Including Gender: An APS Guide to Gender Analysis and Gender Impact Assessment

2.2 How to conduct gender analysis

On this page

  1. Questions to ask in gender analysis
  2. Gender equality in universal systems
  3. The need for gender-disaggregated data
  4. Step-by-step process for using evidence and data to inform gender analysis
  5. Embedding safety by design in Australian Government systems and services
  6. Safety by design principles
  7. Targeted and proportional analysis
  8. Reusing gender analysis
  9. Using artificial intelligence in gender analysis
  10. References

Gender analysis involves interrogating data, considering research, and exploring stakeholder or community views to identify and assess the different ways a proposal may impact people based on gender and intersectional considerations.

Gender analysis is a standard part of effective policy design. Gender analysis is most impactful when incorporated from the beginning of proposal development, as it helps inform a policy from the outset. When undertaken later in the process, gender analysis will still help surface impacts and possible unintended consequences that can be addressed in design and implementation. Overall, the earlier analysis is undertaken, the easier – and more effective – it will be to design policy that has gender equality benefits.

Back to top

Questions to ask in gender analysis

1. Are there existing gaps or inequalities between women and men?

  • Do women, men and gender diverse people have different rates of participation, access or outcomes in the policy area?
  • Are there differences for people based on age, race, disability, income, sexual orientation, gender identity or geographic location?
  • What are the drivers of the gaps in participation, access and outcomes?
    • For example, a key barrier to women’s workforce participation is access to workplace flexibility to accommodate caring and other responsibilities.

2. Will women or men be more affected by the proposal?

  • Will women, men and gender diverse people participate in, access, and benefit from the proposal’s services, resources, opportunities and decision-making in the same way?
    • Consider both the people who will use it and the workers delivering the proposal.
  • Could the proposal inadvertently make existing gender gaps bigger by continuing the status quo or reinforcing norms and stereotypes?
    • For example, funding support for the male-dominated construction industry could contribute to a widening of the gender pay gap. Or programs targeted at families may assume women will take on unpaid caregiving and men take on paid work.

3. Will particular people or groups be affected differently or excluded from the proposal?

  • Will considerations such as age, race, disability, income, sexual orientation, gender identity or geographic location affect how people participate in, access, or benefit from the proposal?
  • Are there barriers to participation or access for particular people or groups? Can the proposal remove or reduce those barriers?
    • For example, factors like money, safety, racism, discrimination, or caring duties.
  • Have you consulted with the people or groups that will be most affected, or affected differently?

4. Does the proposal support the Australian Government’s commitment to gender equality?

  • Does the proposal align with Working for Women? If so, how?
    • Refer to Working for Women’s priority areas, outcomes, indicators and actions that drive change.
  • Does the proposal benefit or pose risks to gender equality?
    • For example, could the proposal increase the risk of gender-based violence?
  • If the risks to gender equality cannot be addressed in the proposal, will action be considered in the future?
  • Has gender equality been considered at all stages of the proposal (development, implementation, monitoring and evaluation)?
    • How will delivery and evaluation account for and monitor differences in how women, men and gender diverse people participate in, access and benefit from the proposal? Will disaggregated data be collected and evaluated?

Tip!

To determine if a proposal benefits or poses risks to gender equality, ask whether it:

  • challenges or reinforces gender stereotypes and traditional roles
  • contributes to ending, or inadvertently creates greater risk of, gender-based violence
  • supports women and men to better balance unpaid work, or reinforces the gendered division of unpaid labour
  • helps close the gender pay gap, or risks making it wider
  • encourages greater workforce gender balance, or contributes to workforce gender segregation
  • helps close the gap between women and men’s retirement incomes, or risks making it wider
  • makes it easier for women and men to access healthcare and achieve equal outcomes, or risks making it harder
  • helps close gender gaps in representation, leadership and decision making roles, or risks making them wider.

Remember

Gender analysis is targeted and proportional.

Use data and evidence to guide your analysis.

Back to top

Gender equality in universal systems

Systems, programs or services may be designed to be equally accessible to everyone, that is, to be universal. However, that does not mean that all people will benefit from them in the same way. Equal access does not always mean equitable access or outcomes. Some people or groups may have different experiences across the systems, programs and services that governments are responsible for (such as health, education, payments and taxation) because of gendered barriers or impacts.

For example, although public transport is available for everyone, women are more likely than men to use public transport to travel to work.1 In the health sector, services like diagnostic imaging and blood tests are available to everyone, but may be used more frequently by women for example, for antenatal care. While men have greater vulnerability to various health disorders, such as heart disease, they are less likely to seek health care.2 These differences in usage need to be taken into account when assessing whether a proposal is likely to have a differentiated or disproportionate gendered impact. Analysis should include available data and evidence, including qualitative evidence from stakeholders and users.

A useful approach is to consider which people and groups will be affected by a proposal and how they may be impacted differently, rather than just looking at the system, program or service as a whole.

If the impact is not known or data is not available, at a minimum a proposal should outline how gender disaggregated data can be collected to understand gendered impacts over time.

Back to top

The need for gender-disaggregated data

Finding high-quality gender disaggregated data and evidence is a good starting point for gender analysis. It strengthens analysis and provides evidence for the impacts of policies. Basing policy in evidence is fundamental to good policy development. Appendix B contains links to a range of useful data sources.

Where gender disaggregated data is not available, policy makers can infer the likely gendered impact using other data. If there is limited data available, policy makers can use qualitative evidence such as research, expert analysis, stakeholder feedback or consultation to ensure the experiences of all women are considered.

Back to top

Step-by-step process for using evidence and data to inform gender analysis

Note the steps outlined are an iterative process.

Step 1: Define the policy problem and data needed

Articulate the policy problem the proposal intends to address.

Identify credible data, particularly data disaggregated by gender and intersectional factors, available on the policy problem. Data may be from government and non-government sources.

Identify any data gaps. Gaps may be due to lack of gender/sex disaggregated data, or unavailable and incomplete data, including lack of timeliness, accuracy, quality or granularity of the data.

  • Acknowledge any limitations associated with the data being referenced. Where quality disaggregated data does not exist, consider how this data can be collected in the future.

What if there are limited data sources?

Lack of data does not mean the lack of a policy problem, especially where stakeholders or subject matter experts have highlighted one. While quantitative data is important to quantify the impacts of policy, policy makers should look for other ways to understand the full extent of the issue.

It is critical for government to hear the views of experts and people’s lived experience to deliver practical changes to improve gender equality. Qualitative evidence such as feedback or other observations from stakeholders, subject matter experts or service providers should be considered alongside quantitative data. For example, a service-delivery program that does not collect gender disaggregated data can rely on feedback from service recipients and providers to show the experiences of the people the program targets. This feedback could highlight gendered impacts where people have different experiences with the services. Intersectionality can be considered through direct feedback from women with diverse lived experiences. There may also be international evidence or comparable policies to draw from. Policy makers should consider how qualitative and quantitative data can be collected as part of policy implementation and evaluation to expand the evidence base in future.

Step 2: Access the data

Data can be accessed through a variety of means, including through open data, customisable data and non-published data. Policy makers can contact relevant APS data teams to identify and access relevant non-published data.

  • Ensure the data is well organised, accessible to review, and comes from reputable sources. Include the source of data and format the reference to align with the Australian Government Style Manual.
  • Include a mix of qualitative and quantitative data where possible. See Appendix B for examples.

Step 3: Analyse the data

Use data disaggregated by gender, sex and other intersectional considerations. Observe patterns or relationships in the data and provide logical and plausible explanations for observations.

Consider any potential biases or limitations associated with the way the data is analysed, collected or presented.

If needed, seek feedback from peers or experts in the field to review and validate draft analysis. Refer to the ABS Standard for Sex and Gender for further guidance.

Step 4: Assess the implications for gender equality

Consider what the data and evidence shows about the current state of gender equality in the policy area. Consider if the proposal will directly or indirectly have differentiated or disproportionate impacts on people based on gender and other intersectional considerations. Refer to Working for Women and consider how the proposal may align with the strategy’s priority areas, outcomes, indicators or actions that drive change.

Back to top

Embedding safety by design in Australian Government systems and services

As part of conducting a detailed intersectional gender analysis, and considering the design of gender equality actions, policy makers should assess the risk that the proposal or flow-on impacts may cause harm to users, including inadvertently or through deliberate weaponisation. Policy makers should develop proposals with safety front of mind, and design actions to mitigate the risk that government systems and policies may be used to cause harm.

Safety by design is a proactive and preventative harms mitigation approach that can be applied during the design or improvement of systems. It helps identify, prevent and monitor possible harms to users. ‘Harm’ could arise as a result of someone directly misusing the system or unintentionally due to the design of the system.

Back to top

Safety by design principles

The safety by design approach developed by Australia’s eSafety Commissioner aims to proactively manage risks of harm to users of digital and technology services. Since the release of principles and tools by the e-Safety Commission, safety by design is increasingly being adopted in a range of settings to proactively prevent harm.

As part of the Audit of Commonwealth Systems announced by the Prime Minister following the 6 September 2024 National Cabinet meeting, the government is taking steps to embed safety by design to prevent the misuse of government systems to perpetrate family and domestic violence.

The following principles underpin a safety by design approach to inform the creation or improvement of systems:

  1. government role in ensuring systems work in the best interests of users and taking reasonable action to prevent harm
  2. user empowerment and autonomy
  3. transparency and accountability.

Recognising that reform and design can be time and resource intensive, the application of safety by design principles should centre on taking realistic, actionable and commensurate measures when designing new systems or developing new policies. Safety considerations need to be balanced with other design objectives including usability, accessibility, privacy and effectiveness.

For further guidance on embedding safety by design in a proposal, Australian Government agencies can access additional resources on the Gender Analysis Community of Practice on GovTeams or contact OFW-SystemsAbuse@pmc.gov.au.

Back to top

Targeted and proportional analysis

Gender analysis should be targeted and proportional to the scope, value and impact of the proposal. If the initial exploration of data and evidence does not find implications for gender equality, or differentiated or disproportionate gendered impacts, and this can be demonstrated, then it may be appropriate to end the analysis at this point. If the proposal meets the criteria for Gender Impact Assessment, it may be appropriate to complete an abbreviated assessment (see the Template for direction). This will direct effort to proposals with the greatest potential to improve gender equality and avoid placing undue burden on policy makers.

Back to top

Reusing gender analysis

Over policy and budget cycles, Cabinet Submissions and NPPs will come forward in the same policy area with similar considerations for gender equality. In these instances, it may be appropriate to repurpose previous gender analysis. For example, findings that relate to the health outcomes of women living in rural areas may be relevant to a number of different proposals. It is important to update and tailor the gender analysis to the specific proposal coming forward and give fresh consideration to the design of gender equality actions, mitigations and the overall impact of the proposal on gender equality.

In instances where the same proposal is reconsidered, the previous Gender Impact Assessment can be reattached. The assessment must be updated based on any variations to the proposal, including changes to context and recommendations.

Back to top

Using artificial intelligence in gender analysis

Policy makers must follow the Protective Security Policy Framework requirements when handling Protected information and Cabinet documents, including when using generative artificial intelligence (AI) tools.

The public expects government to be an exemplar of safe and responsible AI adoption and use of AI technologies. Policy makers have an elevated level of responsibility for the use of AI and are held to a high standard of ethical behaviour, in line with the APS Code of Conduct and existing frameworks and legislation. For further guidance, refer to the Digital Transformation Agency Policy for the responsible use of AI in government (PDF 579 KB) and your departmental policy on approach to AI.

It is important that policy makers develop gender expertise and literacy to ensure outputs produced by generative AI are accurate and do not reproduce gender bias to inform evidence‑based policy advice. In line with guidance provided by the Digital Transformation Agency, policy makers should closely review outputs produced by generative AI.

Back to top
  1. ABS, Census of Population and Housing, 2021↩
  2. The health of Australia’s males, Summary - Australian Institute of Health and Welfare, 2023↩
Back to top