Sex and gender identity statistical standards: Findings from public consultation July–August 2020

Sex and gender identity statistical standards: Findings from public consultation July–August 2020 summarises the findings from the analysis of submissions, and explains the processes used to collect and analyse feedback. It also provides information on the next steps we will take in updating the statistical standards for sex and gender identity.

During the consultation period for the sex and gender identity statistical standards, we heard from over 1,400 individuals and organisations within Aotearoa New Zealand and overseas – demonstrating just how important this work is.

I would like to thank all who engaged in the consultation for their time and effort in making a submission. Your input in this process has been invaluable, and as an organisation we have learned a great deal.

This document summarises the views shared with us by submitters during the consultation period. It reflects the diversity of views and experiences in this area.

We know that finding an approach that works for everyone is a challenge, however we are pleased with the high level of support the proposed changes received. This, along with the constructive feedback on how we could make improvements, has put Stats NZ in a good position to update the final standards. They will be fit for purpose to better facilitate and guide collection on these topics across the government data system.

Our role at Stats NZ is to reflect our diverse society in data. We are committed to updating our statistical standards so they are inclusive and provide best practice guidance to meet the information needs of Aotearoa New Zealand.

Rachael Milicich (pronouns: she/her/hers)
Deputy Chief Executive | Deputy Government Statistician

Read this report online or download the PDF.

Executive summary
Next steps
Purpose
Methodology
Findings
Glossary
Appendix 1: Sex and gender identity statistical standards: Submission form
Appendix 2: Themes and sub-themes from thematic analysis
Appendix 3: Decision diagram 

Executive summary

We are planning updates to the statistical standards for sex and gender to ensure consistent collection of information on these topics.

During the consultation period (2 July to 13 August 2020), we received 1,424 submissions.

Most submissions were from members of the public. Group/organisation submissions were received from the health, education and social sectors in particular. Most submitters agreed with the proposed changes to the standards.

Submissions raised issues in relation to the following high-level themes:

Visibility of minority populations in data
Submissions highlighted the need for a more inclusive approach to data collection for trans, intersex, Māori and Pacific peoples.

Data quality
Submitters identified many factors relevant to data quality, discussing the definitions, question design, response options, classification and coding of responses. There was also discussion on the wider view of sex and gender data in the context of administrative and survey collections.

Why sex and gender data are important
Submitters talked about the range of uses for sex and/or gender data, the contexts in which this data may be used, and the importance of this data in the areas of research, policy and service delivery.

Personal experiences interacting with government
We heard about how changes to our collection practices would impact people on a personal level.

Next steps

The public consultation process has provided a wealth of information to inform the review of the sex and gender identity statistical standards. Overall, there was strong support for the proposed changes from a range of stakeholder groups. We are pleased with the high level of thoughtful engagement with the consultation, and believe it reflects the importance of this work. While views differed in some areas, feedback tended to view the proposals as a positive step forward from the existing standards for sex and gender identity.

The next steps are to develop and apply recommendations for updates to the standards. The need for change to a more inclusive approach was emphasised to us consistently in submissions, and we intend to give effect to our proposed changes in line with this feedback.

Submissions have also highlighted some areas that require further work on our part including:

• The need for further engagement with Māori was highlighted to us by many submitters. We will work with Māori stakeholders to determine the best way to progress this aspect of the work.
• Many submitters highlighted the importance of user guidance and implementation support. We will review how we offer this support and think about further ways we can aid the bedding in of the updated standards across the data system.

Creating recommendations following public consultation is a balancing act. While we seek some consensus, our findings show a wide range of views. In general, we are guided in our next steps by the following points:

• Making sure changes address issues with the existing standards outlined in the consultation paper is a key objective of the review. Any amendments to the proposals will be considered in this respect. Key issues to note here are the standards’ ability to meet data needs, while supporting inclusive data collection practices.
• Consistency with international practice is a key consideration in any changes made to the statistical standards.
• Submissions from other agencies who are likely to adopt these standards will influence the recommendations.
• Views of the wider New Zealand public are an important consideration.

We plan to release the new standards in early 2021. The Government Statistician signs off on the statistical standards, independent of the Minister for Statistics. Stats NZ will work on the development of an implementation plan following approval of the final standards.

Purpose

Sex and gender identity statistical standards: Findings from public consultation July–August 2020 summarises the findings from the analysis of submissions, and explains the processes used to collect and analyse feedback. It also provides information on the next steps we will take in updating the statistical standards for sex and gender identity.

We would like to acknowledge and thank the organisations and individuals around Aotearoa New Zealand and overseas for their time and effort in making submissions.

Background on the consultation

Stats NZ is reviewing the guidance we provide for collection of sex and gender data. Updated standards will provide requirements and guidelines for how to gather, organise and report information on sex and gender. Reviewing the statistical standards for sex and gender outlines the need for the review and includes the terms of reference for the advisory group.

Public consultation is an integral part of the review process. We needed to understand people’s thoughts and perceptions on proposed changes to the statistical standards for sex and gender identity. Public discussion can be contentious and polarising, and this was reflected in the feedback we received.

A total of 1,424 completed submissions were received during the six-week consultation period. This is made up of 1,336 submissions from New Zealand and 88 from overseas. We received 1,383 submissions via the online form and 41 via email.

In general, most submissions agreed with the proposed changes as set out in the consultation paper. We received many suggestions for where improvements could be made, as well as criticism and disagreement with the approach.

See Sex and gender identity statistical standards: Consultation for the public consultation documentation.

Methodology

This section outlines the process used to collect public submissions, and the methodology to analyse public submissions.

Consultation process

We asked the public for feedback on changes outlined in the consultation paper and invited submissions. Public consultation began on 2 July 2020 and was open for six weeks. Invitations to the consultation were emailed to government agencies, as well as groups and individuals who had previously engaged with us on these topics.

The consultation was also available to the public via the Stats NZ website and social media channels. Submitters were able to complete the survey online (see survey form in Appendix 1), email a PDF version, or email us a written submission. Late submissions were accepted on request for two weeks following the close of the online form on 13 August 2020.

Submissions received during the consultation process were then qualitatively analysed and categorised into themes and sub-themes.

How we analysed submissions

The first part of the analysis involved manual coding, where three analysts coded all submission responses and identified common themes. Each analyst worked independently, and the outcomes were re-analysed to identify the final list of major themes and corresponding sub-themes. This thematic analysis allowed us to categorise the key ideas discussed in submissions. Appendix 2 lists the themes and sub-themes identified through this analysis.

The second part of the analysis involved machine coding. Submissions were imported into an inhouse qualitative analysis tool which used R, a programming language and software package for data analysis. This enabled natural language processing (NLP) to categorise submissions into common categories. We used results to check the manual coding process was accurate.

Findings

In general, submitters agreed with the changes proposed in the consultation paper. In this section, we show agreement ratings for each question, and describe themes identified in written feedback. Note that some submissions emailed directly to us did not include agreement/disagreement ratings with the proposed topics. We could not include these submissions in agreement/disagreement tables.

Quotes are used to give context. Organisations may be identifiable through quotes from submissions. However, we do not name individuals when using quotes through which they could be identified. The use of a quote in no way implies or represents the opinion of Stats NZ.

Some submissions were on a topic related to, but not directly about, the statistical standard for sex and the statistical standard for gender identity. These were considered out of scope of the current work but noted for future analysis.

When analysing submissions, it became clear that some related directly to the decision diagram (see Appendix 3) in the consultation paper. We have discussed these comments in the sections where this feedback was given.

Gender by default principle

We asked submitters whether they agreed or disagreed with a principle of ‘gender by default’ for data collection. This means that in cases where data collections require either sex or gender information, in most cases the recommended approach would be for collection of gender. Just under 80% of submitters either strongly agreed or agreed with this approach.

Table 1: To what extent do you agree or disagree with the gender by default principle in the proposed standard?

Submissions revolved around multiple themes, namely:

• Agreement with the principle as:
• a more inclusive approach to data collection
• an opportunity to improve data quality 
• having potential to improve research and policy outcomes. 
• The importance of both variables for specific purposes
• Disagreement with the principle, including: 
• concerns that data quality would worsen 
• that gender data was not suitable for all purposes 
• various legal and ethical concerns. 
• Issues affecting how the principle would work in practice (neither agreement nor disagreement). 

Agreement with the principle of gender by default

Agreement aligned with the following themes:

Inclusive data collection

Submitters recognised the need for data collection practices to be inclusive of people and how they identify in respect of their gender. This would improve the representation of minority groups in data, have a positive impact in terms of mental health and wellbeing, go some way toward normalising diverse genders, and removing the stigma people in these groups face. The ability for someone to self-identify when giving information about themselves was a recurring theme. There was recognition that collection of gender rather than sex was a mechanism to achieve this.

I believe these changes are going to make a huge impact on how transgender and gender diverse people are seen as a whole. This will hopefully impact the public mindset once it’s been shown how many of us there really are out there, just living our lives. Also the data collected will surely bring about policy changes and so importantly access to the health services we are in desperate need of. Thank you. (Individual)

Some submitters described to us their personal experiences of interacting with agencies and completing forms. They discussed the uncertainty and harm of these experiences.

As a trans nonbinary person, I found filling out the last census distressing, because the only options offered to me were male and female, and the on-line version didn't even allow me to just not answer the question, as I would have on a paper form (as I commented to friends later, it felt like I was being forced by the government to tell a lie). For that reason, I'm personally very encouraged to see that Stats NZ is finally recognising that genders outside the binary exist and need to be counted. (Individual)

Inclusive data collection through a principle of gender by default was seen as keeping with a human rights-based approach to data collection.

If the need to collect data exists, the Commission supports the proposal to collect gender by default, rather than sex. This practice is inclusive of indigenous, culturally specific, and non-binary genders. A rich diversity of gender identities and expressions exist in Aotearoa and the Pacific, some of which transcend Western/Pākehā binary concepts. The Expert Mechanism on the Rights of Indigenous Peoples (EMRIP) has emphasised that, “Indigenous peoples should be permitted to identify as distinct groups within States and States should take positive measures to ensure the collection of disaggregated data on indigenous peoples. (Government organisation)

An opportunity to improve data quality

Some submitters supported the principle of gender by default as an opportunity to improve data quality, through clarification of both gender and sex concepts.

These are useful and practical changes that we would actively implement in the health and disability system as part of improving the patient and consumer experience. Confusion over terms and inconsistency in recording sex and gender information among health providers has been an ongoing problem for us, which the proposed changes promise to fix. (Government organisation)

Some submitters suggested accuracy would be improved through the collection of gender data. This came back to the ability to self-identify with a diverse range of genders – a constraint of relying solely on sex data as prescriptive of a person’s identity.

As someone who works with data, including Stats NZ data, it's very important to me that data which is collected is as accurate as possible, and represents the surveyed population as clearly as it can. This policy ensures that the data collected by Stats NZ means what it purports to mean and is as unambiguous as possible. Additionally, it allows for more nuanced understanding of the gender diversity in New Zealand's population. (Individual)

We believe that the proposed approach will improve data quality. Under the previous standard, there was no clarity as to whether respondents were answering about their gender or sex, limiting the reliability of the data. The new proposed question clearly communicates to respondents that they are being asked for information about their gender and, as such, data users will be able to be more confident in the data. (Government organisation)

An opportunity to improve research and policy outcomes

Submitters noted a range of reasons why gender and/or sex information is collected and the importance of collecting gender information when required.

We submitted on Statistics NZ’s (Stats NZ) proposed sex and gender statistical standards because domestic violence is a gendered issue, so we collect gender information. We are invested in contributing to a data system that ensures sex and gender data meets information needs, is accurate, inclusive, and consistent with human rights best practice. (Group/organisation)

Submitters gave examples of instances where research, policy or service provision may be improved by the collection of gender data.

It is more inclusive than current standards and provides more appropriate information for almost all population based research, as well as for planning and delivering health services... As this dataset is built up over time it will enable government to track the experiences of members of the transgender, gender non-binary, and intersex community and ensure that social policies and services are enhancing health and wellbeing outcomes for all populations and reducing inequities. (Government organisation)

Some submitters identified the need to improve measurement of inequities and saw collection of gender data as an opportunity to do this.

All societies must strive to become more equitable and just for its peoples; failure to do so is a glaring failure by those societies. These methods promote equity and visibility for underserved populations. (Individual)

Consistency or standardisation was also seen as a way to improve research possibilities.

This makes perfect sense to us. We are a group who collect demographic, health and well-being information from adolescents and young adults. We ask about gender rather than sex at birth, so other data collection using this method would help us to compare data easily. (Group/organisation)

The importance of both variables for specific purposes

Some submitters discussed the importance of collection of sex data in specific cases, or the value of both variables without prioritising one over the other.

We endorse the changes to the statistical standards for both sex and gender identity as they will enhance the ability of public health researchers to collect, analyse and report accurately and inclusively on both sex and gender specific factors using Statistics New Zealand data. The roles of both the biological and sociocultural factors in disease prevention and health outcomes may therefore be better understood. Interventions that meet the needs of different population groups in New Zealand can be tailored accordingly. (Group/organisation)

I'm a healthcare researcher, and sex is relevant as the conditions I study differentially affect people depending on their chromosomal make up (primarily XX and XY). At the same time, the ways people live with the conditions I study are also affected by their social and cultural contexts, and so gender is also highly relevant. So asking about both sex and gender is important. Until now, we've recorded sex by default, and not taken the time to carefully record people's self-identified gender. The gender by default principle will improve research practices, and is a step in the right direction, in my view. (Individual)

Disagreement with principle of gender by default

Disagreement aligned with the following themes:

Concerns data quality will worsen

Some submitters thought that data quality would worsen as a result of the principle. These submissions drew a comparison between the concepts of gender and sex, with the view that sex provided a more reliable, robust measure.

It is our opinion that the proposed changes to the standards of sex and gender identity will compromise the integrity and accuracy of the data collected and used by government agencies. (Group/organisation)

The concern is that gender by default will have the opposite effect by obscuring rather than enhancing the collected information. It is well within the competency of NZ Stats to collect data on underrepresented groups without using gender as a default. (Individual)

Maintaining existing practice to ensure consistency of collection over time was highlighted by some submitters. These submitters were concerned about the break in time series and that the impact of changing data practices was not properly considered.

The comparisons between the answers on sex and gender and their analysis against other variables (such as ethnicity, age, health, education, crime etc) should yield important information. However, sex is the objective measure and will ensure continuity of the information over time in cross tabulations. (Individual)

Concerns gender is not suited to the purpose of collection

The purpose of data collection was raised by submitters. Some suggested that sex should be the default variable, as opposed to gender.

Outside of Stats NZ surveys, some submitters noted the range of medical and health related needs for sex data rather than gender data. Other submitters noted other fields where data analysis by sex was viewed as more suitable to data analysis by gender.

The reason I have put "Agree" rather than "Strongly Agree" is that I do believe that for very specific medical circumstances, collecting information about sex assigned at birth is important – however even then, sex assigned at birth can have diverse meanings (eg those with XY chromosomes with complete androgen insensitivity are assigned female at birth)" (Individual)

We would like to see a more contextual approach for sex at birth data rather than viewing this as an exception. Collecting sex at birth data for health-related purposes where necessary and for census collection, may be important for health care, particularly reproductive services including maternity and midwifery. (Group/organisation)

Legal and ethical issues

A range of submissions discussed human rights concerns associated with the principle.

Given sex is protected characteristic in the Bill of Rights Act and our obligations under CEDAW, where does the legal basis of “gender by default” arise? This is a question whose ramifications go far further than the collection of statistics. (Individual)

Submitters also detailed the confidentiality, privacy and security aspects that would be required alongside implementation of the principle.

However, I just want to stress that privacy and safety are paramount when requesting this information. I work with young trans people in Wellington. I have fielded many questions over the years about how to avoid outing yourself when applying for work, when getting a flat, when opening a bank account. I have also supported people who have faced discrimination and mental or even physical harm after being outed. So just from this experience I want to stress that asking if someone is trans - while it is an important question to ask in a lot of circumstances - is a big deal, and potentially impacts on the safety of the person you are asking. (Individual)

On the other hand, submitters also discussed international agreements, as well as local obligations in respect of gender identity and gender expression.

In 2018, the Committee on the Elimination of Discrimination Against Women (the Committee) recommended that the New Zealand government, “set up a centralized system for the collection, analysis and dissemination of comprehensive data, disaggregated by sex… [and] gender identity…” As such, gender and sex at birth data are needed for the government’s response to the Committee, in addition to domestic planning. Censuses and national population surveys are prime instruments for the collection of data. (Government organisation)

Our obligations under Te Tiriti o Waitangi were also cited in respect of the principle, and the standards in general.

We note that the proposed standard does not seem to be strongly informed by te ao Māori concepts of gender and does not refer to Māori data sovereignty principles or the Te Tiriti o Waitangi responsibilities of Crown agencies. It may be necessary for Stats NZ to work further with Māori organisations such as Te Mana Rarauranga, (Māori Data Sovereignty), Te Arawhiti (Māori Crown Relations) and Tiwhanawhana to develop sex/gender measurement strategies that work with te ao Māori concepts of gender, sex and sexuality. (Group/organisation)

Gender by default in practice

Submitters identified a range of factors that would impact data quality, which did not pertain to agreement nor disagreement with the principle. Rather, data quality would be conditional upon certain factors, like guidance for users of the standard, and on topics like age suitability of the questions and respondent behaviour.

However, data collectors should always ask the question “Is collecting this data necessary?” for questions about gender as well as for assigned sex. Gender should not be asked for without good reason. For many transgender people, and non-binary people in particular, being asked about their gender can be a microaggression: harmless in itself, but when taken in the context of a lifetime of having the validity of their gender questioned, the question can do harm. The guidelines should state that there must always be a clear reason for asking for a person’s gender, and ideally that reason should be stated on the form. (Group/organisation)

In future reviews, it would be good to consider the needs of children and young people, and to test questions with them to see whether they are age appropriate. It’s possible that different questions may be better understood by young people. (Government organisation)

Once people get used to the style of the question I think it will elicit more useful information. (Individual)

‘Gender’ concept definition

We asked submitters whether they agreed or disagreed with the concept definition we had proposed for gender. Much of the feedback suggested minor changes that could be made, while some feedback suggested major changes were necessary and that the definition was conceptually flawed. The majority (82%) of submitters either strongly agreed or agreed with the definition.

Table 2: To what extent do you agree or disagree with the proposed definition for gender?

Submissions revolved around multiple themes, namely:

• Agreement with the definition as:
• a more inclusive approach
• reflective of lived experience 
• a broadening from the existing definition of gender identity. 
• Disagreement with the definition, including: 
• with emphasis on conceptual issues 
• highlighting the need for inclusion of Māori and Pacific identities 
• questioning whether people with no gender fit into the definition. 
• Other aspects to consider (neither agreement nor disagreement).

Agreement with the proposed definition

Agreement aligned with the following themes:

A more inclusive approach

Submitters who agreed with our definition felt that it captured the meaning of gender and respected the diverse nature of the concept. These submissions noted how inclusive the proposed definition is, and how it gives people the ability to self-define their gender in accordance with the definition.

Quite simply – trans women are women, trans men are men, non binary people are valid. (Individual)

Reflective of lived experience

Some submitters spoke about how the definition applied to their personal experience.

Your definition is consistent with my own experiences as a cisgender woman who has friends with different genders. My transgender and non-binary friends have different gender identities, expressions, and their current gender doesn't reflect their recorded sex at birth. So, the definition you propose matches my lived experiences. (Individual)

A broader concept than gender identity

Some submitters supported the broadening of the concept and saw the existing concept and definition of gender identity as too narrow.

We strongly support the broadening of the ‘gender’ concept to encompass the existing concepts of ‘gender’ and ‘gender identity’. We feel that the use of the word ‘identity’ has the effect of othering transgender people. A broader ‘gender’ concept is more inclusive. (Group/organisation)

Disagreement with the proposed definition

Disagreement aligned with the following themes:

Conceptual issues

Some submitters felt that the definition conflated sex and gender, through use of the terms ‘male’ and ‘female’.

There was some criticism focused on the fluid and/or subjective nature of gender (as per our definition), and its suitability as the basis of statistical collection.

Some submitters disagreed with specific aspects of the definition – such as the inclusion of gender expression and gender identity and the conflation of these more specific terms.

Others felt that the concept itself and its definition reinforced unhelpful stereotypes about gender roles and expectations, and the nature of concepts like gender identity and gender expression.

Gender refers to a set of socially defined behaviours and expressions which have been - erroneously in my opinion - attributed to the two sexes. Despite some earlier relaxation of these stereotypes, I have observed them becoming more rigid again in recent years. These behaviours should not be used to constrain how people express their personalities or present themselves. (Individual)

We also disagree with your claim that a person’s gender can change – a person’s gender expression can certainly change, but a person cannot change their gender identity. This is why trans people transition (!) – because they cannot change their gender identity. Your current proposed definition of gender is likely to reinforce the erroneous belief of many ignorant people that transgender people choose to be transgender. (Group/organisation)

Inclusion of Māori and Pacific identities

Submitters saw a need to work with Māori and Pacific groups in the development of this definition, to ensure its suitability for a New Zealand context. Specifically, it was noted that culturally specific identities do not draw rigid lines between sex, gender and/or sexual identity.

It may be more appropriate for Aotearoa to develop identity standards and classifications that recognise there are culturally specific Māori, as well as Pasifika, identities and terms that convey a mix of gender and/or sexual diversity. (Group/organisation)

We agree with the proposed definition of gender, acknowledging that this is based on Western worldviews. We recommend that Stats NZ undertakes further consultation with Māori and Pacific experts to develop appropriate ways of acknowledging how Māori and Pacific peoples frame these concepts. (Group/organisation)

The definition should include acknowledgement of indigenous Māori and Pasifika concepts of gender which fall outside male/female/non-binary categories. It is important to recognise our Aotearoa context for educational purposes, inclusivity, and clarity. (Individual)

Accordingly, submitters reinforced the need to integrate Te Tiriti o Waitangi in the development of standards in the future.

Inclusion of agender/people with no gender

As drafted, the proposed definition notes that “some people may not identify with any gender”. While this was acknowledged, some submitters questioned whether the definition, or any associated question, would account for people who are agender, or people who have no gender.

However, having 'another gender' alongside 'male' and 'female' is not sufficient; […] believes that Statistics NZ needs to mandate an option for people with no gender as a fourth option. This option should ideally read 'no gender', as this is a clear definition, and 'agender' is not universally used by people with no gender. (Group/organisation)

Other aspects to consider

A range of submissions did not focus on agreement or disagreement with the definition but noted other aspects.

Some submitters said that the definition should explicitly state that no medical or surgical procedure is required for someone’s gender to be valid.

[…] considers it important to acknowledge that gender is self-defined (does not require any need to undergo medical, legal or other steps in order to be ‘valid’). (Group/organisation)

Some submitters noted that inclusion of the term ‘non-binary’ may not be technically correct, given its meaning as someone who identifies outside of the gender spectrum.

I agree with the statement. However 'non-binary' is not usually a gender identity in itself, but a description of the person's gender identity as it lies outside the binary standard of 'male' or 'female'. Also, some people who could be described as non-binary don't use the term, for example agender people or others with a lack of any feeling of gender consider themselves to not be non-binary or binary of gender by way of not having a gender at all. (Individual)

Another gender

We proposed the use of the ‘another gender’ category for classifying responses other than male or female. This would replace the existing ‘gender diverse’ category.

A text response option was recommended, giving respondents the ability to further define a gender within this (eg Another gender, please specify: ______). Most (74%) submitters either strongly agreed or agreed with use of this category.

Table 3: To what extent do you agree or disagree with the use of 'Another gender' in the standard?

Submissions revolved around multiple themes, namely:

• Agreement with use of ‘another gender’ category, including that:
• it is a more inclusive and enduring approach 
• there is the ability to further define and/or include additional categories
• there is added clarity and consistency of data collection.
• Disagreement with use of ‘another gender’ category, including:
• concerns about ‘othering’
• concerns about the classification worsening data quality
• that the existing practice should be maintained.

Agreement with use of ‘another gender’ category

Inclusive and enduring approach

Inclusiveness was a key theme reflected in the feedback we received on the use of the term ‘another gender’ in the proposed classification.

Some submitters told us that ‘another gender’ is a neutral and broad umbrella term that would be inclusive of genders that fall outside of the binary categories of male and female.

We strongly support the proposed use of ‘another gender’ to replace the classification of ‘gender diverse’. This is plain English that elegantly conveys the multiplicitous nature of gender, without forcing respondents to select an umbrella term they may not relate to. The proposed use of ‘another gender’, as opposed to ‘gender diverse’ feels much more inclusive of the cultural diversity within LGBTTQIA+ communities in New Zealand. (Group/organisation)

Some submitters told us that the use of male and female gender categories explicitly including both transgender and cisgender people is a more inclusive and accurate approach than the existing practice. Many described this as reflective of their lived experience of gender in society.

A major limitation of the current gender standard is that the concept of “gender diverse” conflates transgender experience with non-binary identities. “Gender diverse” is defined as an umbrella term which includes all transgender people, but the suggested way of asking about gender lists “gender diverse” as a third discrete option, in opposition to male and female options. In practice, many transgender men and women select the male and female options, rather than “gender diverse” which is generally interpreted as referring to non-binary or other genders. “Gender diverse” is further complicated because it is not a term that many people positively identify with, so it can produce confusion for people whose gender sits outside of male or female, but who do not regard themselves as “gender diverse”. (Group/organisation)

Some submitters agreed with using ‘another gender,’ but questioned its suitability for those who do not identify with any gender.

‘Another gender’ is a good umbrella term to catch other diverse genders and more inclusive of cultural terms than 'gender diverse'. However, there are some people that don't identify with a gender at all who may not appreciate this term. Guidance should be clear that these people - such as those that might use terms like agender - are included. (Individual)

A common thread from submitters reflected on language people use to describe their gender as it changes over time. This can make consistent data collection difficult. Submitters supporting the change said that the term ‘another gender’ is general enough that it should endure well as language continues to evolve. Some submitters also talked about the need to socialise this terminology with the public.

I wasn't sure about this at first but reading the explanation it makes sense. Capturing the full range of gender identity is important, and I can see how this terminology allows for that without being locked into specific terms that may shift over time. (Individual)

Ability to further define

Some submitters supported the use of ‘another gender’, on the condition that they would be able to further define their gender using a text response option. Feedback reflected the importance of self-determination in accurately describing who they are.

A write-in option is preferable wherever possible, and should be the default. The language around gender is still evolving, so no set of tick boxes can ever encompass all of the ways people describe their genders. (Individual)

There were mixed views from submitters on whether ‘another gender’ would be inclusive of culturally specific identities, such as takatāpui and fa'afafine. The use of a text response option to further define these identities was supported. Some submitters felt that identities such as takatāpui should be included as response options in the question alongside another gender.

I think my main concern is that there might be situations where you can't write in what the "another gender" IS to further define that. And that, to me, is dangerous territory. That's where you'll end up erasing a LOT of our diversity, ESPECIALLY the Māori and Pasifika indigenous gender identities (and I say this as Pākeha). And then I'd worry about your obligations under te Tiriti... (Individual)

We also recommend Stats NZ consider including takatāpui and other culturally specific Māori terms as well as Pasifika terms as separate response options (in addition to “Another gender”), and amending the guidance so that these terms are always listed as separate response options if the open ended response option is not offered (eg, “Another gender, eg, takatāpui, fa’afafine, non-binary, agender, genderfluid, etc”). (Group/organisation)

Adds clarity and consistency

Submitters noted that the move to include another gender would improve the accuracy of data collected, as it provides clarity for some respondents. This was thought to allow for a more consistent or standardised approach to data collection. Feedback described how binary transgender people were often unsure whether to put themselves down as ‘gender diverse’ or as male or female.

‘Another gender’ is an improvement upon gender diverse. It avoids the term ‘other’ and makes clearer that binary trans people can self-define as male or female, in line with a human rights-based approach. This will improve response accuracy and enable more detailed analyses to identify inequalities experienced between different genders. (Individual)

‘Another gender’ is helpful wording in that it is clearly a general umbrella term that gives space for a range of non-binary and indigenous genders, and does not appear to be a positive identification term in the way that “gender diverse” does. For someone whose gender is not male or female, this term is more open-ended and inclusive than a term like “gender diverse” or “non-binary”. It is also helpful because by implication, it makes it clearer that trans women are women, and trans men are men. (Individual)

Disagreement with use of the ‘another gender’ category

Concerns about ‘othering’

Some submitters thought that ‘another’ gender was too similar to ‘other’ and potentially othering or alienating.

However, the phrase “another gender” can be read as “othering” people whose genders are outside the binary. Additionally, some Agender people, who have no gender, may not feel they are included within the “another gender” classification. (Group/organisation)

Concern classification may worsen data quality

Some submitters were concerned that classifying transgender and cisgender responses within the male and female categories could worsen the quality of data collected. Some thought this approach confuses the concepts of sex and gender further than existing practices, and that separate categories are still required in the classification. Some noted that data collected historically was related to sex, and this way of describing the classification would be a departure from that.

Some felt that along with the gender by default approach, this may reduce representation of the transgender population.

It will make transgender populations LESS rather than more visible. To identify transgender populations, we need BOTH sex at birth AND gender identity. (Individual)

Maintain existing practice

Some submitters said they would support the addition of an ‘another gender’ category to the gender classification alongside the existing ‘gender diverse’ category, rather than replacing it. The feedback reflected on the fact that the ‘gender diverse’ option is adopted already in many places, and some people may identify with this term more than ‘another gender’.

I'm open to this idea, however, I believe both 'another gender' and 'gender-diverse' should be provided as options, to ensure people can choose the term they most identify with. (Individual)

Some either felt neutral about this change or felt it was unnecessary as they didn’t see any issue with the existing term of ‘gender diverse’.

I liked the use of gender diverse as it felt inclusive. The term another gender is equally fine. Both options are good. (Individual)

Many submitters said that while they may not understand why this change is necessary, the opinion of gender minorities in New Zealand who it affects should inform its suitability.

Two-step method

We proposed a two-step method that involves asking a question about sex at birth and a question about gender. Most (74%) submitters either strongly agreed or agreed with the use of the two-step method.

Table 4: To what extent to you agree or disagree with use of the two-step method in the standard?

Submissions revolved around multiple themes, namely:

• Agreement with the proposed method, with:
• an emphasis on best practice and inclusivity
• concerns about how and when questions are asked.
• Disagreement with proposed method, with: 
• concerns around the two-step method itself.
• Other aspects to consider (neither agreement nor disagreement) include:
• operationalisation of the method and the use of the data.

Agreement with the proposed method

While most submissions agreed with the proposed two-step method, we found that some endorsed the approach fully, while others voiced concerns.

Best practice and inclusivity

Submitters who fully endorsed the two-step method identified a range of reasons why it should be followed – specifically, that both gender and sex at birth are relevant and collecting both would lead to a more complete picture. This could then be used for better decision making and funding allocations.

Other submitters saw the two-step method as providing more representative data, including information on transgender and non-binary individuals. They referred to this approach as the ‘right’ thing to do and agreed that it represented best practice.

We support the importance of collecting both sex at birth and gender data in population representative data collection such as the Census. Population data is essential for health services to measure differences in access and health outcomes based on characteristics that determine health, which include gender and sex at birth, and other characteristics such as, for example, ethnicity. Measuring the size of populations, including Transgender and Non-binary people, is essential to plan, resource and improve services that address their needs. Currently no national population data exist on Transgender and Non-binary people or on Intersex people, and this has a negative impact on, among other issues, planning, resourcing, improving and delivery of health services to these populations. (Group/organisation)

Having a two-step method will mean that there will be a better gauge of how many transgender people are in the population, allowing the government to have a more accurate measure and therefore better planning around meeting transgender people’s needs. (Individual)

Some submitters supported the two-step method because it would mean that data about sex is still collected, which they considered important for biological/health information needs. Some submitters endorsed the two-step method over and above the ‘gender by default’ principle. They emphasised that there were clear information needs for sex. Some submitters proposed that the two-step method be the default collection process.

…gender and biological sex information is both very important to have for different reasons. We need to know how many people there are of the different genders for societal reasons and also need to know how many people there are of different biological sexes for health reasons. (Individual)

Some submitters proposed that collecting both gender and sex would normalise the practice – saying that aside from addressing transgender information needs, it would promote inclusivity and self-identification. Some thought that the two-step method (particularly in terms of asking sex at birth and allowing for self-identification of gender) may provide the opportunity for people to become familiar with the concepts and accustomed to seeing two questions.

This is very important for collecting data that will benefit trans people through increased visibility and accurate statistics. This is inclusive because people can self-identify for the gender question without losing information about the numbers of trans people. This also has the benefit of normalising the concept of gender as distinct from sex at birth for anybody filling in the form. (Individual)

Concerns about how and when questions are asked

Some submitters had concerns about the way questions are asked, and which surveys might use the questions.

Others had concerns about the gender question, particularly the terminology used and the response options.

Agree, however you are continuing to conflate sex with gender by using the terms fe/male. This is going to be very confusing for the majority of people who have not been involved in discussions about sex and gender. (Individual)

Some submitters disagreed with the question design for sex at birth, particularly around using ‘recorded’ or ‘assigned’, whereas others did not like that the response options were binary and thought more options should be included.

There’s a lot of scope for this to go wrong. It would need to be clear in the wording that this refers to assignation at birth rather than implying an objective existence at birth, ie, the distinction between AMAB/AFAB (assigned male/female at birth) and ‘born a man/woman’. (Individual)

While some submitters agreed with the two-step method, they raised concerns about its application. While supportive of the ‘gender by default’ principle, they thought the decision diagram may be ignored in favour of the two-step method. That is, that surveys would automatically use the two-step method even when they should not be collecting either gender or sex, or should just be collecting gender.

I want to impress how vital it is that we do not consider sex assigned at birth to be the important category, beyond the differences between transgender and cisgender experiences. (Individual)

So long as it is only used when absolutely necessary. (Individual)

Disagreement with the proposed method

Those who disagreed with the two-step method either disagreed with the ‘gender by default’ principle as well (discussed above), or only the two-step method.

Concerns about the two-step method itself

Submitters who disagreed with the two-step method thought the proposed focus was incorrect or incomplete.

These submitters thought the two-step method contradicted the ‘gender by default’ principle. They proposed either a different two-step method (eg asking gender first or making sex at birth an optional step), or doing away with the two-step method altogether. Those who wanted to do away with the two-step method were likely to propose asking a gender question only, as they believed gender to be more relevant.

Sex at birth is not relevant for any data collection. If you want to know if someone is transgender then that should be the question. Asking for sex at birth is invasive, does not actually give the information you want and alienates transgender people. (Individual)

While we acknowledge that there are some very limited circumstances in which the two-step method may be effective as a means to collect data, we are not certain that there are any circumstances which it is entirely justified or worthwhile… (Group/organisation)

Some submitters disagreed with the two-step method because they did not think it would address information needs on transgender and intersex individuals. Instead they favoured the use of a gender question and a trans status question or an intersex variation question, depending on the data needs. They also questioned the implications of asking a sex at birth question and its impact on trans individuals.

If you want to know if someone is trans, there should be a question asking if the person is cis or trans (and separate from the question about specific gender). (Individual)

It is also liable to incorrectly flag intersex people who were incorrectly sexed at birth. With further medical information a person registered as F might later be found to be M (eg, complete androgen insensitivity) … A direct question rather than inferring from sex would be more accurate. (Individual)

Some submitters considered a transgender status question particularly important for contexts where asking sex at birth may not be appropriate, but where the information about whether someone is transgender is required. Some thought that without guidance, people would find non-standard solutions.

The consultation document suggests that a transgender status question should never be asked, and that the two-step method can only be used in very limited circumstances. This means that in situations where transgender demographic data is needed, but there is no need to know a respondent’s sex assigned at birth, researchers could respond by:

- using the two-step method in an attempt to appear inclusive and follow perceived best practice, causing unnecessary offence and distress, and limiting response rates. This may result in transgender people being underrepresented in results due to opting out, and carries the risk of poor analysis by less experienced researchers since it is more complex to use the two-step method to ascertain which respondents are transgender.

- not asking about transgender status because they see that their data collection does not fit the limited circumstances that are appropriate for the two-step method, and they see no other guidance about appropriate ways to ask about transgender identity. This would mean researchers could not identify transgender respondents, limiting the value of their research for gaining insight on transgender lives and views. 

- developing their own transgender status question, potentially with inappropriate wording or non-comparable results. This may result in offence, distress or confusion if a question is worded badly. It would also limit the comparability of data sets since transgender status would not be asked about in a standard way. 

Without providing the option of a transgender status question, the standard would have limited value for ensuring a range of good data and information is collected about transgender people in Aotearoa. (Group/organisation)

Other aspects to consider

Operationalisation of the method and the use of the data

Some submitters did not voice agreement or disagreement with the two-step method and instead focused on how the questions would be asked and how the data would be used.

Submitters raised concerns about privacy, security, and confidentiality and the potential for misuse of sensitive data. Due to the nature of the questions, some submissions focused on the potential for misuse of data. Specifically, they were concerned data could be used against individuals (eg to ‘out’ someone as trans) if the data is not kept private/secure/confidential.

…it is a very personal and delicate topic, it can be a very triggering experience for those who view their birth sex as being a point of trauma, and would need to be presented in a way that feels comfortable and safe, as well as necessary information. (Individual)

Privacy and elimination of risk to the individual (through discrimination or other harm) must be ensured when collecting this data. (Individual)

Some submitters thought the two-step method would put undue burden or stress on trans individuals. That is, with the use of the two-step method, some trans individuals would be pressured to ‘out’ themselves as being trans. Submitters thought we had not consulted widely enough with trans and other minority groups.

This is going to be upsetting, infuriating or traumatising for thousands of people, and will only prolong the stigma and discrimination currently being experienced by trans people in particular. (Individual)

I understand that you may want the data for data collection purposes, to establish how many people are in the community, and for valid reasons, but I’m not sure the way you are going about it will be supported by the trans community. (Individual)

The harm of outing: The proposal exposes people to the risk of being outed as trans (to governmental or related agencies) against their will. (Group/organisation)

Some submissions also raised the possibility that, due to the sensitivity of the questions, respondents were unlikely to be truthful, especially if unsure their data would be protected. If there is ambiguity in the definition of a concept or the phrasing of the question/response options, then people may be confused as to how they should answer. This would mean that the data is unreliable, of low quality, and therefore unusable.

Trans people may not respond to this with the info you want… (Individual)

Sometimes sex at birth is hard to define and so might be hard for people to answer honestly. (Individual)

‘Sex at birth’ concept definition

We asked submitters whether they supported the introduction of a specific concept, definition and question module based on ‘sex at birth’. This was recommended for use in the two-step method.

There was a high level of agreement for the change to the sex standard. Some submitters supported the change, but with specific conditions on how it would be used or with suggested amendments, while others did not endorse the change. 70% of submitters either strongly agreed or agreed with use of the concept.

Table 5: To what extent do you agree or disagree with use of the sex at birth concept in the standard?

Submissions revolved around multiple themes, namely:

• Agreement with the proposed sex at birth concept, with:
• improved clarity and ability to meet data needs 
• suggestions on specific terminology amendments
• further user guidance required. 
• Disagreement with the proposed concept, with:
• concerns about sensitivity, privacy and confidentiality
• the definition 
• the response options.

Agreement with the proposed concept

Improved clarity to meet information needs

In combination with the proposed gender concept, some submitters thought the shift to a concept specifically about ‘sex at birth’ would add helpful clarity, enabling data collection that more accurately reflects diversity. Related to this, use of ‘sex at birth’ was mentioned by some submitters as best practice for the two-step approach, where both gender and sex information are collected.

Some submitters thought this change would lead to less conflation of the concepts of sex and gender.

Suggestions on terminology

Preference for the wording ‘sex assigned at birth’ was expressed by some submitters. It was suggested this would further improve the specificity of the concept and question. Some also told us that this terminology was more commonly used, and more appropriate for transgender and intersex people, as it reflects the social designation of sex from birth.

Specifying sex ‘assigned at birth’ was thought by some to be less tied to assumptions about a person’s biology/physical characteristics, which may have been different to what was assigned or recorded at birth. For this reason, answering a sex question that does not specify the point in time may be difficult to answer. Related to this, submitters also queried why ‘recorded’ or ‘assigned’ were not included in the question itself, rather than an example under the question.

Some submitters preferred the wording included in the proposed definition - ‘sex recorded at birth’, saying this was useful in recognising the administrative/external nature of the process.

Further user guidance required

Some submitters suggested further guidance should be given when collecting gender and sex data in different contexts.

Many submitters said clear user guidance about where and how information on sex at birth is requested should be provided in the final standard. This related to concerns about data being collected in inappropriate contexts. The need for education and support for those implementing the standard was suggested.

Stats NZ should also make it clear that if an agency or organisation wants to ask a person about their sex assigned at birth, they must only do this when the reasons are clearly justified, such as accurately representing the size of the trans population in a population-based survey, or for understanding the number of trans people who were assigned male at birth or assigned female at birth for healthcare access reasons. Stats NZ should clearly note that when collecting information about an individual for individual purposes (not nationally representative data collection), a transgender status question is more appropriate. Clear guidance must be provided as to which is appropriate in which situation and must stress that there are very limited times when asking sex-assigned-at-birth might be useful, and that it is unnecessarily invasive to ask this in almost all circumstances. Guidance around privacy and confidentiality should also be provided. Any questions about sex assigned at birth should always be voluntary. (Individual)

Some submitters agreed with the proposed guidance for collecting sex at birth information, namely that this would only be suitable to collect in the context of the two-step approach (where a gender question is also asked). Other submitters thought collection of sex at birth data only in the two-step approach was too limited, and would not allow for data needs to be met in all relevant instances.

Some submitters noted that sex at birth data may not be informative enough in some contexts, for example medical care where more detailed information would be necessary.

Disagreement with the proposed concept

Confidentiality, privacy and sensitivity concerns

Concerns around sensitivity, confidentiality and privacy came through in feedback on this concept and question. This was both from those who supported use of the concept, and those who opposed it.

Some submitters were supportive of the use of this question and concept, but only by exception. Submitters said that the reason for collection must be clear, with enough privacy safeguards around the collection and storage of the information.

Because asking about sex assigned at birth is particularly sensitive and carries risks of harm, we suggest that whenever this question is asked, an explanation should be provided of why this is being asked, and how the information will be used and protected. (Group/organisation)

Some said that asking for sex at birth information may put transgender people at risk of having their transgender status disclosed where this is not relevant or safe. Some also felt this question and concept may be offensive or invalidating and not respectful of self-determination.

We agree that birth-assigned sex should only be asked in those limited circumstances where identifying transgender and cisgender populations is required. The guidance accompanying the standard should make it clear that asking trans people to provide this information can be distressing, traumatic and potentially dangerous, and agencies should only collect information about sex assigned at birth when the reasons are clearly justified (such as accurately representing the size of the trans population, or understanding the number of trans people who were assigned male at birth or assigned female at birth for healthcare access reasons) and when the data can be collected and stored confidentially. (Government organisation)

For some submitters, these concerns meant they did not support the collection of this data in any context.

Disagreement with response options

Some submitters thought that only having male and female response options to the question on sex at birth was not inclusive, noting particular concern for intersex people. This was also reflected in the feedback on the intersex variation question, where some submitters asked why intersex could not be built into the existing sex question response options. Many submitters simultaneously noted that intersex advocacy groups would be better placed to give an informed view.

Disagreement with definition

Some did not think the definition of sex at birth captured the meaning of the concept well. They thought the definition did not describe the physical and/or biological basis of sex. Some submitters also felt that the phrase ‘at birth’ was not necessary as they did not think that sex could change.

Intersex information needs

We asked submitters to what extent they agreed that the inclusion of a specific intersex variation question in the standard would meet information needs for the intersex population. 79% of submitters either strongly agreed or agreed with this approach.

Table 6: To what extent do you agree or disagree that this approach will meet information needs for the intersex population?

Submissions revolved around multiple themes, namely:

• Agreement with the approach, for:
• representation and visibility of intersex people 
• meeting intersex data needs.
• Disagreement with the approach, as:
• a separate question may be othering
• this is private medical information.
• Other aspects to consider (neither agreement nor disagreement) include:
• concerns about a lack of understanding of what intersex means
• further user guidance 
• diverse ways intersex people identify
• further consultation needed with intersex advocates/experts.

Agreement with the approach

Representation and visibility

Some submitters told us inclusion of an intersex variation question in the standards was an important step toward lifting the visibility and representation of intersex people in Aotearoa New Zealand. This related to representation in official statistics, as well as visibility more generally.

Submitters typically described the lack of data available about the intersex population in Aotearoa New Zealand, and the potential benefit that such information could have for decision making, funding and policy decisions affecting intersex people.

Intersex people are largely forgotten in data, its important they are represented accurately. (Individual)

There is a severe lack of intersex population data both globally and in NZ; little is known even about the size of the population. Collecting this data is vital and we support the proposed standard. (Individual)

We hope that enabling data collection about Intersex people in population data will be the first step to improved visibility and understanding of the experiences of Aotearoa New Zealand Intersex people, and that it will enable improvements in access to health care and other public services that address their needs and support their aspirations. (Group/organisation)

Meeting intersex data needs

Some submitters supported the approach, agreeing that a distinct question would be suitable as recommended practice for collecting intersex data. Feedback referenced documents such as the Darlington Statement, Yogyakarta Principles, and Malta Declaration. These acknowledge that intersex people have both a gender and sex assigned at birth.

As intersex is neither a sex nor a gender it is important that this question is held separate from these categories. (Group/organisation)

The Yogyakarta Principles plus 10 recognise that sex characteristics are distinct from other grounds such as gender identity or gender expression. Classifying intersex people as a third sex or gender does not respect diversity or the right to self-determination. Intersex is not a sex or a gender, and there is no one intersex sex; it is an umbrella term to describe over 40 distinct variations in sex characteristics. (Government organisation)

Trying to group intersexuality with sex at birth essentially erases anyone who is intersex but was assigned to be a binary at birth (sometimes due to a decision being made to raise an intersex child as a certain gender (which I don't agree with). Having a separate question about someone's sex at birth and whether they are intersex keeps all the information intact and is more inclusive of intersex people that are still labelled as a binary sex on their birth certificate. (Individual)

Disagreement with the approach

Disagreement with separate question

Some submitters thought that a separate question for collecting data about intersex people was not necessary and that it may be othering for intersex people. Some suggested including intersex as a distinct category in the sex classification. Many noted that they would defer to intersex experts on this issue.

I think this should be integrated into the sex question rather than separate. I don't think forcing an intersex person to choose between M/F and then later stipulating they're intersex is appropriate, considering forced surgeries at birth etc. However, I'm not intersex myself, and intersex people may hold a different opinion. (Individual)

Private medical information

Some submitters told us that having an intersex variation was a personal medical issue, and contexts in which this is relevant would be very limited and therefore not required in the final standards.

Do you ask questions about other medical conditions? As a health professional I have dealt with these conditions and generally the people concerned see it as a medical condition and are usually quite private about it. (Individual)

Other aspects to consider

Concerns about a lack of understanding of what intersex means

Many submitters described a lack of awareness or understanding about what it means to be intersex. There was concern that this would limit people’s ability to answer an intersex variation question in an informed way, making it difficult to get accurate data. This was often combined with emphasis on the need for further guidance and education on the topic.

I think this question is very technical in essence. I understand it might best fit for the intersex community, but if you add it to the census, I imagine there will be a sizeable number of cis-gendered individuals who erroneously select that they are intersex. If you are going to add to census, ensure enough work has been done to validate the accuracy of this question. (Individual)

We support the wording of "Were you born with a variation of sex characteristics (otherwise known as an intersex variation)?" as proposed by the review, provided that it is accompanied by a clear definition. The terminology used here is the most straightforward, but may require an explanation for some respondents, as intersex awareness and education are not strong in Aotearoa, and some respondents may be aware of an intersex variation that they have but not be aware that it is called this. (Individual)

Related to this, submitters often cited the reason they responded as ‘neutral’ was that they did not feel informed enough to have a view either way.

Further user guidance is needed

Many submitters commented on the need for further guidance to ensure people answering this question understand what is being asked. They also reflected the need for more guidance on which contexts were appropriate to ask the question.

The standard must also provide advice and guidance about the contexts where it is appropriate to ask about whether someone is intersex. The standard is a reference point for many different researchers, agencies and data analysts, and will be used in a wide range of contexts. Intersex advocates will be better placed than we are to explain in which contexts it is or is not acceptable or helpful to ask if a person is intersex. (Group/organisation)

Like sex at birth, information about intersex variations is sensitive and should not be collected without a good reason. We agree that further work will be needed to develop guidance around where it is appropriate to include an intersex variation question in any administrative data collection. Such decisions should be heavily weighted in favour of the requirements and advice of intersex communities and organisations. (Government organisation)

We also support including definitions and examples of intersex variations when asking this question; many people who are intersex often do not realise and we would hope that as well as collecting information about this population, asking this question would also raise awareness and reduce stigma in a considerate manner. (Group/organisation)

Diverse ways intersex people identify

The diverse ways intersex people identify with respect to sex and/or gender came through in the feedback on this question. Submitters said that for this reason, no single approach would work well for everyone, including the proposed approach.

Individuals will have their own way of understanding and connecting to their intersex variation and bodily experiences. It is critical to understand that there is no way to minimise such a broad diversity when describing the intersex community. (Individual)

Intersex people make up a wonderfully diverse community. They are born with sex characteristics (including genitals, gonads and chromosome patterns) that do not fit typical binary notions of male or female bodies. They represent diversity as men, women, and non-binary people; that is people who live beyond the binary of male-female. Intersex folks have varied understandings of, and relationships to, their sex and gender. Like everyone, they have a range of sexual orientations; including those who are heterosexual. This is important to point out because not everyone who is intersex feels like they are part of the rainbow. They are parents, guardians, siblings, children, elderly and whānau. Being intersex is not a gender identity for the majority of people born with variance of their sex characteristics. (Individual)

Further consultation needed with intersex advocates/experts

Feedback encouraging Stats NZ to engage in further consultation with intersex advocates and experts was consistent across submitters.

We encourage Stats NZ to undertake further targeted and resourced consultation with intersex people, organisations, and communities to determine that the wording of such a question and any associated guidance will be inclusive of people with intersex variations. (Government organisation)

As Stats NZ indicates in the discussion paper, further consultation with intersex communities is required. At face value, this option looks sensible and straightforward, subject to the possibility of being amended in response to the needs of people with lived experience of intersex variations. (Group/organisation)

Further information shared

We reserved a section for submitters to provide comments that would assist us in the review of the standard. Almost all the submissions were related to the previous sections/topics, primarily emphasising earlier comments. For this section, we focus on topics or issues that do not directly relate to the questions posed, but are related to sex, gender, and data collection by government agencies.

Sexual identity

While we explicitly focused the consultation paper on gender and sex, submissions also mentioned sexual identity. Those who supported the proposed ‘gender by default’ principle, particularly in the inclusion of gender and sex at birth questions (ie, the two-step process), were also likely to support the inclusion of a sexual identity question, particularly in the census. They proposed that, in order to create policies and allocate funding for all minority groups, Stats NZ would need to collect data about sexual identity minorities as well. Most of the submissions supported the inclusion of gender, sex, and sexual identity questions in government surveys, particularly the 2023 Census.

Some submitters raised concerns about the intersection of gender, sex and sexual orientation. This was beyond the scope of the review.

The 2023 Census

Although the consultation paper did not mention it, the 2023 Census was in the media around the time the consultation went live. Several submissions focused on the inclusion or exclusion of gender and/or sex in the 2023 Census. Most of those who agreed with the ‘gender by default’ principle were supportive of including gender in the 2023 Census, with some supporting use of the two-step process. Those who disagreed with the ‘gender by default’ principle raised concerns that the next census would not collect sex, and only collect gender.

Consultation process

Some submissions offered feedback on the way the consultation was conducted. While some commended how the consultation was run, others thought there had been a clear bias toward certain groups, or that key demographics had been missed.

Those supportive of our consultation process thought we had sought advice from the right groups. They also commended our intent to follow international standards and best practice. Some, however, thought we had not fully consulted with the intersex community or advocates/representatives of cultural minorities (eg, takatāpui, fa’afafine).

Some submitters thought the consultation process was biased towards certain groups, particularly those representing the trans community. Many of these submissions were from individuals or groups who were opposed to the ‘gender by default’ principle.

Data system leadership and Stats NZ

Some submissions commented on Stats NZ’s role within government – specifically, Stats NZ taking the lead on guidance for data collection and data use. Some submitters thought Stats NZ would be setting a good example with the ‘gender by default’ principle and encouraged us to provide more guidance and standardisation for other government agencies.

Other submitters disagreed with the proposed ‘gender by default’ principle and were concerned that data about sex would not be collected across all government agencies, academic institutions, or other private organisations.

International context

In the paper explaining our recommendations, we cite international practice to provide context for what we consider ‘best practice’. While some submitters thought this made sense, some noted that while some processes may work for countries like Canada (which we had cited as an example), this would not necessarily be applicable to Aotearoa New Zealand.

We also note that we received submissions from individuals and organisations in Aotearoa New Zealand as well as from overseas. Some submitters have encouraged us to ignore how overseas submitters are trying to influence NZ processes, while others have encouraged us to listen to overseas submitters who may have more information about how these processes have worked in their countries.

Glossary

natural language processing – also known as NLP, applies computational techniques to analyse natural language. NLP uses machine-learning text-mining and topic-modelling to find keywords, concepts, patterns, and themes in large amounts of text.

qualitative analysis – the process of analysing, understanding, and interpreting meaning in non-numeric, textual data. This includes the analysis of naturally expressed opinions or views by people.

R – programming language and environment for statistical computing and graphics.

thematic analysis – also known as topic analysis; a common form of qualitative analysis. It is used to identify meaningful patterns across a dataset.

Terms related to standards and classifications

classification –a statistical classification is a way to group a set of related categories in a meaningful, systematic, and standard format. The statistical classification is usually exhaustive, has mutually exclusive and well-described categories, and has either a hierarchical or a flat structure. A statistical classification usually contains codes and descriptors.

data system – refers to people and organisations that collect and use data. See NZ Government data system map as an example.

statistical standard – provides a comprehensive set of guidelines for surveys and administrative sources collecting information on a particular topic. See What are statistical standards and classifications for further information on statistical standards.

Appendix 1: Sex and gender identity statistical standards: Submission form

Submission guide

Thank you for taking the time to fill in our submission form.

Please read the consultation paper outlining the proposed changes to the statistical standards and then let us know to what extent you agree or disagree with these. We encourage you to further explain the reason for your rating using the free text field – This will allow us to gain further insight about what you like about the proposed changes, and what could be improved.

Please complete your submission form by 5pm Thursday 13 August 2020.

If you prefer you can send your thoughts to us by email to identity@stats.govt.nz.

Privacy statement

We will collect personal information from you, including your contact information and any information you supply in your submission. This information helps build a better understanding of New Zealanders’ views on proposals for change to the statistical standards for sex and gender identity. We will analyse the information to understand the views of different groups and the range of people we have reached.

We keep your personal information secure by protecting it from outside sources, making regular back-ups of our data and using the best security systems.

You have the right to ask for a copy of any personal information we hold about you, and to ask for it to be corrected if you think it’s wrong. If you’d like to ask for a copy of your information, or to have it corrected, please contact us at identity@stats.govt.nz.

A findings paper summarising the submissions we receive will be published on our website www.stats.govt.nz. We intend to use direct quotes from submissions to illustrate our findings. We do not intend to identify or name individuals through use of direct quotes but may name groups in our findings.

Stats NZ may be asked to release submissions under the Official Information Act 1982. This Act has provisions to protect sensitive information given in confidence, but Stats NZ can’t guarantee the information can be withheld, particularly from groups or organisations. If you don’t want information contained in your submission to be released, you need to tell us which information should be withheld and explain why. For example, you might want some information to remain confidential because it’s personal.

Privacy, security and confidentiality of survey data has more information about your privacy.

Contact details

1. Name of person or group this submission is from: _________________

2. This submission is made:
• by an individual
• on behalf of a group
• on behalf of a business or organisation
• on behalf of a government organisation
• other: _________________

3. Where do you usually live? (for organisations – where are you based?)
• New Zealand
• Overseas

4. Contact email: _________________

5. Contact phone number: _________________

6. Are you happy for us to contact you?
• Yes
• No

7. Gender by default principle

We propose that the ‘gender by default’ principle is adopted in the updated standard. This is an approach that defaults to the use of gender data as opposed to sex at birth. Collection of sex at birth information should be viewed as an exception.

In most cases a person’s gender – their social and personal identity – is most relevant for policy making and research rather than their sex at birth. Gender based analysis is used in a range of areas, from income equality to health and education. Recent guidance has recommended that in most cases when sex or gender information is required, gender is most relevant to collect.

For details see: Proposed solution – ‘gender by default’ principle. 

To what extent do you agree or disagree with the gender by default principle in the proposed standard?
a. Strongly agree
b. Agree
c. Neutral
d. Disagree
e. Strongly disagree

8. Please explain the reason for your rating: _________________

9. 'Gender' concept definition

We propose the following gender definition:
'Gender refers to a person's social and personal identity as male, female, or another gender such as non-binary. Gender may include how a person describes themselves ('gender identity'), and/or the gender a person publicly expresses ('gender expression') in their daily life. A person's current gender may differ from the sex recorded at their birth and may differ from what is indicated on their current legal documents. A person's gender may change over time. Some people may not identify with any gender.'

For details see: Proposed solution – an overarching concept of ‘gender’.

To what extent do you agree or disagree with the proposed definition for gender?
a. Strongly agree
b. Agree
c. Neutral
d. Disagree
e. Strongly disagree

10. Please explain the reason for your rating: _________________

11. Another gender

We propose ‘another gender’ as the category for classifying responses as opposed to ‘gender diverse’. This both renames that classification (which is currently known as ‘gender diverse’) and limits it to those who specify their gender as ‘another gender’.

In the New Zealand context, the term ‘gender diverse’ is often used and understood as an umbrella term, similar to the terms trans or transgender. However, some trans people may not use the term and not all gender diverse people may identify as trans (Oliphant, 2018).

This consistent approach also avoids ascribing an umbrella term that may not be a good fit for some respondents. It will also more clearly indicate that data reported in this third ‘another gender’ category does not represent all transgender people (as many will have selected male or female responses) and avoid the confusion created under the current ‘gender diverse’ classification.
For details see: Proposed solution – ‘another gender’ in gender question.

To what extent do you agree or disagree with the use of 'Another gender' in the standard?
a. Strongly agree
b. Agree
c. Neutral
d. Disagree
e. Strongly disagree

12. Please explain the reason for your rating: _________________

13. Two-step method

We propose use of the two-step method in the updated standard.

This involves asking a question about sex at birth, combined with a question on gender. The two-step approach is considered best practice for use in population representative data collections, where reflecting the transgender population is required. It is also the approach implemented by Statistics Canada in some of their surveys.

For details see Proposed solution – two-step method for identifying transgender and cisgender populations.

To what extent to you agree or disagree with use of the two-step method in the standard?
1. Strongly agree
2. Agree
3. Neutral
4. Disagree
5. Strongly disagree

14. Please explain the reason for your rating: _________________

15. ‘Sex at birth’ concept definition

We propose introducing a specific definition and question module based on ‘sex at birth’ for use in surveys, used solely in the two-step method (where identifying transgender populations is required).

Sex at birth refers to the sex recorded at a person’s birth (eg, recorded on their original birth certificate).

For details see: Ambiguity in the current sex standard.

To what extent do you agree or disagree with use of the sex at birth concept in the standard?
a. Strongly agree
b. Agree
c. Neutral
d. Disagree
e. Strongly disagree

16. Please explain the reason for your rating: _________________

17. Intersex information needs

We propose adoption of an intersex variation question where intersex population data is required.

Where intersex population data is required, international best practice is to use a separate question asking whether a person was born with an intersex variation.

For details see: Collection of intersex population data is complex.

To what extent do you agree or disagree that this approach will meet information needs for the intersex population?
1. Strongly agree
2. Agree
3. Neutral
4. Disagree
5. Strongly disagree

18. Please explain the reason for your rating: _________________

19. Further information you would like to share

Is there any other information you would like to share to assist us in the review of these standards?

Appendix 2: Themes and sub-themes from thematic analysis

‘Gender by default’ principle

• Agreement with the principle as: 
• a more inclusive approach to data collection 
• an opportunity to improve data quality 
• having potential to improve research and policy outcomes. 
• The importance of both variables for specific purposes
• Disagreement with the principle, including: 
• concerns that data quality would worsen 
• noting that gender data was not suitable for all purposes 
• highlighting various legal and ethical concerns. 
• Issues affecting how the principle would work in practice (neither agreement nor disagreement).

Gender concept definition

• Agreement with the definition as:
• a more inclusive approach
• reflective of lived experience 
• a broadening from the existing definition of gender identity. 
• Disagreement with the definition, including: 
• with emphasis on conceptual issues 
• highlighting the need for inclusion of Māori and Pacific identities 
• questioning whether people with no gender fit into the definition. 
• Other aspects to consider (neither agreement nor disagreement).

Another gender

• Agreement with use of ‘another gender’ category, including:
• a more inclusive and enduring approach 
• the ability to further define and/or include additional categories
• added clarity and consistency of data collection.
• Disagreement with use of ‘another gender’ category, including:
• concerns about ‘othering’
• concerns about the classification worsening data quality
• maintaining existing practice.

Two-step method

• Agreement with the proposed method, with:
• an emphasis on best practice and inclusivity
• concerns about how and when questions are asked.
• Disagreement with proposed method, with:
• concerns around the two-step method itself.
• Other aspects to consider (neither agreement nor disagreement) include:
• operationalisation of the method and the use of the data.

‘Sex at birth’ concept definition

• Agreement with the proposed sex at birth concept, with:
• improved clarity and ability to meet data needs 
• suggestions on specific terminology amendments
• further user guidance required. 
• Disagreement with the proposed concept, with:
• concerns about sensitivity, privacy and confidentiality
• the definition 
• the response options.

Intersex information needs

• Agreement with the approach, for:
• representation and visibility of intersex people 
• meeting intersex data needs.
• Disagreement with the approach, as:
• a separate question may be othering
• this is private medical information
• Other aspects to consider (neither agreement nor disagreement) include:
• concerns about a lack of understanding of what intersex means
• further user guidance 
• diverse ways intersex people identify
• further consultation needed with intersex advocates/experts.

Further information shared

• Sexual identity
• 2023 Census
• Consultation process
• Data system leadership and Stats NZ
• International context

Appendix 3: Decision diagram

A step-by-step guide to determining if and how to collect sex and gender data

Text alternative for Decision diagram, a step-by-step guide to determining if and how to collect sex and gender data

The decision diagram outlines a process survey developers and other data collections can use to select an appropriate question or questions about sex and gender, depending on the information need.

It contains three steps as three flow diagrams.

Step one asks the user to consider whether sex or gender information is in fact needed.

Step two asks the user to consider whether it is necessary to identify transgender and cisgender respondents in the collection.

Step three asks the user to consider whether intersex population data is required.

Question examples are provided in each step to show the user appropriate ways to ask for gender, sex at birth, and intersex information from respondents.

ISBN 978-1-99-003229-5

Media enquiries

Natalie Quirke
04 931 4444
communications@stats.govt.nz