European Commission

Directorate-General for Research and Innovation (DG RTD)

Not applicable

Not applicable

Not applicable

RTD-GENDERINRESEARCH@ec.europa.eu; RTD-PUBLICATIONS@ec.europa.eu

Not applicable

Not applicable

03/06/2026

03/06/2026

01/06/2026

She Figures provides a range of comparable, pan-European statistics on gender equality in Research and Innovation, and has been released every three years since 2003. 

A large portion of the indicators included in She Figures present and explore the following themes:

• The presence of women in research across different sectors 

• Horizontal segregation by sex across different fields of research and development and research occupations 

• Vertical segregation by sex in academia, i.e. the (under-) representation of women in the highest grades and research posts and as heads of academic institutions 

• The relative working conditions of female and male researchers 

• Steps taken by research institutions to promote gender equality internally 

Each edition of She Figures also aims to provide better understanding of emerging issues by introducing additional indicators.

Further information about She Figures publications, including downloadable reports and other publications can also be found on the webpage of the Publications Office of the European Union.

The Gender Statistics Database includes She Figures indicators since 2015 (reference year 2012).

See also the original reports:

She Figures reports

• She Figures 2015
• She Figures 2018
• She Figures 2021
• She Figures 2024 

She Figures handbooks

• She Figures handbook 2015
• She Figures handbook 2018
• She Figures handbook 2021 
• She Figures handbook 2024

The following classification systems are relevant for the She Figures indicators included in the Gender Statistics Database:

• OECD (2015) Fields of Research and Development (R&D) Classification (FORD) (available here: the Frascati Manual (OECD, 2015)): 

• Natural sciences 

• Engineering and technology 

• Medical and health sciences 

• Agricultural sciences 

• Social sciences 

• Humanities 
• She Figures additionally uses the classifications “multi-disciplinary” and “unknown” for indicators based on the Women in Science data, so that data can also be provided for applications whose field is unknown or cover more than one specific field respectively

• European Framework for Research Careers (see https://cdn5.euraxess.org/sites/default/files/policy_library/towards_a_european_framework_for_research_careers_final.pdf ), which classifies researchers as: 

• R1: First Stage Researcher (up to the point of PhD) 

• R2: Recognised Researcher (PhD holders or equivalent who are not yet fully independent) 

• R3: Established Researcher (researchers who have developed a level of independence) 

• R4: Leading Researcher (researchers leading their research area or field) 

• International Patent Classification (IPC) level 1 (see https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Glossary:International_patent_classification_(IPC)): 

• Human necessities 

• Performing operations and transporting 

• Chemistry and metallurgy 

• Textiles and paper 
• Fixed constructions

• Mechanical engineering, lighting, heating, weapons and blasting 

• Physics
• Electricity

• Grading system of research positions (based upon national mappings, as described in the She Figures handbooks):
  • Grade A: The single highest grade / post at which research is normally conducted within the institutional or corporate system 
  • Grade B: All researchers working in positions which are not as senior as the top position (A) but definitely more senior than the newly qualified PhD holders (C); i.e., below A and above C
  • Grade C: The first grade/post into which a newly qualified PhD (ISCED 8) graduate would normally be recruited within the institutional or corporate system
  • Grade D: Either postgraduate students not yet holding a PhD (ISCED 8) degree who are engaged as researchers (on the payroll) or researchers working in posts that do not normally require a PhD

For details, see the She Figures reports and handbooks(links under Section 3.1 Data description).

The She Figures indicators included in the Gender Statistics Database cover all the fields of R&D and IPC Classifications listed at Classification system.

The indicators on international mobility and labour market conditions of researchers cover researchers only in the higher-education sector (i.e. those working at Higher Education Institutions)

Details are provided in the She Figures reports and handbooks (see Section 3.1 Data description).

Below we list only the statistical concepts and definitions that are relevant to understand and interpret the She Figures indicators included in the Gender Statistics Database.

Gender dimension in research:

The definition used for this indicator is based on that developed by the European Commission’s Expert Group on Gendered Innovations / Innovations through Gender. This defines research that has a gender dimension as that which integrates sex and/or gender analysis into research (where "sex" refers to biological characteristics and "gender" refers to cultural attitudes and behaviours that shape the ideas of "feminine" and "masculine") .

Intersectionality in research:

The method to identifying research that takes an intersectional approach is based on a shortlist of keywords from the European Commission’s Gendered Innovations 2 report.

Definition of Researcher:

The She Figures reports apply the definition of “researcher” from the Frascati Manual (OECD, 2015), according to; which (par. 5.35) “Researchers are professionals engaged in the conception or creation of new knowledge, products, processes, methods and systems and also in the management of the projects concerned”.

Internationally mobile researchers (for indicators based on the MORE surveys, available only up to 2021):

• ‘International mobility during PhD’ applies to researchers who have moved abroad for at least three months during their PhD to a country other than the one where they completed (or will obtain) their PhD. In She Figures 2015 to 2021, the derived indicator is based on a direct question in the relevant edition of the MORE survey.

• ‘International mobility in the post PhD career stages’ applies to researchers who have worked abroad for more than three months at least once in the last 10 years, since obtaining their highest educational qualification (PhD or other). In She Figures 2015 to 2021, the derived indicator is based on a direct question in the relevant edition of the MORE survey. 

Part-time and precarious employment (for indicators based on the MORE surveys, available only up to 2021):

• “Part timeemployment” covers respondents who self-reported any of these three statuses: ‘part time: more than 50 %’; ‘part time: 50 %’; ‘part time: less than 50 %’. 

• “Precarious employment” includes: 
  • Researchers who indicated that they have a fixed term contract of one year or less 
  • Researchers who indicated that they have no contract 
  • Researchers who indicated that they have an ‘other’ non-fixed term, non-permanent type of contract (often associated with student status), unless they stated explicitly that they had a contract of indefinite duration. 

The MORE survey’s definition of ‘precarious’ employment differs from that of the Labour Market and Labour Force Statistics, which describes as ‘precarious’ contracts with duration of three months or less (https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Beginners:Labour_market_-_employment).

Active authors:

In the context of these indicators, “active authors” are defined as those that produced 10 or more papers in the last 20 years (2000-2019) and at least 1 paper in the last 5 years (2015-2019) OR those who produced 4 or more papers in last 5 years (2015-2019). 

Fractional authorship:

“Fractional authorship” is a means of distributing publication and citation counts equally across multiple authors. For example, a given publication with 2 authors would be counted as 0.5 publications for each author.

Gender dimension of a publication:

Research that explores the gender dimension is defined as research that “integrates sex and gender analysis into research, whereby “sex” refers to basic biological characteristics of females and males and ‘gender’ refers to cultural attitudes and behaviours that shape ‘feminine’ and ‘masculine’ behaviours, products, technologies, environments, and knowledge.” This definition is provided by the Gender Innovation Project (http://ec.europa.eu/research/swafs/pdf/pub_gender_equality/gendered_innovations-KINA25848ENC.pdf#view=fit&pagemode=none)

Compound Annual Growth Rate (CAGR):

Compound annual growth rate (CAGR) is defined as the year-over-year constant growth rate over a specified period of time. Starting with the first value in any series and applying this rate for each of the time intervals yields the amount in the final value of the series. Throughout the term CAGR is also referred to as ‘(yearly) growth rate.’

Field-Weighted Citation Index (FWCI):

FWCI is an indicator of citation impact of a publication based on the actual number of citations received by an article compared to the expected number of citations for articles of the same document type (article, review or conference proceeding paper), publication year and subject field. When an article is classified in two or more subject fields, the harmonic means of the actual and expected citation rates is used. The indicator is therefore always defined with reference to a global baseline of 1.0 and intrinsically accounts for differences in citation accrual over time, differences in citation rates for different document types (reviews typically attract more citations than research articles, for example) as well as subject-specific differences in citation frequencies overall and over time and document types.

In general, the Field-Weighted Citation Impact (FWCI) for a publication is defined as:

FWCI = C_i/E_i

Where:

C_i: citations received by publication i

E_i: expected number of citations received by all similar publications in the publication year plus following 3 years

When a similar publication is allocated to more than one discipline, the harmonic mean is used to calculate E_i.

Actions and measures towards Gender Equality:

In relation to assessing the number of Research Performing Organisations (RPOs) that have taken actions and measures towards Gender Equality, an RPO was judged to have taken actions or measures towards Gender Equality if one or more of the following phrases (translated into the relevant national language) were found on their website:

• Gender Equality
• Gender Equality Plan
• Equal opportunities officer
• Equal participation officer 
• Eliminate/Prevent sex discrimination
• Eliminate/Prevent harassment
• Harassment policy
• Gender diversity committee
• Gender diversity office
• Gender diversity task force

Units – Head Count & Full Time Equivalent (as defined in the Frascati Manual (OECD, 2015) (sections 5.49 and 5.58)):

• Head Count (HC) (§5.58): the number of persons engaged in R&D at a given date, the average number of persons engaged in R&D during the (calendar) year, or the total number of persons engaged in R&D during the (calendar) year.
• Full-time equivalent (FTE): the ratio of working hours actually spent on R&D during a specific reference period (usually a calendar year) divided by the total number of hours conventionally worked in the same period by an individual or by a group.

Boards:

• Scientific board of research organisation: A publicly or privately managed and financed group of elected or appointed experts that exists to implement scientific policy by, among other things, directing the research agenda, resource allocation and management within scientific research. 
• Administrative / advisory board of research organisation: A publicly or privately managed and financed group of elected or appointed experts that exists to support the research agenda in a non-executive function by, among other things, administering research activities, consulting and coordinating different actors and taking a general advisory role.

The statistical unit varies by the source of data:

CORDIS: The statistical unit is a research project.

MORE: The statistical unit is a researcher.

PATSTAT: The statistical unit is an inventorship or patent application, depending on indicator.

Scopus^TM: For indicators on authorship, the statistical unit is an author of a publication. For the indicator on gender dimension in research and innovation content, the statistical unit is a publication.

Web scraping (for data on RPOs taking actions toward gender equality): The statistical unit is a Research Performing Organisation (RPO)

The statistical population varies by the source of the data:

CORDIS: Horizon 2020 projects that include a research subject involving human or animal subjects.

MORE: Researchers aged 25 and older and residing in the reference country.

PATSTAT: For inventorship indicators: inventorships in PATSTAT; for patent-application indicators: patent applications in PATSTAT.

Scopus^TM: For indicators on authorship, the statistical population is all academic authors with a first name available  within the Scopus^TM abstract and citation database. All author IDs for whom no first name data was available were excluded from the analysis. For indicators on active authorship, only active authors are included. For the indicator on gender dimension in research and innovation content, the statistical population is all publications as recorded in the  Scopus^TM abstract and citation database.

Web scraping (for data on RPOs taking actions toward gender equality): Two groups of RPOs were included:

• Higher education institutions (found in the ETER project)
• Public Research Organisations (the public bodies and research organisations that participated in projects under Framework Programme 7 (FP7) and Horizon 2020 (H2020) Framework Programme for R&I; found in the CORDIS database and revised from statistical correspondents)

Women in Science (WiS) questionnaires: The statistical population differs by indicator.

• For indicators on the proportion of female and male academic staff at different grades and in different fields, the statistical population is academic staff or researchers in the HES (where available, researchers in the HES was used). Full details of which population was used for each country is available in the methodological notes of the She figures reports.  
• For indicators on the proportion of female heads of universities or assimilated institutions which can deliver PhDs, the statistical population is heads of universities or assimilated institutions which can deliver PhD.
• For indicators on the proportion of women on boards, the statistical population is board members within the boards of umbrella, national-level research performing organisations (RPOs) and research funding organisations (RFOs).
• For indicators on funding success rate difference between women and men, the statistical population is applicants and beneficiaries of research funding within national funding bodies. The full list of national funding bodies included for each country can be found in the methodological notes in the She Figures reports.

The EU Member States, in addition to candidate countries (Albania, North Macedonia, Montenegro, Serbia and Turkey) and Associated Countries (Armenia, Bosnia and Herzegovina, Faroe Islands, Georgia, Iceland, Israel, Moldova, Norway, Switzerland, Tunisia, Ukraine and the UK).

Time coverage varies by indicator. See the individual datasets and the published She Figures reports (links under section 3.1 Data description).

Not applicable

The EU is committed to advancing gender equality in research and innovation. In particular, the promotion of gender equality and the integration of the gender dimension in research and innovation are established as cross-cutting objectives under Regulation (EU) 2021/695 establishing Horizon Europe, which repealed Regulation (EU) No 1291/2013. 
 
More recently, the 2020 ERA Communication and the subsequent European Research Area policy agenda reaffirmed and strengthened the EU’s commitment to gender equality and gender mainstreaming in research, including through institutional change, Gender Equality Plans, and more inclusive research and innovation systems.

Not applicable

Not applicable.

No direct identification of individuals is possible from the indicators.

Datasets are freely available to the public via the EU Open Data Portal within one year after data collection.

Not applicable.

Datasets are freely available to the public via the EU Open Data Portal within one year after data collection.

No regular news releases.

She Figures reports published every three years. See links under section 3.1 Data description.

She Figures is not published as an online database. A selection of indicators is available via EIGE’ s Gender Statistics Database..

Some She Figures indicators are computed from survey micro-data. The underlying micro-data are not available for public access.

Not applicable.

The methodology is documented in the She Figures handbooks and reports (links under section 3.1 Data description).

Details are provided in the She Figures handbooks (links under section 3.1 Data description).

Details are provided in the She Figures handbooks (links under section 3.1 Data description).

Based on the European Statistical System (ESS) quality criteria, the She Figures indicators can be considered of high quality in terms of relevance, timeliness and punctuality. In fact, the She Figures indicators are highly relevant for a wide range of users, from national governments, the EU, and international and national non-governmental organisation. The She Figures indicators use the most recent available data to describe the current situation in the single countries and at the EU level and are published no later than one year after the data collections.

 Some weaknesses have been identified in terms of accuracy and comparability over time / across countries for some She Figures indicators. Further details on this issue are provided in sections 14 Accuracy and 16 Comparability.

The users of She Figures data include EU policy makers, national governments, and international organisations. The publications provide an insight into the situation regarding gender equality in Research and Innovation at the pan-European level. It aims to give an overview of the gender equality situation in research and innovation, using a wide range of indicators to examine the impact and effectiveness of the policies implemented in this area.

Not applicable.

Indicators are complete relative to relevant regulations and guidelines.

For details, see the She Figures reports and handbooks (links under section 3.1 Data description). Here we briefly discuss the main limitations by source of data.

CORDIS and Women in Science (WiS) questionnaires: As described in Geographical comparability, there are country-specific differences and/or a lack of a clear definition for key terms such as “academic staff”, “boards” and “funds”. Measures have been taken to try to harmonise the approach taken for each country, which may impact the accuracy of the data to some extent. Any differences in approach by country are recorded in the She Figures handbooks.

PATSTAT: Bibliometric indicators and indicators on inventions and innovations are computed from databases that do not explicitly indicate the sex of the author/patent applicant. Different methodologies were used to infer the sex of authors/applicants for the Scopus/Web of Science and PATSTAT databases.  Overall, these procedures ensure a good level of accuracy for the matched names but may fail to provide matches between authors and sexes in some cases.

Scopus^TM: As noted in the She Figures handbook 2018, bibliometric indicators can only give insights about the outputs of male and female authors in different countries but should not be interpreted as accurate measure of scientific production. Especially for social sciences and humanities, a reasonable proportion of research outputs take the form of books, monographs, and non-textual media. As the She Figures bibliometric indicators refer only to journal articles, they cannot fully account for the scientific production of researchers in those fields. Moreover, bibliometric indicators can only account for the scientific production of named authors and not of all “researchers” (based on the OECD, 2015 definition, provided at point 3.4 above) in a country. For instance, research in the corporate sector would not be included in these bibliometric indicators. The bibliometric indicators exploring the sex and gender dimension of research output can be considered accurate. For the 2018 She Figures, the output generated by the different bibliometric queries used to construct these indicators was validated by a pool of gender experts, and the number of false positives, based on the expert assessment, was less than 1%. Despite this accuracy, these indicators can still suffer from bias because they depend on the fields of publications presented in the bibliometric databases used as data sources.

Finally, both bibliometric indicators and indicators on inventions and innovations are computed from databases that do not explicitly indicate the sex of the author/patent applicant. Different methodologies were used to infer the sex of authors/applicants for the Scopus/Web of Science and PATSTAT databases. Details can be found in the She Figures handbooks. Overall, these procedures ensure a good level of accuracy for the matched names but may fail to provide matches between authors and sexes in some cases.

MORE:Accuracy is generally good. The main source of error for these indicators is sampling error (see section 14.2), which is common for indicators computed from survey data.

Web scraping (for data on RPOs taking actions toward gender equality): The results of this indicator are estimates. The accuracy of this indicator was calculated at 86 % during an exploratory web-scraping phase. This means that the indicator correctly assigned organisations as having or not having taken actions and measures towards gender equality in 86% of the cases. 

There may also be some limitations to accuracy in terms of the accuracy of the population included, as not all organisations funded by FP7 or H2020 are research organisations. However, the input of the statistical correspondents contributed greatly to the removal of such organisations from the lists.

Finally, some of the search phrases may be more or less common across countries (and under-/over-estimate the indicator). In order to have comparable research in all countries, country-specific terms were excluded even if they may have had increased accuracy at country level.

For details, see the She Figures reports and handbooks (links under section 3.1 Data description). Here we briefly discuss the main limitations by source of data.

CORDIS and web scraping: Not applicable

Women in Science (WiS) questionnaires: Not easily quantifiable. The questionnaires did not sample the statistical population directly but rather surveyed a selection of Statistical Correspondents.

PATSTAT and Scopus^TM: Sampling errors are reported also for bibliometric and invention indicators. The sampling errors assume that the bibliometric / patent database (respectively) are a random sample of all publication / patent applications in each subfield / IPC category (respectively). Sampling errors were used to compute confidence intervals for the bibliometric / inventions indicators, which are reported in the She Figures publications.

MORE: Indicators computed from the MORE surveys have a low sampling error (lower than 5%) and hence can be considered accurate if computed at the country and at the European level. However, as specified in the MORE3 methodological report, indicators computed at other subpopulation levels (e.g., field of science, gender, career stages) are not guaranteed to have the same accuracy, except at the EU level. See the discussion in: European Commission, DG for Research and Innovation, Survey on researchers in European Higher Education institutions – Annex to MORE3 study

Non-sampling errors for the She Figures indicators included in the Gender Statistics Database may be related to processing errors such as cleaning errors or mis-assignment of gender, or the presence of outliers.

For details, see the She Figures reports and handbooks (links under section 3.1 Data description). Here we briefly discuss the main limitations by source of data.

The She Figures data collections take place every three years. Data refer to the most recent point in time available (this varies by data sources).

The She Figures publications are released according to schedule.

For details, see the She Figures reports and handbooks (links under section 3.1 Data description). Here we briefly discuss the main limitations by source of data.

MORE:

 The same methodology was followed in all countries, ensuring excellent cross-country comparability.

PATSTAT:

Indicators on innovation and inventions (e.g., patent applications) are partially comparable across countries, as the percentage of patent applicants to which a sex could be attributed varies by country. Details on the percentage of matched sex-name pairs can be found in the She Figures handbooks.

There may be bias in favour of some countries, as more applications could be attributed to some countries and less to others. However, since the indicators are ratios of variables referring to the same given country it is expected that such bias does not affect the cross-country comparability of them.

Scopus^TM:

Bibliometric indicators are partially comparable across countries. In fact, although the methodology to compute the indicators is the same across countries, the extent to which names of the authors could be matched to their sexes varied by country.

Web scraping (for data on RPOs taking actions toward gender equality):

As noted in Accuracy overall, the same search terms were used for each country (see Statistical concepts definitions for a list of terms) to support comparability between countries.

However, there are some limitations to comparability as it is unknown how common each search term is in each country (so the indicator may under-estimate the prevalence of gender equality measures and actions in countries where other terms are more frequently used).

Additionally, lists of national RPOs were not prepared in the same way (some are curated lists of FP7/H2020 participants, others are the lists of public organisations surveyed for the production of R&D statistics), which may also limit comparability.

Women in Science (WiS) questionnaires:

Regarding the indicators on the proportion of academic staff at different grades, it is important to note that these data are not always completely cross-country comparable as the seniority of grades is not yet defined in the same way across countries. Furthermore, it is not always possible to distinguish research staff from teaching staff, although the target population for ‘academic staff’ is researchers in higher education institutions (excluding staff involved in teaching or administration only and not at all in research).

Regarding the indicators on boards, no common definition of boards exists and the number of boards varies significantly between countries. Boards of organisations that are performing research and boards of organisations that are funding research are both included in the final computations (although these are distinguished in the methodological annex of the She Figures report).

Regarding the indicators on funding success, no common definition of funds exists, and the total number of funds varies significantly between the countries and over the time period being considered. To increase comparability of the data, it was requested that data should cover all publicly managed research funds (funds granted by institutions in the public sector, excluding private sector funding) and funds which allocate funding exclusively on a first-come, first-served basis, (i.e. without other selection criteria) were excluded.

Each She Figures publication contains an appendix with a correspondence table establishing comparability between indicators in the current and previous editions of She Figures. Data are displayed in EIGE’ s Gender Statistics database in accordance with these tables. Indicators that are not comparable from one edition to the next (even when they retain the same title) are displayed as separate tables in the Database. Data from multiple editions are shown side-by-side only when they have been deemed to be comparable over time

For most She Figures indicators, no alternative data sources exist.

For indicators on contract types and international mobility, different sources (such as the now-discontinued MORE survey and Labour Force Surveys) use different definitions, which prevent their comparability. For this reason, data on indicators using different sources to replace the MORE survey should not be compared to earlier data.

For example,, Eurostat defines “precarious working contracts” as those which are three months or less, while MORE defined precarious working contracts as those without contract, with fixed-term contracts of up to one year, or with other non-fixed term, non-permanent contracts.

Each She Figures indicator included in the Gender Statistics Database has full internal coherence, as it is based on the same data source. Data sources differ across indicators

Variables are being edited and corrected based on set of logical edits at data entry stage. No revisions are done after the publication for the data.

No fixed revision schedule.

The data sources of the She Figures indicators included in the Gender Statistics Database are:

•       European Commission’s Women in Science (WiS) questionnaires on the representation of women at various stages of an academic career and in various decision-making bodies
•       The Scopus^TMabstractandcitation database
•       The Worldwide Patent Statistical Database (PATSTAT) of the European Patent Office (EPO)
•       CORDIS (Community Research and Development information Service), a public database on research funding in the EU
•       Web scraping of research organisation’s websites
•       Eurostat
•       European Commission’s MORE Surveys on the career paths, working conditions, and mobility patterns of EU researchers (only up to She Figures 2021)

The frequency of data collection varies by indicator. See the She Figures reports and handbooks (links under section 3.1 Data description).

The methods of data collection vary by indicator. See the She Figures reports and handbooks (links under section 3.1 Data description).

The She Figures  handbooks detail the coherence and validation checks carried out to detect and correct non-sampling errors and outliers. See links under section 3.1 Data description.

Details are provided in the She Figures handbooks (links under section 3.1 Data description).

Not applicable