The Scottish Government

« Previous | Contents | Next »

Listen

LEARNING FROM EXPERIENCE: LESSONS IN MAINSTREAMING EQUAL OPPORTUNITIES

SECTION 2 MAINSTREAMING IN PRINCIPLE AND PRACTICE

In Section 1 we noted that mainstreaming as a strategy is often confused with mainstreaming as a tool. Rees distinguishes between mainstreaming principles (such as integrating equality, taking a holistic approach, promoting values of participation and consultation, building ownership; and 'visioning' equality) and mainstreaming tools (such as gender monitoring, gender impact assessments and awareness raising). The tools are needed to put the principles into practice, but the existence of tools alone does not necessarily imply that mainstreaming is underway. It is also important to note that mainstreaming is not only concerned with process but also with outcomes; it is a social justice-led approach to equalities which seeks to transform structural discrimination and contribute to the creation of a fairer society. ⁴⁴

There is a need, therefore, to clarify the uses and purposes of tools. The first point to note is that mainstreaming tools are diverse both in terms of type and in terms of what they are intended to achieve. Most importantly, tools will only be useful if there is an understanding of wider contexts of aims, objectives, structures and the like. It is useful to think of mainstreaming in terms of principles, integrated systems, framework tools and discrete tools and techniques.

Principles involve the underpinning values and goals of mainstreaming; integrated systems involve the strategies, structures, processes and mechanisms through which the aims and objectives of mainstreaming are operationalised and implemented; framework tools operate at a lower level and can be part of integrated mainstreaming systems, they involve the use of a cluster of tools to achieve a broad aim whereas discrete tools are used at specific instances to achieve more specific aims. An example of framework tools (sometimes called mechanisms) would be Gender Impact Assessments (GIAs) where a collection of discrete tools such as disaggregated data, consultation exercises, manuals, and so on are combined with awareness-raising and training to aid policy-makers to integrate a gender perspective into their work. While this level of tool lies above that of the discrete tools, it is not of itself mainstreaming. This is an important point as confusion about this distinction is noticeable in many case studies. Successful use of a GIA may mean that the needs of some women have been mainstreamed into service design and provision in that area, but mainstreaming requires a whole set of structures, procedures, and cultural changes as well as the use of these frameworks for achieving specific aims.

This section first focuses upon systems of mainstreaming and examples of framework tools. All of the above require, as well as many other things, the collection of information and statistical data on women. We therefore look at gender disaggregated statistics and Gender Equality Indicators, in particular the GDI and GEM, as tools which have been developed for this purpose. Finally, the range and purposes of mainstreaming tools and techniques are considered.

MAINSTREAMING SYSTEMS: STRUCTURES, PROCESSES AND MECHANISMS

The Gender Management System developed by the Commonwealth Secretariat, provides a model for the integration and strategic implementation of an equalities perspective at national government level.

The Commonwealth Secretariat Gender Management System Series

This series of guidebooks and manuals details the Commonwealth's model for implementing a mainstreaming strategy for gender equality. The Gender Mainstreaming System (GMS) has been designed to help governments implement a mainstreaming approach to gender equality and to suggest frameworks for monitoring and evaluation. While the handbooks are aimed primarily at governments and advancing gender equality through promoting political will, the importance of creating partnerships between government, the private sector, and civil society is emphasised. GMS is based on a 'stakeholder' approach and on three broad principles: empowerment (representation in decision-making bodies and control over resources), integration (transformation of existing structures), and accountability.

The concept and methodology of GMS has been developed to assist Commonwealth governments in realising the vision of the 1995 Commonwealth Plan of Action on Gender and Development, and is connected with the move from a Women In Development (WID) to a Gender and Development (GAD) approach within development theory. Unlike the former approach, which attempted to incorporate women into structures designed to 'male' norms, the GAD approach attempts to change the structures and the norms on which they have been based. It recognises not only the differences between women and men but also differences within these groups and the possibility for multiple lines of discrimination /disadvantage. Mainstreaming is seen as a systematic and holistic approach which, by integrating a gender perspective into all areas of an organisation's work, can transform that organisation. Despite being designed from a development perspective, these handbooks are useful to all governments interested in implementing a mainstreaming strategy. The series covers the sectoral areas of development planning, finance, the public service, education, trade and industry, agriculture and rural development, information and communications, as well as equal employment opportunities policy, gender-sensitive indicators, and a general GMS handbook. Forthcoming handbooks will continue with the sectoral approach as well as introducing 'themed' booklets on issues with particular relevance to gender equality. The handbooks are accompanied by quick guides which provide key points and recommendations. ⁴⁵

The GMS has three main components; structures, mechanisms and processes, which need to be developed within an enabling environment. The system is diagrammatically represented in Figure 2.1.

Structures

Turning first to structures: six key components are identified which work to drive, co-ordinate and deliver mainstreaming.

A Lead Agency: usually the national women's policy machinery e.g. Ministry for Women's Affairs, Office for the Status of Women.

Roles: initiating and strengthening institutional arrangements; co-ordination; monitoring; advocacy; communications; media relations; reporting.

A Gender Management Team: includes representatives of the Lead Agency, other key government ministries, and civil society.

Roles: leadership for implementation; defining broad operational policies; indicators of effectiveness; time frames; monitoring performance.

Gender Focal Points: designated senior staff members in each ministry or department.

Roles: Co-ordinates gender activities (e.g. training); responsible for promoting mainstreaming within their ministries.

Inter-Ministerial Steering Committee: members are the Gender Focal Points and representatives of the Lead Agency.

Roles: Ensures mainstreaming in all ministries is effected and that strong links are established between ministries.

Parliamentary Gender Caucus: comprise committed members of legislature.

Roles: Raises awareness; lobbies; promotes equal participation of women and men; scrutinises parliamentary structures, procedures and matters under debate.

Gender Equality Commission / Council: representatives of civil society such as academic institutions, NGOs, professional associations, media and other stakeholders.

Roles: actively represents and advocates the interests of civil society in government policy making and implementation processes.

Levers

Within the GMS framework various levers of change are identified. These can be classified as:

Communication levers: information flows are crucial which not only give an accurate picture of current gender differentials within an organisation and society as whole and allow a diagnosis for appropriate action; but also continue to provide information which can inform ongoing decision making;

Awareness levers: these levers are necessary to challenge and change institutionalised discrimination and stereotypical norms and values within an organisation and within society more generally;

Incentive/boundaries levers: with the increasing emphasis upon performance -based public service, these levers use incentives ( that is, rewards for positive achievements) and boundaries (which define the limits of acceptable behaviour and the sanctions the organisation will enforce if behaviour is unacceptable - that is, discriminatory).

Figure 2.1 Commonwealth Gender Management System

Source: Commonwealth Secretariat Gender Management System Handbook, 1999, p.13

Mechanisms

The following specific frameworks and tools are highlighted.

Gender Analysis: which assesses how policies, plans, and programmes impact differentially on men and women and upon different groups of women and men ( communication and awareness levers).

Gender Training: promotes gender awareness and sensitivity ( awareness lever).

Management Information System: communicates critical information throughout the GMS, e.g. gathering and dissemination of sex disaggregated data ( communication lever).

Performance Appraisal System: is modified to become gender sensitive. Thus performance targets will explicitly include gender equality goals. This ensures accountability is the basis of decisions on what incentives or sanctions can be applied in each case ( incentive/boundary lever).

It is argued that through the appropriate use of a combination of tools which trigger these three types of lever a process of cyclical and planned change can be implemented. Each repetition of the cycle results in progress and brings the organisation closer to its goal of gender equality and gender equity.

The system is disseminated through a series of handbooks for specific sectors (e.g. finance, education) which provide models for implementing a GMS in the given area, including examples of what this might mean in practice. The handbooks explore the connections between gender and the sector in question, discuss relevant tools and how to use them and give recommendations for action. Case studies are used to demonstrate good practice and to identify requirements, needs, and problems. Each handbook sets this information out in the context of the development and aims of GMS. The handbook for education includes analytical tools and guidelines for conducting a gender impact analysis of government policies, plans, programmes and projects in the formal and non-formal educational sector. The Guide combines indicators and key questions which can be tailored to the specific context and task. Several of the handbooks contain glossaries of key terms used. Models for establishing a GMS, engendering national budgets and macroeconomic policy, creating 'gender-aware medium-term development plans' and promoting gender awareness in the planning cycle are also given.

Other examples of integrated systems include Canada's Gender Based Analysis model and British Columbia's Gender Lens model. The structures, processes and mechanisms in place in these systems are considered in the country examples sections. Here we consider some of the framework tools used to facilitate mainstreaming: Guides to Gender Analysis and the British Columbia Gender Lens guidelines are aides to policy-makers, highlighting gender issues and questions to be considered at all stages of policy development, implementation and evaluation.

GENDER LENS AND GENDER BASED ANALYSIS

The Canadian Federal and British Columbian Provincial guides to gender analysis are aides for policy-makers to be used at all stages of the policy cycle. They raise questions and suggest actions which should be taken at all stages of the policy development, implementation, communication and evaluation procedures. In Canada, the term Gender-Based Analysis is synonymous with mainstreaming as understood in the British and European contexts. As is evident from the Canadian case studies in Section 6 of this report, mainstreaming is at an advanced stage in Canada and has benefited from high levels of political support and resources.

The British Columbian Guide to Gender-Inclusive Analysis, 'Gender Lens' breaks the policy-making process down into eight phases, for each of these key issues and questions relating to gender equality are raised in order to help policy makers integrate a gender perspective into their work and to develop evidence-based policy and decision making. The eight phases are:

1. Identify the Issue

2. Define Goals and Options

3. Define Information and Consultation Needs

4. Conduct your Research

5. Develop and Analyse Options

6. Make Recommendations/Decision-Seeking

7. Communicate the Policy/Programme

8. Evaluate the Policy/Programme.

The guide includes hypothetical examples to demonstrate the gender impact of a particular policy/programme at each stage as well as quotations from women relating to each stage outlining some of the issues and problems raised. For each stage the guide addresses three questions. ⁴⁶

• What to consider
• What to ask yourself
• How to ...

The Federal level Gender-based Analysis Guide: Steps to incorporating gender considerations into policy development and analysis comes in similar style and format to the British Columbian guide. It is supported by a Gender Analysis Backgrounder which can be found at the web site of the Women's Bureau Human Resources Development Canada. ⁴⁷ This booklet sets out the reasons for doing GBA, what this means for policy-makers and factors to be considered. It looks at the public policy implications of gender trends in the economy, and addresses frequently asked questions about GBA. These critical questions are countered with factual examples and explanations. In this instance, six steps are identified, which are reproduced in the following figure.

Figure 2.2 Summary of Canadian Gender Analysis Steps

Source: Gender-based Analysis Guide, Summary of Gender Analysis Steps, HRDC, March 1997.

Gender analysis questions are identified for each stage and hypothetical examples are used to illustrate the questions. ⁴⁸ Further discussion of the development of mainstreaming in Canada can be found in Section 6, together with examples of mainstreaming in action.

GENDER IMPACT ASSESSMENT: THE NETHERLANDS AND FLANDERS

Gender Impact Assessment tools are another example of framework tools or mechanisms which involve the use of several techniques. GIAs are based on tools designed to assess the likely impact of policies on the environment (Environmental Impact Assessments). They are ex-ante procedures which means that the impact on gender relations is assessed before the final decision on a given policy proposal is taken. Modified GIAs have also been used to assess the impact of existing policy on women and men in order to create baseline data and as awareness-raising or educational tools. GIAs work to expose the gap between assumptions about policy target groups and reality, challenging the gender-neutral or androcentric bias of much policy and reformulating it to take account of gendered reality. Thus GIAs are often promoted as tools which enable policy-makers to make better policy, that is, relevant to the needs of the target group, effective, and able to achieve its aims. Gender Impact Assessment is based on the argument that policy needs to be firmly rooted in an analysis of the actual experiences and diversity of the population.

GIAs can have weak or strong effects depending on how and when they are conducted. GIAs conducted early in the policy process have a stronger effect on the formation of policy than those conducted near the end or after policy has been made. Serious attempts to increase participation and consultation as part of this process will also increase the impact of a GIA. Reactive analyses, which aim only at being non-discriminatory will have a lesser effect than proactive analyses which aim to enhance equality.

Gender Impact Assessment has been pioneered in the Netherlands by academic Mieke Verloo. ⁴⁹The Dutch GIA (the EER) assesses the potential impact of a policy proposal before the final decision on that proposal is taken. Ideally, the EER works to develop a 'virtuous circle' in the policy-making process whereby the results of the EER are fed back into the policy-making system to create better policy proposals.

EER is based upon a theoretical analysis of the factors which influence and contribute to gender inequality in Dutch society. It has two parts, the first asks whether the policy is relevant to gender issues and establishes the baseline situation in the policy area. The second uses the theoretical framework to analyse the likely impact of the policy in terms of gender equality.

The second stage is carried out by considering three variables: the basic structures central to gender inequality; the basic processes which constitute and reproduce gender inequality; and the central structures for advancing gender equality. More specifically: stage one considers the division of labour and the organisation of intimacy; stage two addresses the distribution of and access to resources and the existence of "rules" (interpretations and norms connected to gender); stage three assesses the impact of the policy upon equality, (equal rights and (un)equal treatment of (un)equal cases) and, autonomy (possibility for women to make decisions about their own lives).

Most EERs have been carried out by academics, sometimes in conjunction with civil servants. The tool requires a fairly high level of gender expertise and its effectiveness would be compromised by the lack of this. The tool, when used successfully, also works as an 'eye opener' or awareness-raising educational tool. It demonstrates to policy-makers the unintended consequences of their policies and the need to take gender into account. To date, five years from the development of the tool, nine EERs have been conducted and two are in process. The areas involved have included the Ministries of Education, Justice, Agriculture, Environment and Fisheries, and Domestic Affairs. As it has only proved possible to conduct EERs on a very small proportion of the new policies which have emerged, the process of selection is seen as very important. Verloo suggests that screening committees could be established to consider all policies and then decide whether an EER is necessary (as is the case with environmental impact assessments in the Netherlands); or that instruments such as SMART be used to assess whether a policy has gender relevance. ⁵⁰

Some positive outcomes resulting from the use of EERs have included: ⁵¹

• Demonstrating that what had been viewed as a 'gender neutral' plan to restructure secondary school education would, in reality, reinforce educational gender segregation;
• Showing that a supposedly 'gender neutral' reform of the electoral system would result in the reduction in the number of elected women politicians;
• Highlighting that a strategy for chronically sick people had been designed around the needs and lives of young men, whereas most of the chronically sick are older women.

A recent review ⁵² of the instrument concluded that it is in principle a good instrument, though changes are necessary to the context in which it is used. The tool needs to have a more central position in the policy-making process, being used at the earliest possible point, political support needs to be increased, and the findings of the tools need to be translated into alternatives, recommendations and guidelines for use by the people who commission it. The evaluation found that EERs were most successful when applied jointly by external experts and internal policy makers.

Following the EER, other instruments have been developed and used. For example, the Gender Impact Test, a simpler tool, has been developed and used at the local and regional level. In 1998, the Dutch Association of Local Authorities also designed a GIA for the local level. ⁵³

As noted earlier, the Dutch GIA (EER) is underpinned by sophisticated theoretical understandings of gender and has been developed within the context of high levels of academic expertise about gender relations in the Netherlands and a specific political history of emancipation policy. As such it is not easily 'exportable' to other countries with different policy and political contexts. ⁵⁴

The Flanders GIA was developed in 1997 by academic researchers who were asked to produce a tool similar to the EER. Its developers had thought the Dutch EER was unsuitable for Flanders due to four reasons: the high degree of specialisation needed; the cost in time and money; the dependency on the quality of information; and the omission of a requirement for the construction of alternatives. ⁵⁵ The instrument was modified and simplified from the more sophisticated Dutch prototype to take into account the fact that Flanders had little or no tradition of gender expertise or equality policy; equalities work is further hampered by the lack of gender disaggregated statistics which map the relative positions of women and men in society at either national or regional/community level; and by a lack of political and bureaucratic commitment. The instrument is more educational and process-oriented than the Dutch EER. It does, however, still require some knowledge of gender issues for its correct implementation and therefore depends on the administration's capacity to develop such knowledge.

The tool developed consists of three stages: stage one, to trace the gender dimension of a policy proposal; stage two, to estimate its size; and stage three, to formulate alternatives where necessary. To enable the tool to be used by policy-makers who do not necessarily have any previous experience of gender work, the first step is designed to require less expertise and to stand alone if necessary. The Flanders GIA exposes inconsistencies between the basic assumptions of the policy proposal and the real situation in the policy field in terms of gender. At present, due to the lack of gender expertise it is used primarily to raise awareness by confronting policy-makers with a mechanism which makes them look at their own policy proposal from a different angle. The instrument is accompanied by a handbook, explaining the aims and concerns behind such a tool, as well as its operation. ⁵⁶ Unlike the Dutch EER, the tool has not been used since it was completed in 1997. ⁵⁷

Complexity versus Usability

The differences between the Dutch and Flemish GIAs, resulting from the different contexts of these countries, give rise to an issue around the use of 'experts' versus 'ordinary policy-makers'. The mainstreaming strategy aims to integrate a gender perspective into normal work at all levels of an organisation. Therefore existing staff should be enabled to incorporate this into their daily work. While bringing in expertise is recognised as a necessary and desirable measure to aid staff and to guide mainstreaming work, if gender analysis remains solely in their domain, it may not be correct to say that mainstreaming has been implemented. ⁵⁸ This issue creates a tension in tool design/creation which is evident in comparing the Dutch and Flanders examples. On the one hand, there is a desire for a sufficient level of complexity to create a subtle and useful tool; on the other hand, there is a need for simplicity so that the tool can be used effectively and without sole reliance on 'experts'. If a tool is too complex, it is likely to be reduced to simple questions or check lists which do not really address the issue. This was seen in the case of the Dutch Gender Impact Test, developed for local government level. This problem is also apparent in the tool discussed below, Levy's Web of Institutionalisation.

WEB OF INSTITUTIONALISATION

The Web is a complex framework for gender mainstreaming which requires a high level of understanding of gender equality, relations and power structures. It was designed by Caren Levy of University College, London (UCL) to facilitate the mainstreaming of gender in planning, policy, programmes and projects, primarily in the context of development work. The tool is used to assess the extent to which mainstreaming is underway, and the strengths and weaknesses of the process, in a specific context. It also considers the extent to which practices have become institutionalised, that is, the extent to which the process has been absorbed into institutional norms. The Web can then provide the basis for defining strategic action by highlighting Routes Through The Web which will strengthen the mainstreaming process. ⁵⁹

The Web comprises thirteen elements which each represent a site of power. The analyst considers these sites of power and the relationships between them in order to develop a picture of the opportunities or resistance to the change which gender mainstreaming is attempting to bring about. For some purposes, the elements in The Web can be grouped into different spheres of influence, for example, the citizen, the policy, the organisational and the 'delivery' spheres. The form the elements take is context specific. The tool needs to be combined with methodological tools for understanding gender relations in society and in policy intervention. These could be around gender roles, access to and control over resources, gender interests and need, and their combination in different policy approaches. A set of indicators to monitor the process of gender mainstreaming can be defined from each element in The Web.

The problem of complexity versus useablility is raised by Levy herself,

A key challenge has been how to develop methods to apply The Web that are manageable without losing the necessary complexity of the tool. This has been done primarily through the experience of training and is an ongoing process. ⁶⁰

THE "3RS METHOD"

In 1995 the Swedish Government allocated funds to the Swedish Association of Local Authorities (SALA) to develop its JAMKOM project. The JAMKOM initiative seeks to establish methods of analysing and promoting equalities work in local government and to share and publicise good practice. The '3Rs method' evolved out of this initiative as a tool for incorporating a gender perspective into the work of local authorities. The method has also been used in other Nordic countries.

The primary objective of the method is to increase understanding about the division of power between men and women and the structures and norms which hinder gender equality, and to identify the ways in which these need to be changed. The tool enables the user to conduct a systematic gender analysis, assessing the position of women and men in a given policy field, and to define priorities for action. The tool could also be adapted for conducting analyses based on other factors such as race or age. The '3Rs' stands for Representation, Resources and Realia. Representation and Resources are quantitative variables, whereas the third variable, Realia is qualitative.

The idea behind the method is that a systematic review of men's and women's 'representation' in different places and positions within the council department's or public board's field of operations, and of the distribution and utilisation of resources, should trigger discussions about why the 'products' of local councils - that is, goods, services and jobs - are as they are, and raise questions about who gets what, and under what conditions?

According to project co-ordinators based at SALA, the pilot schemes that have been carried out by the councils/boards have shown that the method has been successful in raising awareness. The surveys and analyses that have been carried out have led to the rethinking of gender equality in the councils' or boards' spheres of operation, including the writing of specific gender equality objectives into their normal guidelines. Some have also decided to include gender equality in their budgets, to collect new and continuous information in the form of statistics and different types of customer surveys, and to actively monitor the impact of the different measures. The 3Rs method has been adopted, and some municipalities have decided to spread the method to other committees and administrative authorities.

The method, in brief, consists of the following elements:

Representation

Representation refers to the ability of women and men to get their voices heard within a given policy area. The council departments or boards must map the levels of representation of men and women in decision making positions, starting with the composition of the committee/board itself and the administration. The process then moves to 'mapping' contacts involved in the policy process and producing sex disaggregated 'contact lists'. These mapping exercises are all designed to raise awareness of gender imbalances of representation. A study by SALA using this method showed that women politicians were more likely to have contact with other women than with men, whereas the majority of the contacts of men politicians are men. This research therefore demonstrates the importance of having a high percentage of women in local government as this is one way of improving women's chances of having their 'say'.

Resources

The Resources analysed by the R test are money, time and space. These measurements are used to supplement the information gathered in the Representation phase of the analysis. The analyst looks at the resource allocation within organisations (time women /men speak at meetings, pay comparisons) and in service provision. One example of this is that certain committees/boards have constructed 'salary trees' showing the total number of women and men and their position on the salary scale. One committee studied the gender distribution among artists exhibiting at the local art gallery, and the information was completed under the resources section of the test by recording how much money was paid to the female and male artists respectively. The lack of resources has been identified as a barrier to involvement in decision-making processes.

Realia

Realia is qualitative, this stage is concerned with exposing and challenging the norms and values embedded within local authority structures and practices which underpin decision-making and which may reinforce gender inequality. These are complex - sometimes hidden - issues which can be difficult to grasp. Various methods are used to make these issues transparent. On the face of it, some aspects of Realia may seem trivial, but they add up to a larger picture of services being designed to a male norm which exclude the realities of women's lives and needs.

Some committees/boards have looked at the Realia by studying, together with the staff, such things as morning assembly at day-nurseries, and classroom dynamics. Others have visited different establishments for which they are responsible. The politicians have formed multi-party groups and gone to places like sports centres and noted things such as lighting, hair-dryers and sauna space. They have also interviewed employees about the amount of attention they give to girls and boys, women and men. The amount of time and attention given to boys and girls by the staff of sports centres has been used to expose assumptions that boys are better at / more interested in sport than girls are.

One Swedish town planning committee that took part in piloting the 3Rs method has used 'mental maps' to help build up an understanding of how different categories of citizens felt about a particular urban renewal area. This involved groups and individuals who could not be reached through the types of groups and organisations that the committee routinely consulted.

According to project organisers:

Politicians who have been involved in different studies have expressed their satisfaction. Some have said that they have received greater insight into their field of operations and that they understand better what gender equality really means and how they can specifically work for it. ⁶¹

GENDER DISAGGREGATED STATISTICS AND INDICATORS

We turn now to look at some specific tools. Gender disaggregated statistics and other gender-sensitive data underpins mainstreaming. They have crucial uses in awareness-raising, informing policy making, monitoring and evaluating the impact of policies and programmes, and assessing trends over time. The production of good, transparent gender data was a priority action area agreed by governments as part of the Beijing Platform of Action which emerged from the UN Fourth World Conference on Women in 1995.

A small number of governments, notably some of the Nordic countries and Canada, produce comprehensive official gender statistics as a matter of course. In the case of Sweden they are published annually to allow for the monitoring of progress across all government policy areas. Many other governments are committed to improving the collection, collation and dissemination of gender statistics although progress is uneven. In some instances, women's NGOs have filled the information gap, for example between 1993 and 2000, the Scottish women's organisation Engender produced an annual Gender Audit. Drawing from varied sources, ranging from government statistics and academic research to annual reports from voluntary organisations, the publication pulled together available information and gender statistics on a range of areas and offered commentary on the position of women in Scottish society. ⁶²

In Sweden, gender disaggregated statistics are a central plank of its mainstreaming strategy. Statistics Sweden, the official statistics service first set up a special unit for the production and promotion of gender statistics in 1983. Since 1992 statistics collected on individuals have been disaggregated by sex in an annual publication: Men and Women in Sweden, which summarises information in areas such as: health, crimes, work, income, pay, power, education, smoking, the elderly and so on, together with international comparisons . This publication is seen as invaluable in mainstreaming training and policy analysis; it is also seen as an important resource for civil society. ⁶³

In 1996 - as part of the overall gender mainstreaming strategy - work was carried out in the Ministry of Labour to present labour market gender statistics. For example, unemployment statistics had routinely been presented in the following categories: Total; of which women; of which younger people; of which older people; of which handicapped; of which immigrants. This presented all but middle-aged, non-handicapped, non-immigrant men as "special" or "problem" groups. The statistics have since been rearranged into three columns: Women, Men and Total. These are each cross-tabulated by other equality dimensions such as race/ethnicity, age, disability and so on. 'Young, immigrants and handicapped are women and men. And now we find that civil servants and politicians pose new questions, because of the visibility of women and men.' ⁶⁴

Statistics Canada's role in compiling and disseminating gender-disaggregated data is seen as a crucial component in the development of relevant policy research for gender-based analysis. In partnership with the Policy Research Fund at Status of Women Canada (see Section 6), Statistics Canada published Finding Data on Women: A Guide to Major Sources at Statistics Canada, a document which provides information on gender-disaggregated data sources that can be used in carrying out gender-based analysis.

Indicators

Indicators 'summarise a large amount of information in a single figure, in such a way as to give an indication of change over time, and in comparison to a norm.' ⁶⁵ Successful mainstreaming therefore requires the development and use of gender-sensitive and/or equality indicators. The following section draws upon the discussions of a symposium on gender equality indicators involving Health Canada, Human Resources Development Canada, Statistics Canada, Status of Women Canada and information drawn from the Commonwealth Secretariat GMS reference manuals. ⁶⁶

Three primary potential applications of indicators within mainstreaming work are:

• As an input: using the indicators to inform analysis

Gender disaggregated data can give a comparison of the situation of women and men in the policy area which will then inform gender analysis. For example, analysis of transport policy in Sweden revealed that while women were the main users of public transport, provision followed male patterns of travel, therefore, 'public transport answered to the needs of those who do not use it'.⁶⁷

• As a results measure: to measure the success of analysis

For example, an indicator measuring the ratio of women to men in all levels of an organisation can be used to measure the success of a policy designed to overcome segregation in the workforce. On a national level, the GEM indicators (see below) can be used to measure and track the position of women relative to men in terms of ability to participate actively in economic and political life. Thus GEM can be used to measure the success of a range of policies aimed at improving women's participation.

• To raise awareness: as a tool to sensitise policy-makers and the public to problems and the need for specific gender/equal opportunities analysis

Providing and publicising disaggregated data can turn a private issue into a public or social issue; an example of this would be the dissemination of statistics on domestic violence or the gender pay gap. Collections of gender disaggregated statistics across a whole range of areas builds up an overall picture of patterns and trends in gender inequality.

However, there are particular issues which relate to the use (and misuse) of indicators. Indicators are not 'facts' which exist 'out there' for the policy analyst/maker to use. Rather, they are created and 'validate particular world views and prioritise selected areas of knowledge.' ⁶⁸The lack of statistical and other data on the position of women in various policy fields has been noted at several points in this report and in the country examples. The majority of existing indicators are designed around gendered assumptions, which tend to lead to the 'invisibility' of women. These assumptions can be summarised as follows:

• Many traditional indicators assume that women and men experience the world in the same way.
• Alternatively, women figure in traditional indicators primarily as mothers or caregivers, further relations among the family are often obscured by a focus on the head of the household, often defined as the male breadwinner.
• The main indicators used at an international and national level are GDP (Gross Domestic Product), ISH (Index of Social Health), and the HDI (Human Development Index). All of these are 'gender invisible'.

The need for information on the position of women has led to the development by the United Nations of new indicators, notably the Gender Related Development Index (GDI) and the Gender Empowerment Measure (GEM). The former gives gender disaggregated data and adjustments of the Human Development Index for women and men. The HDI compares 175 countries on three basic measures: longevity, educational attainment and life expectancy. The GEM measures women's participation in the economic, political and professional fields. It aims to examine whether women and men are equally able to participate actively in economic and political life.

These national level indicators can be used to give comparisons between countries on different aspects of gender equality. It is important to note that very different results will be obtained depending on the specific indicators used. For example, The United Nations has ranked Canada first in the world in terms of human development, but this drops to second place when the status of women is factored in. ⁶⁹ Canada ranks 47th out of 55 countries the UN compares in terms of women's wages as a percentage of men's, behind all industrialised countries studied and behind many developing nations. ⁷⁰

The recognition that indicators are not value free raises questions which must be addressed in the development and use of equality indicators. There are three main ways in which gender has been looked at in the construction of equality indicators:

• By measuring the individual attainment of women;
• By disaggregating data by sex and other categories (age, ethnicity, disability);
• By comparing the status of women relative to men.

All of these approaches, as well as the GDI and GEM indicators, tend to take men as the standard against which women are measured. This is a serious point for a mainstreaming strategy which aims at being transformative, rather than just 'adding women in'.

A number of questions or points need to be considered when designing and using indicators in mainstreaming work. They are summarised below:

• Consideration that equality indicators cannot be used as stand alone tools.

They need to be interpreted within an explicit framework and used in conjunction with other tools and resources. They will not constitute policy analysis on their own.

• Is it possible, or indeed desirable, to attempt to reduce complex social and economic phenomena to a single meaningful statement or stylised fact?
• To what degree is there a 'fit' between indicators and social reality? Whose reality do they correspond to?
• How can we develop indicators which meaningfully measure qualitative variables and experiences ?
• Consideration of the need to develop expertise in using and interpreting indicators
• Indicators need to be supplemented by additional information from other sources.
• Indicators need to be combined with clear targets and plans to improve results.

However, when used appropriately, these measures can provide brief and powerful descriptions of women's current situation and benchmarks against which success of particular initiatives and the need for further work can be assessed. Other examples include the New South Wales scorecard (see Section 3), performance indicators (used in the past in the Australian Women's Employment Strategy), and the Finnish Equality Barometer.

TOOLS AND TECHNIQUES

Mainstreaming involves the reorganisation of policy processes so that gender and/or equalities issues are dealt with in 'mainstream' policy making by mainstream policy actors. This necessitates the development or adaptation of a range of tools and techniques. The Council of Europe classifies tools into three broad categories: analytical tools and techniques; educational tools and techniques; and tools and techniques which facilitate consultation and participation. Most tools and techniques have been developed within the context of gender equality, although many, but not all, are amenable to use in generic equalities work.

Analytical tools are designed to 'expose' the problem.

Tools include: disaggregated statistics; surveys and forecasts; research; check lists; guidelines and terms of reference; gender impact assessment and differential impact analysis models; indicators; and monitoring tools.

Educational tools are designed to raise awareness, to transfer knowledge, and to support training.

Tools include: awareness-raising and training courses; follow-up action; mobile or flying experts; manuals and handbooks; booklets and leaflets; and educational material for use in schools.

Consultative and participatory tools are designed to improve the quality of policy-making and deepen democracy.

For example, working or steering groups and think tanks; directories; databases and organisational charts; participation of both sexes - and all social groups - in decision-making; conferences and seminars; and hearings and consultative forums.

Tools can be differentiated by the purpose they are intended to serve. In reality, many tools could be placed to all three categories, for example, disaggregated statistics are often useful as educational tools or in forming the basis of consultation, as well as for their analytical uses.

The following figure summarises some of the main mainstreaming tools and techniques and their uses.

Figure 2.3 Summary of mainstreaming tools and techniques

Source: Abridged and adapted from Council of Europe, 1998 and USGS, 1998. Updated from materials gathered in current research review.

« Previous | Contents | Next »