily online in Brazil and Peru using panel-based recruitment, while in Bolivia, limited panel availability necessitated predominantly face-to-face interviews. This mixed-mode approach allowed the project to meet sampling targets while accommodating country-specific constraints. The Brazilian data collection concluded last due to challenges in recruiting sufficiently diverse student populations. Limitations As with most empirical studies, this work has several limitations that should be considered when using and interpreting the data. First, the study employed purposive sampling, a non-probability approach that is not intended to support statistical generalization. Although this method is well suited for reaching specific populations, such as women in STEM fields, the findings should be interpreted with caution when extrapolating to broader populations. Second, data collection methods varied across countries. In Brazil and Peru, surveys were administered through online panels, whereas in Bolivia, limited digital infrastructure required the use of face-to-face interviews. While these adaptations were necessary to ensure adequate coverage, they may have introduced mode effects that should be considered when making cross-country comparisons. Third, despite a rigorous data-cleaning process, a small number of responses exhibit atypical or internally inconsistent patterns, such as implausibly early educational milestones, unrealistic work experience timelines, or outlier family data. Although these cases are limited, they warrant careful attention, particularly in analyses involving outliers or incomplete information. As described before, problematic cases were coded for the easy reference of users. Finally, although the survey was translated into Spanish and Portuguese and subjected to back-translation procedures, subtle linguistic and cultural nuances may have influenced item interpretation. This is especially relevant for culturally specific terms and professional classifications, underscoring the importance of contextual sensitivity in cross-national analyses. Ethics Statement The ELLAS project was coordinated by the Federal University of Mato Grosso (UFMT). The research protocol for the application of the questionnaire in Brazil received ethical approval from CEP-Humanidades/UFMT (CAAE: 66,296,922.6.0000.5690; approval date: March 6, 2023), authorizing the use of a Free and Informed Consent Form (CLE). Participation was voluntary, and informed consent was obtained from all participants prior to data collection. Additional ethical approvals were obtained in participating countries. In Peru, the study was approved by the Instituto de Investigación Científica of Universidad de Lima (May 22, 2023). In Bolivia, the protocol was reviewed and approved by an institutional ethics committee at Universidad Mayor de San Andrés (June 27, 2023). Credit Author Statement Indira R. Guzman: Conceptualization, Methodology, Writing, Original draft preparation; Boris Branisa: Conceptualization, Data curation, Writing, Original draft preparation; Guillermo Davila: Conceptualization, Methodology, Writing; Florencia Sánchez Guillén: Data curation, Writing, Original draft preparation and Editing; Leihge Roselle Rondon Pereira: Writing, Reviewing and Editing; Luana Mendes: Execution, Reviewing and Editing; Luciana Bolan Frigo: Reviewing and Editing; Nadia Rodríguez-Rodriguez: Conceptualization, Reviewing; Silvia Amélia Bim: Conceptualization, Reviewing; Cristiano Maciel: Conceptualization, Reviewing, Editing and overall process coordination. Acknowledgements This work was generously supported by the International Development Research Centre (IDRC) of Canada. Project Title: “Latin American Open Data for gender-equality Policies Focusing on Leadership in STEM” by Equality in Leadership for Latin America STEM (ELLAS) Network. ( https://ellas.ufmt.br/ ). Project ID #109798 [ 36 ]. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Footnotes Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.dib.2026.112721 . Appendix. Supplementary materials mmc1.pdf (2MB, pdf) Data Availability ELLAS Platform ELLAS Survey Dataset (Original data) . References 1. UNESCO, «UNESCO call to action: closing the Gender gap in science». 2024. [En línea]. Disponible en: https://unesdoc.unesco.org/ark:/48223/pf0000388641 . 2. I. Guzman, R. Berardi, C. Maciel, P. Cabero Tapia, G. Marin-Raventos, N. Rodriguez, and M. Rodriguez, “Gender gap in IT in Latin America,” 2020. 3. World Bank . World Bank; Washington, DC, USA: 2021. World Development Indicators. [ Google Scholar ] 4. World Economic Forum . World Economic Forum; Geneva, Switzerland: 2022. Global Gender Gap Report 2022. [ Google Scholar ] 5. Leaper C., Starr y C.R. Helping and hindering undergraduate women’s STEM motivation: experiences with STEM encouragement, gender bias, and sexual harassment. Psychol. Women. Q. 2019;43(2):165–183. [ Google Scholar ] 6. UNESCO . UNESCO Institute for Statistics; Paris, France: 2021. Women in Science: fact Sheet No. 55. [ Google Scholar ] 7. International Labour Organization . International Labour Organization; Geneva, Switzerland: 2021. World Employment and Social Outlook: Trends 2021. [ Google Scholar ] 8. Ramírez-Corona N., Aguirre Calleja A.C., Segovia-Hernández J.G., Aristizábal-Marulanda V. «Latin American women in chemical engineering: challenges and opportunities». Chem. Eng. Proc. 2022;181 [ Google Scholar ] 9. Avolio y B., Chávez J. Professional development of women in STEM careers: evidence from a Latin American country. Glob. Bus. Rev. 2023 [ Google Scholar ] 10. T. Memanova y I. Catita, Catalyzing Women in STEM Entrepreneurship: Ecosystem Gaps and Pathways for Support in Central America and Ecuador, Inter-American Development Bank (IDB) and IDB Lab, Washington, DC, IDB-TN-3252, nov. 2025. [En línea]. Disponible en: https://publications.iadb.org/publications/english/document/Catalyzing-Women-in-STEM-Entrepreneurship-Ecosystem-Gaps-and-Pathways-for-Support-in-Central-America-and-Ecuador.pdf . 11. López P., Simó P., Marco J. Understanding STEM career choices: a systematic mapping. Heliyon. 2023;9 doi: 10.1016/j.heliyon.2023.e16676. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Muñoz M., Delgado L. Economic Commission for Latin America and the Caribbean (ECLAC); Santiago, Chile: 2023. Women in Decision-Making Positions in Latin America. [ Google Scholar ] 13. Drummond B., et al. Mapping contextual aspects that influence women in computing in Latin America. Interfases. 2023;18:19–30. n.o December. [ Google Scholar ] 14. Maciel C., Guzman I.R., Berardi R.C.G., Rodriguez-Rodriguez N., Salgado L., Frigo L.B., et al. An open data platform to advance gender equality in STEM in Latin America. Commun. ACM. 2024;67(8):90–92. doi: 10.1145/3653294. [ DOI ] [ Google Scholar ] 15. Zadro L., Williams K.D., Richardson R. «How low can you go? Ostracism by a computer is sufficient to lower self-reported levels of belonging, control, self-esteem, and meaningful existence». J. Exp. Soc. Psychol. 2004;40(4):560–567. doi: 10.1016/j.jesp.2003.11.006. jul. [ DOI ] [ Google Scholar ] 16. Withanaarachchi y A., Vithana N. Female underrepresentation in the cybersecurity workforce – a study on cybersecurity professionals in Sri Lanka». ICS. 2022;30(3):402–421. doi: 10.1108/ICS-08-2021-0129. may. [ DOI ] [ Google Scholar ] 17. Lam S., Yim P., Ng Y. «Is effort praise motivational? the role of beliefs in the effort–ability relationship». Contemp. Educ. Psychol. 2008;33(4):694–710. [ Google Scholar ] 18. Bea M.D., Yi Y. Leaving the financial nest: connecting young adults’ Financial independence to financial security. J. Marriage and Family. 2019;81(2):397–414. [ Google Scholar ] 19. Patterson M.M. Self-perceived gender typicality, gender-typed attributes, and gender stereotype endorsement in elementary-school-aged children. Sex. Roles. 2012;67(7):422–434. [ Google Scholar ] 20. Kinias Z., Kim y H.S. Culture and gender inequality: psychological consequences of perceiving gender inequality. Group Process. Interg. Relat. 2012;15(1):89–103. [ Google Scholar ] 21. Schwartz-Salazar S., García-Sánchez E., Martínez R., Rodríguez-Bailón R. Development and validation of the multidimensional Gender Inequality perception scale (MuGIPS) PLoS. One. 2024;19(4) doi: 10.1371/journal.pone.0301755. p. e0301755, abr. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Ranf K.M., Báguena Puigcerver M.J., Toldos Romero M.P., Beleña y A. «El Recuerdo de Experiencias de Agresión en la Escuela: consecuencias y Diferencias de Género». Psicopatología Clínica Legal y Forense. 2006;6(1–3):133–152. [ Google Scholar ] 23. Spreitzer G.M. An empirical test of a comprehensive model of intrapersonal empowerment in the workplace. Am. J. Comm. Psychol. 1995;23(5):601–629. doi: 10.1007/BF02506984. oct. [ DOI ] [ PubMed ] [ Google Scholar ] 24. Starr C.R. I’m not a science nerd!”: STEM stereotypes, identity, and motivation among undergraduate women. Psychol. Women. Q. 2018;42(4):489–503. doi: 10.1177/0361684318793848. dic. [ DOI ] [ Google Scholar ] 25. Xu X., Payne S.C. Quantity, quality, and satisfaction with mentoring: what matters most? J. Career. Dev. 2014;41(6):507–525. doi: 10.1177/0894845313515946. [ DOI ] [ Google Scholar ] 26. García-Holgado A., Deco C., Bredegal-Alpaca N., Bender C., Villalba-Condori K.O. IEEE Global Engineering Education Conference (EDUCON) IEEE; 2020. Perception of the gender gap in computer engineering studies: a comparative study in Peru and Argentina; pp. 1252–1258. en 2020. [ Google Scholar ] 27. French B.F., Oakes W. Reliability and validity evidence for the institutional integration scale. Educ. Psychol. Meas. 2004;64(1):88–98. [ Google Scholar ] 28. Raver J.L., Nishii L.H. «Once, twice, or three times as harmful? Ethnic harassment, gender harassment, and generalized workplace harassment.». J. Appl. Psychol. 2010;95(2):236–254. doi: 10.1037/a0018377. mar. [ DOI ] [ PubMed ] [ Google Scholar ] 29. Terborg J.R., Peters L.H., Ilgen D.R., Smith F. Organizational and personal correlates of attitudes toward women as managers. Acad. Manag. J. 1977;20(1):89–100. doi: 10.2307/255464. mar. [ DOI ] [ Google Scholar ] 30. Castillo-Mayén y R., Montes-Berges B. Analysis of current gender stereotypes. Anales de Psicología. 2014;30(3):1044–1060. [ Google Scholar ] 31. Nava Quiroz C.N., Bezies Álvarez R., Vega Valero y C.Z. «Adaptación y Validación de la Escala de Percepción de Apoyo Social de Vaux». Liberabit. 2015;21(1):49–58. [ Google Scholar ] 32. Keller B.K., Whiston S.C. The role of parental influences on young adolescents’ Career development. J. Career. Assess. 2008;16(2):198–217. [ Google Scholar ] 33. Luo T., So W.W.M., Wan Z.H., Li W.C. STEM stereotypes predict students’ STEM career interest via self-efficacy and outcome expectations. Int. J. Stem Educ. 2021;8(1):36. [ Google Scholar ] 34. Chan y K.-Y., Drasgow F. Toward a theory of individual differences and leadership: understanding the motivation to lead. J. Appl. Psychol. 2001;86(3):481–498. doi: 10.1037/0021-9010.86.3.481. [ DOI ] [ PubMed ] [ Google Scholar ] 35. Chandy R.K., Tellis G.J. The Incumbent’s curse? incumbency, size, and radical product innovation. J. Mark. 2000;64(3):1–17. [ Google Scholar ] 36. C. Maciel, I.R. Guzman, K.D.S.F.M. Ribeiro, F. Suni-Lopez, L.B. Frigo, L. Salgado, et al., Fostering open data on women in STEM across Latin America: the ELLAS network, Breaking Barriers, Building Bridges, edited collection. https://www.researchgate.net/publication/399159092_Fostering_Open_Data_on_Women_in_STEM_across_Latin_America_The_ELLAS_Network . Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials mmc1.pdf (2MB, pdf) Data Availability Statement ELLAS Platform ELLAS Survey Dataset (Original data) . Articles from Data in Brief are provided here courtesy of Elsevier ACTIONS View on publisher site PDF (411.6 KB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top