How can human factors close the gender data gap?

This commentary paper will describe how the discipline of human factors and ergonomics (HFE) can help to close the gender data gap, which is prevalent across many domains and arises due to a lack of data capturing female metrics and viewpoints. HFE is a domain‐independent discipline that seeks to understand human performance and well‐being with respect to the interactions that humans engage in and the environments that they inhabit. HFE therefore presents an opportunity to understand how gender influences human performance, effective design, social interactions, and environmental factors. This paper argues that a sociotechnical systems approach is essential when reviewing equality, diversity, and inclusivity issues, without which attempts to close the gender data gap will not go far enough. Following the sociotechnical systems approach in HFE, the micro‐, meso‐, and macro‐levels of system design with respect to closing the gender data gap are reviewed. We discuss these issues in relation to a case study example of a crash test dummy. A checklist approach for researchers is presented, which identifies key questions that prompt where gender should be considered in the research process across these levels of sociotechnical systems.

systems, procedures, technologies, and equipment that do not enable females to have equal levels of safety, opportunity, or well-being in comparison to their male counterparts.Instead, a "default male" approach is taken to the design and development of equipment and systems (Criado-Perez, 2019).This is responsible for females' increased injury risk when traveling in road vehicles (Linder & Svedberg, 2019a).Similarly, poor-fitting personal protective equipment (PPE) reduced nurses' protection from coronavirus disease 2019   (Fidler, 2020) and stab vests provide less protection and more discomfort for female police officers due to being modeled on the male body form (Niemczyk et al., 2020).It should be noted that there may be occasions when biases in data collection may also be toward women in areas where women hold the majority population; however, it is the case that generally the bias is in favor of males, especially in engineering and design-based disciplines (Criado-Perez, 2019;Upchurch, 2020).
Gender equality is a United Nations (UN) sustainability target (UNECE, 2022), which highlights the importance and priority that needs to be given to these issues on a global level.The UN states that gender equality must eliminate discrimination against women and girls, empower women, and achieve equality between men and women in human rights, humanitarian action, peace, and security.
Human factors and ergonomics (HFE) can play a meaningful role in solving global social, environmental, and economic challenges (Moray, 1995).As a discipline, HFE holds considerable opportunity to close the gender data gap (Read et al., 2022).HFE is defined by the International Ergonomics Society as "the scientific discipline concerned with the understanding of interactions among humans and other elements of a system, and the profession that applies theory, principles, data, and methods to design in order to optimize human well-being and overall system performance."When we consider the human, their interactions, well-being, and impact on system performance, we must consider the individual characteristics that influence these human experiences to be inclusive.This is the aim of gender-equitable human factors (Parnell et al., 2022).
The discipline of physical ergonomics, which is a subcomponent of the field of human factors, has strived to represent female and male body types to inform the design of systems that are inclusive of both sexes.Design standards include a need to account for the 5th percentile female and 95th percentile male to account for approximately 90% of the human population.This recognizes the different body sizes of males and females; however, it does not recognize the different ways in which males and females may use the products or systems.In 2006, the International Ergonomics Association set up the "Gender and Work Technical Committee" to review the role of sex and gender in ergonomics, which has highlighted the important role of gender in physical ergonomic assessments and encouraged further research into the area (e.g., Habib & Messing, 2012;Laberge et al., 2020Laberge et al., , 2022)).This has highlighted issues surrounding the invisible work of women (e.g., Salerno et al., 2011), the inequality in the employment of women, and their experience of the workplace (e.g., Riel et al., 2017).A recent report into gender and workplace safety by the British Occupational Hygiene Society (2023) has shown that there is still a clear disparity in the health of female workers compared to their male counterparts.Women are more likely to have their health suffer from work activities and they are also more likely to go on long-term sickness leave due to occupational-related health issues (British Occupational Hygiene Society, (2023)).Furthermore, the BOHS report claims that the way in which work-related health issues are reported and the tools used to alleviate them, or lack thereof, are contributing to the gender disparity.
Due to the social component of gender, in contrast to sex, it is important to review the broader societal factors that relate to gender and how this influences the experiences of males and females when reviewing gender disparities.This is where a sociotechnical systems approach is useful, in understanding the broader social, cognitive, and environmental factors that influence how gender influences system performance.The value of HFE is in its targeting of different levels of a system when considering the user experience and wider sociotechnical pressures (Bertalanffy, 1968).Complex systems and systems-of-systems are often segregated into micro-, meso-, and macrolevels to understand the interactions between different elements of the system and the wider environmental context of the domain that they inhabit (e.g., Thatcher et al., 2016).We propose that this is a useful approach for reviewing the gender data gap and the "default male" bias within society, with an aim to ultimately close this gap through a combination of individual, organizational, and societal change.
1.1 | Sociotechnical systems and the micro-, meso-, and macrolevels of human factors assessment Sociotechnical systems theory draws on the development of the complex system approach that argues against looking toward individual elements but stresses the value of interactions between multiple elements that comprise a system and the wider environment within which they are located (Bertalanffy, 1968).A sociotechnical system is one that has both social-organizational and technical components that can be viewed as intra-and interdependent, working toward common goals for joint optimization (Trist & Emery, 2005;Walker et al., 2008).Read et al. (2022) state the value in applying sociotechnical systems theory to the gender data gap issue and the positive impact that HFE can have on closing the gap.The micro-, meso-, and macrolevels of complex systems, as presented in In this commentary paper, we argue that closing the gender data gap requires focused efforts across each of these levels, and we review the methods and tools available within the HFE domain in their ability to enable this.Each level will be considered in turn.

| THE GENDER DATA GAP AT THE MICROLEVEL
HFE analysis at the microlevel is predominantly focused on the characteristics of the individual and how they perform in relation to the system's goals, as well as how they interact with their environment.Research at this level is critical to ensuring that the characteristics of individuals under analysis are representative of the target population, including the diversity within that population.A focus on gender differences in physical ergonomics has identified aspects of the design of equipment and work scenarios that adversely affect females and interventions have sought to adapt the design to meet the needs of females (e.g., Habib & Messing, 2012;Laberge et al., 2020).For example, it was identified that females are more likely to suffer from work-related musculoskeletal disorders, including neck and shoulder pain, when working in office environments (Ekman et al., 2000;Korhonen, 2003).Incorporating an awareness of these differences into the ergonomic setup of office workspaces and developing ergonomic chairs that account for gender differences can allow both males and females to have improved comfort when at work (e.g., Yuan & Tao, 2020).
Yet, more research in this area is needed to identify where designs do not fully meet female needs.A failure to consider the female population when designing and evaluating products and systems has led to numerous gender data gaps that have significant safety and well-being issues for females.One example that has received recent attention is vehicle crash test dummies, which use average male body size to represent the whole adult population within regulatory tests (Linder & Svedberg, 2019a, 2019b).This is taken as a case study example in Section 5. Ergonomics assessments at the microlevel must account for the different body types of females and males.
The view of women as "smaller men"' has been adopted in many domains and further embeds the "default male" status within our societies."Pink it and shrink it" is a term for an identified trend that markets products to females by making smaller and different colored versions of the male equivalent, without accounting for the needs and viewpoints of females (Van Tilburg et al., 2015).This has been common in sports equipment; however, a broader understanding of gender differences is now starting to be realized (e.g., Rasmussen et al., 2022).An understanding that women are not small men has started to feed into the military domain and the design of body armor (Evans et al., 2023).This has required a more in-depth understanding of the gender composition of military personnel and an enhanced anthropomorphic data set.Central to closing the gender data gap at the microlevel is understanding the user population that is being F I G U R E 1 Representation of the micro-, meso-, and macrolevels of complex systems.HFE, human factors and ergonomics.represented within the research and ensuring that it is equitable and reflective of the differing needs of females and males.User-centered design must include a representative user population when sampling and collecting data to inform product design (Kujala & Kauppinen, 2004).For items that have daily consequences (e.g., crash test dummy), the user population should be an equal split of males and females, as is representative of the global population (United Nations, 2022).In more specific areas, females and males are not evenly split (e.g., female representation in Science, Technology, Engineering, and Math (STEM) subjects).When conducting usercentered research within these areas, the representation of females to males should be accounted for in the sampling.This may require a more proactive approach to collect effective and representative samples due to limited access to smaller subsets of the population.In our own work conducting user interviews with commercial airline pilots to inform user-centered design requirements, we sought to include both male and female pilots (Parnell et al., 2021).A total of five male and three female pilots were interviewed, yet data collection took longer to achieve due to increased time sourcing female pilots.Often this can interfere with project and funder deadlines, which leads to more convenience and nonrepresentative participant sampling (Madeira-Revell et al., 2021).This limited sample size also prevented effective data disaggregation by gender.
Awareness of the importance of adequate sampling methods and procedures is required.Sharples (2019) highlights the value of HFE insight and the utility of a qualitative approach to equality, diversity, and inclusivity issues.Interviews and focus groups can obtain indepth information on the perspectives and values of minority groups.
When conducted effectively, they can help to strengthen the voice of those who have not been captured within the standardized approach to system design.

| THE GENDER DATA GAP AT THE MESOLEVEL
The mesolevel is concerned with the collaboration between individuals (i.e., teams), including how they interact and cooperate to achieve a system goal.The application of HFE at this level includes analysis of, for example, the collaboration between military teams (e.g., Roberts et al., 2017;Salas et al., 1995;Stanton, 2011) and the importance of training to improve team performance (e.g., Salas et al, 2009).More recently, there has been a subsection of HFE applied to teamwork within sports teams (e.g., Reynolds & Salas, 2016;Salmon et al., 2009;Salmon, McLean, et al., 2020).This work has included reviewing communication between sports team members (Mclean et al., 2019) and observational methods to assess team performance (Neville et al., 2016).There are specific HFE methods that capture and assess teamwork (Stanton et al., 2017).For example, Event Analysis of Systemic Teamwork (Walker et al., 2006) is a method for assessing teamwork that is becoming increasingly popular within the HFE discipline.The method captures social, communication, and task networks between elements of a system.
The framework has been established and built upon through application to military case studies (e.g., Griffin et al., 2018;Stanton & Harvey, 2017;Stanton & Roberts, 2020), yet such a basis is likely to be largely male-dominant (although the write-up of these studies does not comment on the gender, which in itself perpetuates the gender data gap and reporting, such should be a minimal requirement in attempting to close the gender data gap).
More traditional HFE methods have been adapted to account for team performance, such as the Hierarchical Task Analysis for Teams (Annett, 2004) and Team Cognitive Task Analysis (Klein, 2000).The role of gender in team composition and the impact that it has on teamwork has, however, been little considered within the HFE discipline.In a review of the literature on gender diversity and team performance, Bear and Woolley (2011) found evidence to suggest that gender diversity in teams can improve collaboration, particularly in STEM fields, which have traditionally been male-dominant.
Organizations can play an important role in encouraging more diverse workforces by understanding the different challenges facing male and female workers.For example, when workforces account for work-family balances in the scheduling of work patterns of workers in shifts and atypical work schedules, it can increase gender equity in domains that are more male-dominant (Lefrançois & Probst, 2020).
Yet, the foundations of many applications of teamwork assessments within the HFE domain have been on areas that are male-dominant and have given little consideration for gender, or the different balance that females can bring to gender-diverse teams.For example, the military is still a largely male-dominant domain.Roberts et al. (2017) conducted a study on teamwork within a submarine command and control simulator.A sample of 71 male participants and nine female participants were recruited.While this is not an equal sample, they state this is representative of the target population where females have only recently been permitted to serve on submarines.
However, the results were not disaggregated by gender to show possible differences in averages.The small sample would prevent valuable statistical analysis from being undertaken to understand possible gender differences.Furthermore, assessment of future technology that does not include a larger mix of females within samples may not help to incentivize more females into these fields in the future.
In sports, the role of gender in teams has also been given limited consideration within HFE research.In their study of communication with a football team and the development of a team training method, Mclean et al. (2019) utilized a sample of 25 professional football players, yet the gender of the players is not reported in the paper.It is likely that this was a male football team, due to the male dominance in this sport, further evidencing the "default male" approach taken within research (Criado-Perez, 2019).This was also the case in the assessment of teamwork measures applied to umpires in the study by Neville et al (2016).Here 20 empires participated, and while their experience levels are reported, other individual characteristics such as gender are not provided.In accordance with the Sex and Gender Equity in Research guidelines (Heidari et al., 2016), gender must be reported within the write-up of all research that involves human participants.Where the results are only based on one gender, this must be stated clearly in the title and abstract of the paper to avoid misleading interpretations (Heidari et al., 2016).As a minimum, it is vitally important that all research follows these guidelines to close the gender data gap, yet the HFE domain can, and should, go further than this.To achieve this requires training and guidance for all reviewers of journal paper submissions to ensure these minimum requirements are met.The implications for female participation in dominant male teams is an issue worthy of future research, as is the inclusion of gender when considering team dynamics and how to increase performance through gender diversity.
Increasing gender diversity is also important to consider at the organizational level.There are a number of HFE domains that suffer from gender imbalances within their employees and senior management.For example, in the United Kingdom, females make up only 22% of workers in the transport industry (European Commission, 2021), 11% of the UK regular military forces (Officers Association, 2022), 16.5% of all engineers (Engineering UK, 2022), and 20% of workers in the nuclear sector (Nuclear Skills Strategy Group, 2021).Where they are employed in these sectors, they are also typically in positions of lower responsibility and pay, which limits the impact they can have on decisions and organizational strategy.
Within the medical domain, an area of increasing application to HFE, females make up a more significant portion of workers, for example, 44% of NHS workers are female (NHS Digital, 2019), and within different areas of the NHS, this figure is even higher (i.e., 89% of nurses are female; NHS England, 2021).Yet, even within this field, females are not equally represented in more senior roles (NHS Digital, 2019), which can lead to female voices being ignored within high-level decision-making, as in the case of the Covid-19 PPE procurement (United Nations, 2020b).Efforts are being made across a number of institutions to increase the uptake of females within STEM disciplines, with targeted women-focused initiatives, for example, Women in Transport, Women in Nuclear, and Women in Engineering.This is a positive movement, but reaching gender equity in many of these domains will require a focus on training and changing perspectives on current male-dominant professions from an early age.Therefore, a broader systemic, macro, approach is also required to close the gender data gap.

| THE GENDER DATA GAP AT THE MACROLEVEL
The macrolevel of the system includes the legislation, regulation, and governance of systems as well as the societal and cultural pressures that influence how systems function.The factors at this level have a top-down influence on the gender data gap and how it is realized at the individual level.It is through this perspective that the wider sociotechnical system can be viewed, alongside its contribution to the gender data gap.Legislation and standards can play an important role in ensuring that sex and gender are not discriminated against within the workplace and in society.The UK Equality Act 2010 legally protects people from discrimination based on a number of protected characteristics, including sex.Many other countries have similar legislation, for example, the USA Equality Act and the Australian Antidiscrimination Act.This legislation has enabled males and females to be more equally treated with respect to recruitment, health care, and many other societal systems.There is also now increasing awareness of the importance of equality, diversity, and inclusivity within the workplace and broader society (e.g., Liddy et al., 2022).
Many organizations now have EDI policy and guidance that encourages them to act in a more inclusive manner to understand the different needs of their members and to accommodate them.
Organizations that take responsibility for EDI can increase the inclusivity of their staff and communities, yet there is still a policypractice gap in understanding how to best inform inclusive change within organizations (Scott, 2020).Here, HFE has an opportunity to understand the broader systemic EDI challenges facing organizations and wider society (Sharples, 2019).HFE has developed and applied a number of methods to the study of systemic factors in the analysis of accidents (e.g., Hulme et al., 2019;Salmon et al., 2012) and the assessment of legal practices on human performance (e.g., Parnell et al., 2017).Yet, there has been limited application of systems methodologies to review issues related to inclusivity, equality, and diversity.HFE researchers and practitioners must now start to consider how they play a role in the methodologies that are used and the implications this might have on the gender data gap.
There are several HFE methods that enable systems to be understood at the macrolevel.They can provide an understanding of how policy and laws feed down to the regulators of systems, organizations, and individuals, as well as the broader environment.They can also show how behavior at the individual level may diverge from expected behavior, resulting in adverse events (Dekker, 2016).HFE systems-based methods therefore offer the opportunity to understand how gender differences at the individual level may require different legislation, policy, and regulation across the sociotechnical system to enable more gender equity within our societies.Improved awareness of these possible gender differences by closing the gender data gap can enable a better understanding of where society may be encouraging a male bias and how to adjust it to be more equal.
One of the most popular HFE macrolevel methods is the Risk Management Framework (Rasmussen, 1997), which has been applied across numerous domains as a framework to capture the top-down pressures of government regulation as well as the bottom-up pressures faced by end-users and environmental factors (e.g., Parnell et al., 2017;Salmon, McLean, et al., 2020;Vicente & Christoffersen, 2006).The utility of this framework is its ability to demonstrate the connectivity between different layers of a system in how the day-to-day function of a system performs, as well as how accidents can occur, with the development of the Accimap analysis method (Rasmussen, 1997).An actor map is part of the Accimap analysis, and it identifies all actors with responsibility in the domain of study and/or accident under assessment.When identifying the higher-level actors who have responsibility for equality, diversity, and inclusivity issues, the actor map can be beneficial.Yet, the actor map can also identify actors across the micro-and mesolevels of the system as it Another methodology that has been frequently applied to study the sociotechnical system surrounding HFE issues is cognitive work analysis (Naikar et al., 2006;Vicente, 1995).This methodology comprises several individual methodologies that focus on different aspects of system performance.The individual methods are applied in accordance with the needs and intentions of a specific analysis, in a tailored manner.Work domain analysis provides an initial overview of the actors, functions, and purposes that comprise a system.The output of this is an abstraction hierarchy that provides a structured framework for mapping the high-level purpose and values of a system with respect to the individual objects and their individual functions.This method may, therefore, be useful in mapping out the values of gender equity, as well as other inclusivity values, with respect to the equipment, functions, and environment that comprise the system.This could also be used in combination with actor maps and the Risk Management Framework to identify who has responsibility for equality, diversity, and inclusivity values.In the next section, we consider the micro-, meso-, and macrolevels of the system within a case study of the crash test dummy, whereby a gender data gap had led to significant safety issues for females traveling in vehicles.

| CASE STUDY: CRASH TEST DUMMY
Crash statistics show that females have been exposed to a higher injury risk in vehicle crashes (Bose et al, 2011) and that they are much more likely to sustain whiplash injuries (Kullgren et al., 2013;Linder & Svedberg, 2019a).This is thought to be due to a default male approach to crash test dummy design and vehicle testing (Linder & Svedberg, 2019a).Crash test dummies are used in the assessment of vehicle safety within high-impact frontal and side-impact tests.The models of the dummies used in these tests represent the 50th percentile male, which is then scaled up to 95th percentile male and scaled down to the 5th percentile female, stated to be the equivalent of a 12-13-year-old girl (Linder & Svedberg, 2019b).An average female dummy with female body dimensions and mass distribution is not regulated within vehicle crash testing (Linder & Svedberg, 2019b), with the requirement for only average male metrics set out in UN regulation (UNECE, 2018;regulation numbers 16, 94, 95, 135, and 137).European (EU) policies also use male dummy metrics (Euro NCAP, 2022).
The female body has a different body composition in comparison to males (Carstensen et al., 2012;Linder & Svedberg, 2019b;Young et al., 1983), which impacts on the shape of the upper body and the muscle composition (e.g., Jordan et al., 1999).These differences influence how effective seat belts and head restraints are when an individual is involved in a vehicle crash and, as safety testing is based on male measurements, they leave females more vulnerable to injury and fatalities (Forman et al., 2019).We review this case study from the micro-, meso-, and macrolevels to understand how the gender data gap arose and what could have been done to close it.To do this effectively, an Accimap analysis was conducted to demonstrate the systemic influences, beyond the apparent microlevel issues.
Traditionally Accimaps are developed to review events and causal factors that contribute to an accident, yet here we were not looking at a specific accident.This Accimap aimed to review the events that caused a gender data gap in crash test dummy testing that has resulted in the increased injury risk to women (Kullgren et al., 2013;Linder & Svedberg, 2019b).The Accimap was therefore developed using the detailed review conducted by Linder and Svedberg (2019b), which looked into the use of average-sized male and female occupant models in EU regulatory safety assessment tests and EU laws.This review provided an understanding of the impact of laws and legislation at the higher levels of the actor map framework and Rasmussen's (1997) risk management framework.
References and legislation that were of relevance from within this review were also read to gain a more detailed insight into the legislation and how it was influencing the design of crash test dummies and the crash test procedures.Key events and processes were identified and mapped onto the levels of the Accimap.Events were linked when one event has influence over the development of another event.Once the Accimap was developed, an expert in the methodology, with over 15 years of HFE experience, reviewed it, and required adjustments and clarifications were made.The aim was to provide the actors and events involved in the crash test dummy testing procedure.The Accimap is presented in Figure 2. At the enduser and environment level at the bottom of the Accimap, the roles of the seat belt and vehicle design on the use by male and female occupants are presented.This captures the microlevel interactions between the individual and the vehicle, as discussed in Section 5.1.
It also shows the role the organizational actors have at the mesolevels discussed in Section 5.2 (resource providers and industry/local government levels as shown in Figure 2).The legislation at the national and international levels is presented at the macrolevel, as discussed in Section 5.3.

| Microlevel
There are clear failings within the case study at the microlevel of the system.There was no consideration for the gender split within the demographics of the population that the crash test dummy represents.
The global population is roughly equal in the number of males and females (United Nations, 2022), and therefore a female and male dummy should be used (as a minimum, there are, of course, many intra-gender differences too).The absence of data for female car users prevents gender-disaggregated data on crash severity from being reviewed at the vehicle testing stages.It is only more recently that the statistics resulting from this data gap have been identified (e.g., Forman et al., 2019).The simplicity of not having to involve human users and being able to predetermine the dimensions of the dummy should have enabled the more representative testing to be conducted, but this was not the case.
The reporting of the safety assessment is also misleading as the vehicles are declared safe for all based on the safety testing of only a limited F I G U R E 2 Accimap of the factors that contribute to the gender data gap in crash test dummy testing within vehicle safety regulation.The arrows represent interactions between the events and the actors at the different levels.
T A B L E 1 Actions that can help to close the gender data gap with respect to crash test dummy testing across the different levels of the Accimap and risk management framework.

Accimap level
Action to close the gender data gap

International Committees
International committees must value inclusivity issues and mandate the inclusion of representative samples in data used for policymaking National Committees A coordinated approach to legislation and regulations across different countries in response to missing female occupancy testing is required Government Policy and Budget Directives must highlight the importance of inclusive design and legislate that inclusive design principles are followed Gender equity principles should be included in the standards

Regulatory Bodies and Associations
Regulations must stipulate female occupancy model requirement in vehicle-type testing Inclusivity safety ratings could be developed to incentivize manufacturers to design inclusively Industry and Local Government Vehicle designers must consider the ergonomic differences between male and female bodies Research centers must review statistics by disaggregating by gender and/or clearly stating the sample biases in their tested populations Insurance providers should account for gender-related factors in the assessment of injuries

Resource Providers
A female occupancy model is required Manufacturers should be assessed on inclusive safety Research publishers should refuse to publish work that does not follow the Sex and Gender Equity in

Research guidelines
End-Users There needs to be a public awareness for this issue so end-users know how their vehicles are assessed relative to their own characteristics

Equipment and Environment
Vehicles should be designed with the whole adult population in mind

| Mesolevel
The role of the organization is important to consider at the mesolevel.
The automotive industry has traditionally been heavily male-dominated (e.g., Tranter & Martin, 2013).In the United Kingdom, women only make up 13% of the automotive workforce, and 90% of these women felt that women were underrepresented within leadership positions (Deloitte, 2020).A lack of female representation within the decisionmaking in vehicle testing could be one explanation for the male bias.
Increasing female representation is one way in which more equality can be considered in decision-making processes.This is, however, a slow process that requires a shift in organizational cultural norms and enhanced support for young females in training and education to encourage them into the industry.It is important to consider the gender dominance within the industry of study and statistical resources that can enable these figures to be drawn across a number of domains that are now emerging, due to increased awareness of gender equity within recent times, for example, Deloitte (2020).

| Macrolevel
At the top of Accimap are the International and National Committees that outline the directives and high-level aims of international and national policy.As shown in Figure 2, it is at this top level that the inequality in road safety is initiated through a lack of clarity in the regulation requirements for inclusivity.This feeds down to the development of the regulations and requirements for vehicle-type testing that the vehicle manufacturers must adhere to.
The Accimap includes both the United States and EU regulations that stipulate the need for male occupant model testing only.The impact that this has on female occupants' safety is highlighted by researchers and crash statistics data analysis that reviews genderaggregated data.
F I G U R E 3 Flowchart to determine checklist requirements.
Table 1 presents ways in which the gender data gap and gender inequality within this example can be overcome across each of the Accimap levels of the sociotechnical system.

| CHECKLIST TO HELP CLOSE THE GENDER DATA GAP WITH HFE
As described, HFE and its associated methods can play an important role in closing the gender data gap at each of the levels of a sociotechnical system.Taking these into account, we have developed a checklist that can be used by researchers and practitioners during a research project to understand how they should incorporate gender considerations within their research.This checklist approach is beneficial as it allows a standardized way of considering gender and the ways in which to enact gender-equitable research, which is currently lacking (Madeira-Revell et al., 2021;Read et al., 2022).The development of this checklist has drawn on the researcher's knowledge and experience, as well as the case study presented in the previous section.This checklist aims to go beyond the SAGER guidelines, which focus on the reporting and write-up of data, to encourage the inclusive of gender within the research process itself.
T A B L E 2 Checklist to guide gender equitable research in human participatory research.
Gender Checklist for Human Participatory Research 1. Formulating research idea(s) Consider how gender may impact the research discipline as well as how the current governing of the systems under study may be impacted by gender biases.

Sampling
Understand the gender within your user population by searching databases and statistics resources, do not assume there is an equal gender split.
Conduct representative population gender sampling based on the population breakdown by: • Consider the sociodemographic and other factors that may influence sampling methodology; for example, sampling within universities (or certain departments within them) may not be representative of other populations of people.
• Consider the time and accessibility of research trials-flexibility to different times of day and platforms (e.g., virtual meetings can enhance sampling efforts).
• Allow a sufficient participant recruitment period to ensure balanced/representative samples can be recruited.This can be lengthy and not having enough time can reduce the opportunity for diversity in the sample (e.g., convenience sampling).
• N.B.If the sample is not representative for any reason, this must be stated clearly in the write-up.

Data collection
Ensure ethical acceptance of the study permits the collection of gender data, while also giving anonymity to the participants.
Allow the participant to choose from the following options: Male, Female, Nonbinary, Prefer to self-define (with blank text box), Prefer not to say.The option for including assigned sex at birth could also be included where required, for example, if looking specifically at sex differences.
Ensure the data collection activity is accessible and inclusive.

Data analysis
Disaggregate the data by gender and review the differences between females, males, and other genders identified.
Review the differences between genders, and conduct statistical analysis, where appropriate, to review gender differences.Note that where only one participant of a gender category has been identified, then reporting of this individual will not be anonymous and should not be reported widely.However, the inputs from the individual can be used to inform the analysis, and further data from others in the same gender category should be sought.
Identify the separate gender metrics as well as the total group metrics.

Write-up
Adhere to Sex and Gender Equity in Research guidelines on reporting gender balance throughout the report or manuscript.
Highlight all limitations in sampling within the write-up.
Write up the disaggregated data taking note of the following: • If there are no differences, this still needs to be stated in the write-up.
• Do not make assumptions about gender in the write-up.
Do not apply findings from one gender to another without the evidence to do so.

PARNELL and PLANT | 71
The checklist includes an initial flowchart (Figure 3) that should be followed to understand the focus of the research and which of the two possible checklists are required.The flowchart in Figure 2 starts by asking the researcher if they are conducting research on a particular product or system that involved humans and/or human interactions.If the answer to this is yes, they are directed to our "Gender Checklist for Human Participatory Research" (see Table 2).If  2), analyze data (Data analysis section, Table 2), and consider the wider organizational and societal influences that gender may have on their research (Table 3).As such, we would recommend its use throughout the duration of a project.This is intended to be an initial prompt to researchers and is intentionally domain nonspecific to apply to the multiple domains that HFE is applied to.Ideally, each item should be considered for best practice.Where items have been missed, it is important that these are noted within the write-up of the research, as stated in the "Write-up" section of the checklist, to ensure transparency.This is where the Sex and Gender Equity in Research (SAGER) guidelines are particularly important to be considered alongside this checklist.
Table 3 presents the "Checklist for Systemic Impacts on Gender in Research."This checklist encourages researchers to think about the number of ways in which the broader system surrounding the area of study may be impacted by gender.It is important to consider these when initiating research, as shown in the "Formulating research ideas" section of Table 3.The consideration of these issues is important at the beginning of a research process, to help frame the research study to be more inclusive.They can also be reviewed in relation to the data analysis and write-up stages.
Application and validation of this checklist is required, and we are interested to hear how researchers go about applying this checklist within their own research going forward.Consider the gender composition of organizations involved in the research areas of focus, and their decision-makers, for example, female/male ratios in employment and senior employment roles.

| CONCLUSIONS
Review gender and inclusivity issues within the governance of the system of study, for example, policy reports relating to gender and other sociodemocratic factors.
Consider the gender roles within society and how they may impact the utility and performance of a system, for example, family and caring roles.
Consider the gender composition of researchers in the domain of study and how that may impact the research that has been conducted.

Analysis
Include gender considerations in system modeling outputs, for example, the gender composition of organizations, the gender implications for regulations, and ways to improve gender equality.

Write-up
Detail the wider systemic considerations for gender in the write-up (intro/discussion) or the research, as identified through the above points in the checklist.
research practices should not be overlooked.It is evident that some HFE research has not always considered the role that gender may have on the analysis or area of study.This may, in part, be due to a lack of awareness or unconscious biases that are inherent within our societies, which is why a sociotechnical systems approach is required to review and understand the biases and shift understanding.
This commentary paper has identified how reviewing gender in research across the micro-, meso-, and macrolevels of sociotechnical systems can provide the insights required to help close the gender data gap.The example of the crash test dummy in the case study in Section 5 identified that male dominance in the automotive domain and underrepresentation may have had a role to play in the lack of representative female vehicle safety testing.This feeds into the macrolevel, which presents the importance of reviewing the broader systems interactions and systemic factors that influence how humans experience and perform within sociotechnical systems with respect to gender and the gender roles that are still prevalent within our societies.The methods that are currently widely used to study sociotechnical systems can offer insights into the systemic factors that lead to gender data gaps and therefore can also be used to identify solutions to closing these gaps.The Accimap analysis of the crash test dummy example provides an overview of systemic levels and contributing actors, which is a powerful tool for providing recommendations.
Using the insights from the case study, we have developed a set of checklists to help researchers to understand how to consider gender with respect to their research questions and research practices.The flowchart in Figure 2 provides a process to guide the reader to the checklists.The "Gender Checklist for Human Participatory Research" can be used to assess research targeting gender at both the micro-and mesolevels.This tool will help to improve the consideration of gender throughout the research process.The flowchart also guides users to consider the wider systemic issues within their research, through the checklist for "Systemic Impacts on Gender in Research" (Table 3).This checklist focuses on the conceptualization of the research and the wider gender issues that may surround the domain, as well as how their research can generate meaningful outputs to help close the gender data gap.

Figure 1 ,
Figure 1, are one way in which this can be conceptualized.The levels comprise a holistic model and act as a way of reviewing systems in a bottom-up way with respect to their lower-level individual characteristics, as well as top-down with respect to the influence of societal pressures and their governance.The microlevel of the system refers to individual characteristics, including well-being, individual performance levels, decision-making, and physiological components.This level also captures individual technological components to enable individual performance.The mesolevel refers to the wider teams within which individuals participate and the collective groupings of individuals, such as system review of a particular domain.SeeSalmon,   Hulme et al. (2020)  for a more detailed overview of the Accimap methodology.
population.Females are not made aware of the increased risk they are at when traveling in road vehicles.It is essential that, where biased sampling and data collection occur, these biases are reported clearly to avoid misleading interpretations of the data.
they are not looking at individuals, they are directed to the next question in the flowchart, which asks if they are conducting research into organizations or teams of people interacting together.If the answer to this is yes, they are again directed to the "Gender Checklist for Human Participatory Research," which is also applicable to teams and the mesolevel.Following the direction to this checklist, the flowchart also directs users to the question on considering the broader systemic integration of their study, including standards, legislation, and regulation.The user is then directed to the "Checklist for Systemic Impacts on Gender in Research" (See Table3) if they are aware of the possible impact of systemic factors.If they are not aware of the broader systemic impacts of their area of study, the user is encouraged to think why this may or may not be important to their work and are directed to the "Checklist for Systemic Impacts on Gender in Research" to determine how these issues may impact their research area.The design of the flowchart in this way encourages people to consider the broad systemic implications of their work as well as the systemic nature of gender issues in their field.The Gender Checklist in Human Participatory Research (Table2) steps the user through the different phases of the research process and lists key activities that encourage them to consider gender from the formulation of the research question(s) to the study design, data collection, analysis, and reporting.Working through this flowchart and checklist can help the researcher to understand how to recruit participants (Sampling section, Table HFE holds great potential in bridging the gender data gap across its domains of application.The combination of quantitative and qualitative research methods that are used within HFE enables the performance, requirements, and perspectives of human users to be understood.The user focus is a key, distinguishing, component of the HFE tool kit, and the importance of utilizing a representative sample when using end-users within HFE T A B L E 3 The Checklist for Systemic Impacts on Gender in Research.Checklist for Systemic Impacts on Gender in Research1.Formulating research idea(s)