Statistics for gender equality

Download 80.7 Kb.
Date conversion04.05.2016
Size80.7 Kb.

Dr Esther Breitenbach

University of Edinburgh

Presentation at seminar on ‘Indicators and Statistics on gender equality and the new duty on public bodies to promote gender equality from April 2007’, held on 12 April, 2007, Royal Statistical Society, London

Royal Statistical Society, Official Statistics Committee

Economic and Social Research Council

Gender Statistics Users Group

The general context in which I have been working on gender statistics in recent years is one in which these have been increasingly promoted as a tool in gender equality policy making, and in particular as part of gender mainstreaming. The provision of gender disaggregated statistics and the development of gender equality indicators has been widely promoted in an international context since the mid 1990s: by the UN, stemming from the adoption of the Global Platform for Action in Beijing in 1995; by the adoption of a gender mainstreaming approach by an increasing number of governments and at different levels of governance e.g. EU, national parliaments, devolved administrations, and local government. Key arguments for the development of gender equality indicators have been well rehearsed: they are useful for monitoring change, for making governments accountable, for mobilising people in support of change, and for empowering women. It has also been argued that having a grasp of official statistics and understanding how they are used is part of ‘critical citizenship’.

In this presentation I aim to do two things:

  • Firstly, to make some brief comments on the provision of gender disaggregated statistics in the UK and devolved administrations – how this has developed, and where we are now;

  • Secondly, to discuss the notion of gender equality indicators and some options for their development.

Reviewing the current position

Last year I was commissioned by the EOC to carry out an evaluation of gender statistics across selected topic areas and for a range of geographies within the UK, and this provided an opportunity to assess what progress had been made. Specifically, the remit of the research was to identify publicly available gender disaggregated data for GB, England and Wales, Wales, and Scotland for five topic areas: local government, education, health, the criminal justice system, and transport. The evaluation found that:

  • In general, since the late 1990s, there has been a marked improvement in the provision of gender disaggregated statistics by the Office for National Statistics and by government departments;

  • The types of improvement that have occurred are similar across jurisdictions in the UK, Wales and Scotland [Northern Ireland was not part of this study, though developments on gender statistics have also been taking place there], though there is some difference in the capacity of administrations to produce data. A combination of factors is responsible for this: the asymmetrical distribution of powers between the UK government and devolved administrations e.g. crime and justice data are produced by the Home Office for England and Wales combined, while Scotland produces its own data; sample sizes in UK/GB surveys which may not permit sufficient disaggregation; and resources available for statistical services.

These types of improvement include:

  • Guides to sources of gender and other equality statistics;

  • Compendia of gender disaggregated statistics;

  • One off in-depth studies with gender analysis and/or focus;

  • Consultation with users;

  • A much more systematic approach to the publication of gender disaggregated statistics – within the topic areas considered by the evaluation this was particularly notable in education, health, and criminal justice;

  • Greatly improved accessibility of data through its availability online (though the volume of data available also produces challenges for users).

Part of the remit of the EOC project was to identify gaps. There remain gaps in data in certain areas, and a need for better data was identified in the following areas:

  • Public sector workforces

  • Gendered patterns of childcare and other forms of caring

  • Gender data on poverty and social exclusion

  • Data on the incidence of domestic violence and other forms of violence against women

  • Travel data, linking employment status and parental status to patterns of transport use

  • Data on access to and use of services/resources – housing, childcare, health, transport, and on time costs of accessing services.

There is also an issue concerning multivariate analysis, with there being continuing problems with limitations to disaggregated statistics by gender and ethnicity, and by gender and disability, though more such data are available than was previously the case. Currently there appears to be no systematic approach to analysis combining age and gender (with the caveat that relevant age groups will differ according to topic area), or gender and socio-economic status.

As well as gaps in statistical data, there are other ways of thinking about gaps:

  • Accessibility – no single point of access, data spread across multiple locations, variability in user friendliness and intelligibility, lack of regular publication of some types of data/publication.

  • Capacity – variation in capacity of producers of statistics; variation in capacity of policy makers to understand and make use effectively of statistical evidence; limited number of specialists/dedicated posts.

  • Communication and dissemination – evaluation of this and knowledge about audiences and user groups is limited; engagement with users, such as consultations, etc, do not appear to have been consistently carried out, and it is not clear how effective such consultations have been in leading to improvements.

Having identified that gaps of various kinds still exist, it is important to stress that we now have a very large volume of gender disaggregated statistics being regularly published on a wide range of topics. Thus, gaps in data are not necessarily the major barrier to using gender statistics to inform policy. What is of crucial importance, however, is the identification of areas of policy where there is an absence of significant data, and where this absence makes impossible the effective elaboration of policy objectives or evaluations of their outcomes.

The key conclusions I have drawn from this evaluation of gender statistics are that:

  • There is a need to take stock of the data which are already available and to promote better use of these. This may be regarded as a shared responsibility, between the academic research community, producers of statistics, public bodies, and women’s organisations, etc. Training is needed as an essential component of the promotion of better use. Methods of dissemination are another essential component – if it is not possible to have single point of entry for all data given the volume of data currently available, then it should be possible to have single point of entry for key data, with links to relevant data sets, across a range of significant topic areas.

  • There is a need to find ways of prioritising use and development of gender statistics e.g. identification of key indicators relevant to key areas of gender inequality for regular publication, and to provide a general framework for gender equality schemes and setting of objectives; identification of key areas for improvement of data, new data etc.

  • There is a need to look at ways of structuring engagement with users of gender statistics in a way that takes account of different levels of expertise, and in ways that can be consistently followed through.

In sum: there is a large volume of data available; we need to make better use of this; we need to find ways of focussing and prioritising uses of data in developing indicators and addressing significant gaps; and we need to look ways of improving dialogue between producers and users of gender statistics.

Prioritising and selecting – developing key indicators

The theme I wish particularly to focus on here is that of the development of a selection of key indicators of gender equality, because it seems to me that this would be a particularly important step in both using data better and providing a focus on key areas. However, I think also that it is important that this is seen as an integral part of a multi-faceted strategy on gender statistics and gender equality indicators, which should involve:

  • Regular provision of guidance on data sources;

  • Regular publication of gender disaggregated statistics, including compendia of statistics, in-depth studies, and bulletins for specific topic areas;

  • The development of selected key indicators, which may be simple or composite indicators, and the development of performance indicators.

Progress has been made across the UK and devolved administrations in a number of these areas, particularly with respect to guidance on data sources, improved access to gender disaggregated statistics, occasional publications including compendia of statistics and topic related studies. However, commitments have not been made to regular publications of compendia of data. Of the recommendations listed above, the least progress has been made with respect to indicators. There has not yet been any adoption of a set of key indicators, or much in the way of performance indicators relevant to gender equality or attempts to develop composite indicators.

It must be acknowledged that the term ‘indicator’ can be confusing, as there are different types of indicators with different functions. An indicator can be defined as something that summarises a large amount of information in a single figure, indicating changes over time, and providing comparisons with a norm; or as statistical information which sheds light on a particular economic, demographic or social problem or question. Indicators can be used for a number of purposes: to compare the position of women and men; as a measure of results of policies; as an awareness raising tool. This means that compendia of data, such as the WEU report Key Indicators of Women’s Position in Britain, can be regarded as providing a set of indicators, albeit not ones that are closely related to policy objectives.
It is important to stress that both reports such as the WEU report and UN type high-level indicators, such as the Gender-related Development Index and Gender Empowerment Measure, provide indicators of social trends comparing the position of women and men. They do not provide data which can measure the impact of specific policies i.e. performance indicators. It can be argued that in principle policy objectives should be measured according to performance indicators which assess gender equality impacts. In the UK to date there has been resistance to the elaboration of performance measures with respect to gender equality issues, such as specific target setting. Furthermore, even where targets exist it may be very difficult to work out to what extent specific policies produce change, rather than other factors, or to what extent meeting targets – which may often be crude headcounts – actually produces change in terms of gender equality, in the broader sense of changing the balance of power relations, or access to resources. This is a very complex area, which will inevitably require more attention as public bodies go down the route of carrying out gender impact assessments. But at the present moment, there is limited experience of this and still a long way to go in developing appropriate performance indicators.
My focus here then is on indicators of social trends, and what I want to argue for is a selection of these types of indicators to provide a framework for measuring changes in the comparative position of men and women, for different jurisdictions and/or geographies, and which would be published on a regular basis, preferably annually. Though such indicators would be indicators of social and economic change, not indicators that can be used to judge the effectiveness of specific policies, I would however expect these indicators to be a key part of the framework in which policy options are considered and in which policy decisions are made, just as, for example, unemployment rates condition economic policies, and figures on child poverty or households below average income condition policies on benefits and taxation.
Selected international examples of indicators

I want briefly now to look at some international examples of indicators, before looking at the types of gender disaggregated statistics produced for the UK and for devolved administrations, from which selected indicators might be developed.

Selected examples of indicators

Organisation and



United Nations

Gender-related Development Index (GDI)

Gives gender disaggregated data and adjustments of the Human Development Index for women and men and compares 177 countries on four basic measures: life expectancy, adult literacy, educational participation, and estimated earned income.

United Nations

Gender Empowerment Measure (GEM)

Measures women’s participation in the economic, political and professional fields, using measures of: seats in parliament; female legislators, senior officials and managers; female professional and technical workers; and ratio of estimated earned income.

European Commission

Reports on equality between women and men

Compares the position of EU member states on the following measures: employment rates; unemployment rates; part-time work; reconciliation of work and family life; the pay gap; decision-making; education, training and research; healthy life expectancy and the average age of women at the birth of their first child.

Status of Women Canada

Economic Gender Equality Indicators

A series of indices intended to provide a comprehensive picture at aggregate level. The indices are: Total Income Index; Total After-tax Income Index; Total Earnings Index; Total Workload Index; Paid Work Index; Unpaid Work Index; University degrees granted; Job-related Training Index; and Occupational Returns on Education. These indices use ratios of women to men, enabling a measurement of the position of women relative to men.

Institute for Women’s Policy Research, USA

Index of women’s status

This index is used to compare the status of women in different states. Data are presented for 30 component indicators, as well as for five composite indices, which rank states in each of five domains: political participation; employment and earnings; economic autonomy; reproductive rights; and health and well-being.

Statistics Norway

Gender equality index for Norwegian municipalities

This is an index which combines various direct and indirect measurements of gender equality that show the extent to which women and men participate in politics, education and working life. The indicators are:

  • Kindergarten coverage for children aged 1-5

  • Number of women per 100 men aged 20-39

  • Education levels for women and men

  • Income for women and men

  • Percentage of female municipal council members

Statistics Sweden

EqualX – Gender equality index

This is an index that compares municipalities or counties. The index is a weighted sum based on 13 variables. The variables in the index are as follows:

  • People with post-secondary education

  • People in gainful employment

  • Job seekers

  • Average income from gainful employment

  • People with low income

  • Unequal sex distribution by industry

  • Days of parental leave benefit

  • Days of temporary parental leave benefit

  • Sickness rates

  • Young adults (25-34 years)

  • Women/men in municipal council

  • Municipal executive board

  • Entrepreneurs with at least 1 employee

The UN regularly publishes both the Gender-related Development Index (GDI) and the Gender Empowerment Measure (GEM), both of these being composite indicators ranking a large number of countries. Because these include many countries where statistical data are limited, the elements that make up the composite indices are relatively limited. Developed countries score highly on the GDI index, but may have a different ranking on the GEM. For example, in 2004, the UK ranked 9th on the GDI, but 18th on the GEM. There are also a set of EU indicators, which are published annually. These are tied to key EU policy objectives on women’s economic participation, and compare the position of member states. It would seem to me to be a logical starting point for UK/devolved administration indicators to include both UN and EU indicators, as a means of providing a framework of comparability with other countries.

The other indicators in the table are a small selection, which show possible models for further developments here. The Canadian Economic Gender Equality Indicators provide, for example, a useful way of comparing the division of labour between men and women in paid and unpaid work, and also the differential impact on men and women of caring responsibilities through the Occupational Returns on Education Index. The Institute for Women’s Policy Research in the USA is an index of women’s status rather than a gender equality index, and uses composite indicators to provide rankings. The production of this set of indicators differs from the other examples cited, in that it is produced by an independent women’s organisation, and not a government body or national statistical service. The Statistics Norway and Statistics Sweden models both provide a means of ranking municipalities, though differ in the number of indicators used.
Generally speaking, what these examples have in common is a predominant focus on decision-making, education, employment and pay. Several also include data about childcare, though there is a variation in how this is treated, while some also include data about health, similarly with some variation in how this is treated. This might suggest then that these areas should be regarded as core indicators for gender equality. It also seems to me that the local authority rankings produced in Norway and Sweden offer possible models for development here. Given the introduction of the Gender Equality Duty local authorities will all be producing equality schemes, and this offers a possible opportunity for collaborative effort to produce a set of indicators common to local authorities and which would allow comparisons to be made.
Gender statistics in the UK/devolved administrations

Looking at a selection of reports on gender disaggregated statistics in the UK/devolved administrations produced in recent years, we can see the range of areas typically covered. The table below provides a simple comparison of the chapter headings in four reports, not necessarily following the order in which they appear in the publications, but re-ordered in the table in order to indicate the areas common to all reports.

The reports are somewhat different in character: the WEU report (updated in 2004 – see Aston et al) and Scottish Gender Audit report both provide a wide range of tables with interpretative commentary; the report on gender equality indicators for Northern Ireland recommended regular publication of gender statistics in selected areas, based on a review of existing data; while the Measuring Up report for Wales contains a focussed selection of indicators, indicating changes in gender differences in the last ten years, and uses a ‘traffic lights’ system to indicate where change has moved towards equality, stayed the same, or moved in the direction of greater inequality.
Topics covered in reports in UK/devolved administrations

Topic areas in Key Indicators of Women’s Position in Britain,

WEU, 2002

Topic areas of indicators recommended in Gender Equality Indicators for Northern Ireland, OFMDFM, 2004

Topic areas of indicators in Measuring Up, Bevan Foundation/EOC Wales, 2006

Topic areas in A Gender Audit of Statistics: comparing the position of women and men in Scotland, Scottish Executive, 2007

Population and demography



Population, households and families

Economic activity and employment



Labour Market

Work-life balance

Childcare; other forms of caring


Care and Caring

Education and training

Education and training

Education and learning

Education and training

Financial resources

Earnings; Income

Financial resources

Income and wealth

Participation in public life


Public life

Participation in political and public life






Violence and crime

Crime and safety

Crime and justice



Personal mobility










Notwithstanding the different functions of these reports, it is evident that there is a set of core topics common to all these projects, though there are also some variations in coverage e.g. population and demographic data, attitudinal data, data relating to caring, and housing, and also some variation in how data have been organised. Essentially all these reports have covered political and public life/decision-making, education and training, employment, earnings and other forms of income, childcare, health, crime – including violence against women, and transport.

Taking together the core areas covered by international examples, and the core areas covered in recent reports in the UK and devolved administrations – which have also shown where gender disaggregated data already exist – this would suggest a selection of indicators covering the following areas:

  • Political and public life/decision-making

  • Education and training

  • Employment

  • Earnings and income

  • Caring

  • Health

  • Violence against women.

I would argue strongly for these areas to be regarded as the key areas in which to agree a selection of indicators, though this is not necessarily straightforward, and in some areas better and/or new data are required, for example, in measuring time spent in caring, and in measuring the prevalence of domestic violence and other forms of violence against women. With respect to other areas, such as other aspects of crime and victimisation, transport, housing and homelessness, while it is important that gender data continue to be made available in all these areas, it is less clear how indicators of gender equality might be developed, or whether the data permits this.

There are also conceptual questions. To what extent can selected statistical measures in the topic areas outlined be measures of gender equality? Some of the areas listed above – education, employment, pay, incomes, care – seem to me to be a key cluster of inter-related areas, in which the broad policy goal might be to achieve average equal outcomes, though there are of course a range of policy routes that might deliver such outcomes. However, other key policy areas such as health, crime and justice, housing, and transport, seem to me more problematic when it comes to devising measures of gender equality, and/or gender equality targets. This does not mean to say that there should not be monitoring of gender disaggregated data, and, where appropriate, gender specific data, but aims may be conceptualised sometimes as about gender equality and sometimes about gender equity – to take into account areas where there are gender specific needs or gender specific patterns of behaviour, where it is not possible to make a straightforward comparison between women and men.
For example, domestic violence/abuse is a significant issue for women. This can be conceptualised as a gender equality issue, since domestic violence/abuse inhibits the capacity of women to achieve their potential, and can result in women having to move around to avoid a violent ex-partner, experiencing instability, poverty, and poor housing. But it can also be conceptualised as a human rights issue. Furthermore, while it may be understood as a largely gendered phenomenon in which women are overwhelmingly the victims and men overwhelmingly the perpetrators, there are also male victims and female perpetrators, though research suggests that the patterns and seriousness of abuse tend to be different (see Gadd et al, 2002). There is also the issue of (non-domestic) violence perpetrated by men against other men. If a set of gender equality indicators includes the incidence of violence against individuals, should this only be domestic violence, or other forms of violence?
To take another example, with respect to transport, what might gender equality goals be? Should it be equalisation of car licence holding – with this implying effectively further growth in car ownership/use by women – which conflicts with environmental policy goals? Furthermore, the relationship between transport services and usage, labour market participation, childcare responsibilities, and access to other facilities and services is complex. Arguably policy responses need to be not just about encouraging more use of public transport, but about addressing the ways in which the structure of public transport services interacts with current gender divisions within the labour market, caring and domestic responsibilities, and about the extent to which restructuring services can facilitate changes towards equality, rather than entrenching current divisions. Developing measures of equality impacts of transport policies therefore presents a considerable challenge.
This suggests that in certain areas it will be easier to develop measures of gender equality than in others, especially in those areas where it is possible to make a relatively straightforward comparison between the position of women and men. It is also the case that only in some areas which I have recommended for the development of key selected indicators are statistical data currently adequate to the task. For example, there need to be improvements in data on various aspects of decision-making; individual measures of poverty; time use studies; and the incidence of violence against women. But we can start with what we have got, and work to improve data over time.
Given the range of gender statistics now available, the selection of specific indicators within the broad areas listed above, and whether or not to go down the route of constructing composite indicators, should be the next stage in the process. This would necessarily involve discussion and negotiation between government statisticians, the EOC/CEHR, academic and other experts, and women’s NGOs, etc. However, at this stage some criteria for the selection of indicators that should be regularly reported can be put forward. These are:

  • Quality of the data i.e. reliability, robustness

  • Clarity

  • Comparability (intra-UK, EU, and international comparisons)

  • Frequency of availability

  • Capacity for trend analysis

  • Capacity to provide profiles for relevant groups and areas (multi-variate analysis including ethnicity and disability, etc; UK and constituent countries, regional and local areas).


What would be the process through which such indicators might be developed? If it is accepted that there should be agreement on a relatively small number of selected indicators for regular publication, and which may be deemed to measure progress towards gender equality in our society in general, then ideally these should be a set of indicators that are endorsed by government and government statisticians. Any process of developing such indicators should ideally involve a range of constituencies: government departments and statisticians; key public sector bodies such as local government and health boards; academic researchers; women’s organisations; EOC/CEHR; and trade unions, etc.

The introduction of the gender equality duty is going to result in the production of a large number of equality schemes by a range of public bodies. It is anticipated that many of these bodies will make use of gender disaggregated statistics produced by government statisticians, and by public bodies themselves. This will present challenges for many public bodies both in familiarising themselves with existing data, developing new data, analysing data and so on. A general framework of regularly published indicators at national levels (UK and devolved administrations) would be helpful in giving a focus to the setting of objectives within gender equality schemes, as well as measures against which regional and/or local comparisons could be made. Particular sectors might also wish to give consideration to the development of a set of common indicators e.g. local government could have a set of indicators for local authority areas. Thus the introduction of the gender equality duty provides an opportunity for the selection of key indicators as a common framework to guide public bodies in setting gender equality objectives.
Summing up

To sum up:

  • The provision and availability of gender disaggregated statistics in the UK and in devolved administrations has improved vastly since the late 1990s, and a current key challenge is to make more effective use of the data that already exist;

  • One way of making more effective use of these data would be to develop a set of selected key indicators for regular publication, and it is already the case that in a number of key areas adequate data exist, though in other areas more work needs to be done e.g. care, domestic violence;

  • The context of the gender equality duty provides an opportunity for collaborative work between government departments and statisticians, academics and other gender equality experts and organisations, and public bodies, to produce such a set of indicators;

  • Such indicators should provide the capacity for comparisons to be made internationally - with UN and EU indicators, and within the UK between its four constituent countries;

  • The possibility of developing a set of indicators for ranking local authorities should also be considered.

Above all, the key thing I wish to emphasise is the need for regularly published selected indicators, that are relatively easy to understand, and that are accessible to a range of users and to the general public. This would help increase understanding of the nature and persistence of gender inequalities in our society, would provide a means of measuring change, and would provide a framework for holding public bodies to account in fulfilling the gender equality duty.


Aston, J, Clegg, M, Diplock, E, Ritchie, H and Willison, R (2004) Interim Update of Key Indicators of Women’s Position in Britain, London: Women and Equality Unit, Department of Trade and Industry
Bevan Foundation/EOC Wales (2006) Measuring Up: Progress towards Equality for Women in Wales, Blaenau Gwent: Bevan Foundation
Breitenbach, E (2006) Gender Statistics: an evaluation, Manchester: Equal Opportunities Commission
Breitenbach, E and Galligan, Y (2004) Gender Equality Indicators for Northern Ireland: A Discussion Document, Belfast: Office of the First Minister and Deputy First Minister
Breitenbach, E and Wasoff, F (2007) A Gender Audit of Statistics: comparing the position of women and men in Scotland, Edinburgh: Scottish Executive Social Research
Dench, S, Aston, J, Evans, C, Meager, N, Williams, M, and Willison, R (2002) Key Indicators of Women’s Position in Britain, London: Women and Equality Unit, Department of Trade and Industry
Gadd, D, Farrall, S, Dallimore, D and Lombard, N (2002) Domestic Abuse Against Men in Scotland, Edinburgh: Scottish Executive Central Research Unit
Weblinks for gender equality indicators

UN Gender in Development Programme
Status of Women Canada – Economic Gender Equality Indicators
European Commission – Report on equality between women and men, 2005
Institute for Women’s Policy Research, USA
Statistics Norway – Gender Equality Index for Norwegian Municipalities
Statistics Sweden – EqualX – Gender Equality Index

The database is protected by copyright © 2016
send message

    Main page