Data Corner #3: Gender representation in IT, 2014 vs 2018

Previously on Data Corner, we looked at the male-to-female ratio of Australian university staff and students in 2014 and 2018, with a focus on the discipline where we saw the biggest gender skew: engineering.

We found that at across all fields in both 2014 and 2018, from undergraduates to mid-career academics at Level C, women made up 40–60% of people at each level, which falls within the acceptable range for gender representation. However, in some specific disciplines such as engineering, the proportion of men never dropped below 75%. This paints a very different picture to the overall data.

In this edition, we are putting the spotlight on Information Technology (IT) in recognition of the 10th anniversary of Girls in ICT Day on the 22nd of April 2021. On this day, marked annually on the fourth Thursday in April, the United Nations International Telecommunication Union (ITU) highlights the need to promote technology career opportunities for girls and women in the world’s fastest growing sector.

Based on the scissor graph in Figure 1 below, it’s immediately obvious that IT is a male-dominated discipline, with the proportion of men never falling below 60%. There’s very little change between the 2014 and 2018 data, but we can see a slight improvement in female representation amongst postgraduates and at Level B, and a slight decline at Level C.

Interestingly, the proportion of women at Level A is double the proportion of female undergraduates, but there is a significant drop-off beyond these levels.

Scissor graph of male and female representation for university students and academic staff in IT.
Figure 1. Comparison of the gender distribution of Australian university staff and students in IT in 2014 and 2018. Source data: Department of Education, Skills and Employment (DESE) Higher Education Statistics.

This decline is in line with the overall picture, where we see a decrease in female representation at Level B and above. However, when we compare the gender profile for IT with the gender profile in academia overall, a noticeable difference is the initial increase in female representation in IT between undergraduate and Level A, compared to the largely flat figures at these levels when looking at all disciplines (Figure 2).

Scissor graph of male and female representation for university students and academic staff.
Figure 2. Changes in gender distribution of Australian university staff and students between 2014 and 2018 across all Fields of Education (for students) and Academic Organisational Units (for staff). Source data: Department of Education, Skills and Employment (DESE) Higher Education Statistics.

This suggests that female IT undergraduates are entering postgraduate study and early career academic roles in the field; however, when progressing to senior academic positions, there are challenges in career progression or workplace culture that disproportionately affect women. There is more work to be done to investigate why there is such a stark drop-off in the proportion of women between Levels A and B in IT.

While you’re here, check out our Q&A with Dr Arti Agrawal, Director of the UTS Women in Engineering and IT Program. In the interview, Arti talks about the program’s initiatives for girls and women at all stages in their trajectory – from primary school through high school and university, all the way to industry professionals and professors.

For more activities and ways to get girls involved in IT, take a look at ITU’s toolkit for organising a Girls in ICT event.

Creating equitable academic workloads – lessons from the US

It will come as no surprise that the same inequities in workload allocation we found in our analysis of the SAGE Bronze Award applications are replicated internationally – women academics spend more time on teaching and service than men do. Not apparent in our analysis, but also of no surprise is that underrepresented minorities do more diversity and mentoring work than their majority counterparts. And, as we all know, these inequities have real consequences for progression and retention of women and other underrepresented groups.

The Faculty Workload and Rewards Project (FWRP) at the University of Maryland was a five-year action research project, working with 51 academic units across 20 US universities, to establish equity-minded reforms to workload.

The Project identified six conditions that support equitable workloads and suggested ways to promote these.

ConditionExamples of ways to promote this condition
Transparency – data on workload is visible to everyoneDashboards
Transparent policies & practices for assigning work
Clarity – expectations of workload are clearPublished expectations guidelines
Compensation for key roles
Credit – the full range of work activities are recognised & rewardedCredit systems policy
Credit swap system
Norms – workload equity is normalised and reinforcedPlanned rotation of roles and tasks
Context – workloads respond to individual strengths, interests and demandsDifferentiated workload policy with options for negotiated deviations
Accountability – mechanisms are in place to ensure staff fulfill their workload obligationsReduced & restructured committees
Statement of expectations
Table 1. Conditions and corresponding practices that support equitable academic workloads, as identified by the Faculty Workload and Rewards Project.

They then worked with individual academic units to drive systemic change to reform their workload allocation through training on workload inequity, creating a department workload dashboard, and developing an equity action plan. They supplemented this with voluntary, individual professional development on assignment and uptake of, and reward for, work activities.

A report on the project has been published, with many of the supporting tools also being made available. Additionally, a webinar on the Project is available, providing an overview of the Project from the Principal Investigator, and also reflections from two Heads of Department involved.

LitBits #2: Supporting Indigenous academic staff

We know from SAGE Athena Swan applications that many Australian higher education and research (HER) institutions have strategies in place to improve the attraction, retention and progression of Aboriginal and Torres Strait Islander staff. It got us thinking: wouldn’t be interesting to explore what institutions overseas are doing and thinking in this space for their Indigenous peoples?

In this LitBit, we look at promising initiatives for Indigenous scholars in North America and Aotearoa (New Zealand). First, we must acknowledge that this piece was written from our subject positions as non-Indigenous people. Don’t take our word for what works! When designing initiatives for Indigenous communities (or indeed any equity group), always abide by the golden rule: not about us without us. Partnership and co-design are essential in creating programs that truly respect and meet needs.

Helping everyone “succeed”: What should initiatives aim to achieve?

Consider the example of Māori academics in Aotearoa, who feel a lower sense of belonging in academia.[1] This is partly because their career motivations (and what constitutes meaningful work for them) don’t align with Eurocentric ideals of scholarly excellence. They are driven by a desire to serve and make visible Māori communities in their work – goals which are not always supported by increasingly individualistic and corporatised academies.

Thus, change initiatives must not further alienate Indigenous academics by insisting that they conform to notions of success that are not culturally relevant to them. It’s time for the HER sector to broaden its concept of success beyond institutional achievements such as graduation or promotion. For Indigenous peoples, true success includes being a core part of the academy without sacrificing their cultural integrity.[2,3]

Here are some ways to create institutions that are respectful, welcoming, and inclusive of Indigeneity.

Enhance Indigenous leadership and policies

Western institutions are managed in ways that often conflict with Indigenous governance models and cultural protocols.[2] This could be mitigated by incorporating (more) Indigenous leadership on influential senior committees, such as the Board and the Senate. A handful of British Columbian universities have established Aboriginal Advisory Committees, where Elders, Aboriginal community leaders and institutional representatives work together on Indigenous initiatives.

Institutional Indigenous strategic plans can set goals and articulate policies for promoting Indigenous research, whether by affirming Indigenous methodologies or supporting projects that are relevant to Indigenous communities.

No strategy is complete without the ability to monitor the impact of these policies over time. To enable this, HER institutions should collect standardised, high quality ethnicity data and commit to publicly reporting this data.[4] In Aotearoa, Māori representation in higher education is currently difficult to estimate, due to variable response rates and differences in recording staff numbers (as full time equivalents or headcounts, for instance).

Embed Indigenous knowledges in the staff and student curriculum

Interactions with non-Indigenous peoples form a major part of Indigenous peoples’ everyday environments. A fix-the system approach therefore requires the former’s attitudes to be founded on a truthful understanding of Indigenous histories, cultures and lived experiences. Some Canadian universities (e.g. University of Winnipeg and Lakehead University) have set Indigenous content requirements for all undergraduate students to address non-Indigenous people’s low awareness of Indigenous history and contemporary issues.[2]

Pihama and co-workers recommend formalising cultural competency requirements for non-Indigenous research supervisors.[5] This was based on an observation that Māori postgraduate students frequently encounter tensions arising from the cultural ignorance of non-Indigenous faculty and their consequent dismissal of Indigenous research methodologies.

Promotion

There is a growing realisation that traditional measures of merit in academia are often too narrow and inequitable. For example, McAllister et al. noted that Māori researchers are more likely to obtain PhDs later in life.[4] Indigenous scholars may adopt academically rigorous yet culturally relevant research practices, which are pivotal in bridging relationships with Indigenous communities.[2] However, the nature of their research may mean that Indigenous scholars do not use the same dissemination processes as their non-Indigenous counterparts, and are thus disadvantaged by academia’s emphasis on short-term outputs rather than long-term outcomes, and its unhealthy obsession with single-author publications in high-impact journals.[4] Promotion policies need to better recognise these non-traditional research trajectories.

Institutions should also value service contributions more highly in their performance review and promotion processes.[3] Staff from underrepresented groups tend to take on more committee and service work than their majority colleagues, often at the expense of their research. Institution leaders must communicate that equity, diversity and inclusion is not the sole responsibility of those from underrepresented groups. When all staff take ownership of inclusion work, we can reduce the hypervisibility that minority group staff are often subjected to.

Recruit for diversity, not tokens

Haynes Writer and Watson suggest these good practices for recruiting Indigenous and minority ethnic faculty.[3] Firstly, recruitment panels should include members of underrepresented groups. To avoid overburdening the small number of staff from these groups, institutions can also invite diverse community brokers to join their search committees. Non-majority panellists can provide unique perspectives that help broaden and diversify the candidate pool. They are also in a better position to share cultural insights with the diverse candidates that the institutions seek to hire.

Cluster and target hiring can help create a community of new hires, thus reducing their risk of isolation. Identified positions are not without controversy, which usually arises when individual hires are tokenised and valued only for their diversity. This can be avoided by emphasising the scholarly expertise that they bring to the institution.

Institutions may also opt for grow-your-own strategies, where they invest in developing existing talent. Staff or students from underrepresented groups could be encouraged to pursue higher degrees (e.g. by providing financial incentives) or undertake leadership and teaching opportunities, with the view of transitioning into a longer term or higher-grade position at the institution.

Mentoring

The MAI Te Kupenga Doctoral Programme is a Māori-driven initiative, established in Aotearoa to support Indigenous scholars and challenge “whitestreaming” in tertiary education.[5] Participating scholars engage in hui (Māori cultural gatherings) and workshops, where they can interact with senior Māori academics outside their supervisory teams. MAI Te Kupenga’s collective mentoring model thus connects scholars to a much larger pool of Māori role models. The resulting support networks provide a space where Indigenous knowledges and methodologies are affirmed, and Māori scholars feel comfortable sharing their experiences and strategies for dealing with structural racism. The cross-institutional nature of the program keeps individual institutional structures at arm’s length, and enables the achievement of critical mass.

Mentoring is also a valuable addition to induction processes for new staff. A good mentor can introduce them to relevant networks and familiarise them with the institution’s political context. By taking a more proactive approach in helping employees integrate into their new workplace, institutions can greatly increase retention.

Culturally safe spaces

Indigenous staff and students benefit from having dedicated sanctuary spaces where they can relax their academic personas and find affinity with fellow Indigenous people. On university campuses, these areas may be Indigenous academic units or student services.[2,5] Safe spaces should honour Indigeneity, for example by featuring culturally appropriate artwork and physical spaces, such as circle lounges.[1]

Whether these spaces are located centrally or on the fringes of the campus, in buildings that are well-maintained or require significant repair, centrally funded or reliant on short-term external grants is a powerful symbol of how much an institution values Indigenous presence. Reclaiming institutional space is an important, physical step towards Indigenising the academy.

Indigenisation is not a pipe dream

Altogether, these initiatives acknowledge and attempt to undo the systemic racism that pervades academia. It is heartening to note that Indigenous scholars do not report experiencing issues of racial and cultural discrimination at Indigenised academies,[2] because it proves that mainstream universities too can create similarly inclusive environments.[5] All we need is institutional leadership and commitment.


References

1 Staniland et al. (2020) ‘Fit’ for whom? Career strategies of Indigenous (Māori) academics, Higher Education, 79, 589-604.

2 Pidgeon (2016) More than a checklist: Meaningful Indigenous inclusion in higher education, Social Inclusion, 4, 1, 77-91.

3 Haynes Writer et al. (2019) Recruitment and retention: An institutional imperative told through the storied lenses of faculty of color, Journal of the Professoriate, 10, 2, 23-46.

4 McAllister et al. (2019) Why isn’t my professor Māori?, MAI Journal, 8, 2, 236-249.

5 Pihama et al. (2019) MAI Te Kupenga: Supporting Māori and Indigenous doctoral scholars within Higher Education, AlterNative, 15, 1, 52-61.

Data Corner #2: Gender representation in engineering, 2014 vs 2018

Data Corner #2: Gender representation in engineering, 2014 vs 2018

Previously on Data Corner, we examined the male-to-female ratio of all university staff and students in 2014 and 2018. From undergraduates through to mid-career academics at Level C, we found that women comprised between 40% to 60% of people at each level for both years.

Since we consider 40% to 60% an acceptable range for men’s or women’s representation, this paints a mostly rosy picture of gender distribution in Australian higher education. But if you’ve ever been to a physics or nursing class, you would know that the gender composition of academic staff and students varies wildly by discipline.

In this post, we’ll discuss a classic example of a discipline with a very pronounced and persistent gender skew: engineering. Of all the disciplines we studied,[1] the largest male/female gender gap was observed here.

Line chart comparing the percentage of male and female staff and students in engineering, by level, in 2014 and 2018.
Figure 1. Comparison of the gender distribution of Australian university staff and students in engineering in 2014 and 2018. Source data: Department of Education, Skills and Employment (DESE) Higher Education Statistics.

Let’s take a look at the scissor graph in Figure 1. The first thing we notice is that the overwhelming majority of engineering staff and students are male. This is true at every level: the proportion of men never drops below 75%. We also see that the relative proportions of men and women at each level are quite similar in 2014 and 2018. It’s only at Level B that we observe any significant increase in the proportion of women.

The lack of female students in engineering is particularly striking because when it comes to university student enrolments and completions, women outnumber men in almost every other discipline.[2] Generally, the proportion of women in a discipline starts to fall below 40% after Level B or C, which suggests that there are challenges in career progression or workplace culture that disproportionately affect women. In other words, most disciplines struggle to retain women long enough for them to fill senior academic positions.

That’s not to say that such challenges don’t exist in engineering. Rather, the data indicate that there are bigger barriers to attracting women to study engineering in the first place.

Picture of woman inserting cable into machine.

According to the 2019–20 Youth in STEM Research data, compared to girls between the ages of 12 and 13, a smaller proportion of girls aged 14 to 17 declared interest and confidence in engineering as a subject.[3] Curiously, this was not the case for other STEM subjects, where similar levels of interest and confidence were reported by older and younger girls. In both age groups, boys were more likely than girls to say they were interested in engineering, and to feel that they would perform well in those subjects.

It’s also noteworthy that in the 12 to 13 age group, nearly equal proportions of girls and boys aspired to become engineers (14% and 15% respectively). In the 14 to 17 age group however, a gender gap emerges: only 7% of girls indicated they would pursue an engineering career, compared to 20% of boys. For women aged 18 to 21, that figure is even lower – merely 5% stated they would like a career as an engineer. It’s hardly surprising, then, that less than 20% of engineering undergraduates in 2014 and 2018 were women.

These findings suggest that efforts to draw more women into engineering should target early or pre-secondary students and aim to develop and maintain girls’ interest in engineering throughout high school. Many universities do have school outreach programs to encourage girls to consider further study and careers in engineering. The University of Technology Sydney, for example, runs a suite of programs for primary and high school students as part of their Women in Engineering and IT initiative.[4] Likewise, the University of Western Australia delivers the Girls in Engineering outreach program, where current UWA students and industry partners visit Years 7 to 12 students in schools and guide them through exciting engineering challenges.

For those wishing to improve their engineering outreach programs, researchers from Monash University offer the following advice:

  • Long-term interventions (at least 12 weeks) are necessary to achieve sustained change. For this reason, the UTS engagement programs are designed to be multi-touchpoint instead of one-off events.
  • Measure long-term impact by collecting longitudinal data. We need to know if the students who participate in these programs go on to study engineering when they’re older and if not, why not?
  • Emphasise the importance of non-academic qualities such as interpersonal skills and empathy to becoming a successful engineer.
  • Focus explicitly on engineering instead of science and maths in general. Choose activities or examples that have high personal relevance and social appeal. The University of Queensland’s Women in Engineering program includes workshops that showcase engineering as a way to help people and solve real-world problems. Participating students get to build prosthetic limbs, design affordable flood-proof housing and improve healthcare access for remote communities.

Notes and references

[1] To gauge the differences in gender composition between academic disciplines, we created separate scissor graphs (like the one in Figure 1) for ten Academic Organisational Unit Groups, as defined by DESE, namely: 1) Natural and Physical Sciences, 2) Information Technology, 3) Engineering and Related Technologies, 4) Architecture and Building, 5) Agriculture, Environmental and Related Studies, 6) Health, 7) Education, 8) Management and Commerce, 9) Society and Culture and 10) Creative Arts.

It’s important to note that Academic Organisational Unit Groups are fairly broad categories; for example, Natural and Physical Sciences encompass mathematical sciences, physics and astronomy, chemical sciences, earth sciences, biological sciences and other natural and physical sciences. Thus, there are likely to be variations in the gender composition between the sub-disciplines in each Group as well.

[2] Department of Education, Skills and Employment, Higher Education Data Cube (uCube).

[3] See also: Department of Industry, Innovation and Science (2020) Youth in STEM Research 2019–20: Summary of Results.

[4] Hear more about the design and strategy of the UTS Women in Engineering and IT programme in our webinar Engaging Students as Partners in Gender Equity, Diversity and Inclusion.

Most common issues when allocating workload

Most common issues when allocating workload - header image

According to the Athena SWAN Bronze Award applications from the SAGE pilot, the top 3 issues with staff workload allocation are:

  1. Service work is not evenly distributed;
  2. Workload components are distinctly gendered;
  3. Women are more likely to be overloaded with committee work.

Download the infographic below to learn more about these issues and other common themes.

Data Corner #1: Comparing gender representation in universities between 2014 and 2018

Has women’s representation in higher education improved over the course of the SAGE pilot? It’s still early days, but data from the Department of Education, Skills and Employment (DESE) show promising signs of progress.

Scissor graph of male and female representation for university students and academic staff.
Figure 1. Changes in gender distribution of Australian university staff and students between 2014 and 2018 across all Fields of Education (for students) and Academic Organisational Units (for staff). Source data: Higher Education Statistics, Department of Education, Skills and Employment (DESE).

 

From the scissor graph (Figure 1), we can see that the gender ratios of students and Level A academic staff didn’t change much between 2014 and 2018. At Level B and above, however, there were visible increases in women’s representation. The greatest increase was observed amongst academic staff above Level C. There is, however, still a long way to go to achieve equitable representation at senior levels.

We calculated the 5-year change to the gender gap at each level using this formula:

Equation for gap change

If the gap change has a positive value, it means the gender gap is widening. Conversely, a gap change with a negative value signifies that the gender gap is closing. The heat map in Figure 2 illustrates the changes in the size of the gender gap at each level between 2014 and 2018. Do note that the gap change value does not, by itself, indicate whether the gap favours men or women.

Heat map showing gender gap changes for university students and academic staff by level.
Figure 2. Gender gap changes across academic levels. Source data: Higher Education Statistics, Department of Education, Skills and Employment (DESE).

 

When all disciplines are considered, women make up the majority of undergraduate and postgraduate students, and Level A and B academics. This trend is reversed at Level C and above. However, if we consider gender balance to be 40% women, 40% men, 20% people of any gender, the gender distribution at Level C still falls within the acceptable range; gender inequity is only apparent above Level C (i.e. at Levels D and E). Applying a 40:40:20 approach is inclusive of those identifying outside of the gender binary, which is a key principle of the Athena SWAN Charter. Compared to a 50:50 approach, it also better accommodates minor fluctuations and is more realistic when working with small numbers.

These changes in gender representation suggest good news for the sector, but there are limits to what this data can tell us. Academic staff numbers at Level D and above are only provided as an aggregate in this data set, so it’s unclear what the gender distribution looks like at each individual level. While we suspect that the number of women decreases as level increases, disaggregated data is essential to verify this and herein lies an important attribute of the Athena SWAN approach.

Furthermore, the DESE data appears to exclude research-only staff.[1] We know from SAGE Athena SWAN Bronze Award applications that women are generally underrepresented amongst research-only staff and overrepresented in teaching-only staff. This will likely mean that women’s representation is inflated in this data set. And, of course, this data set only captures binary gender data.[2] We suspect that disaggregation by all genders would not change the overall picture substantially, but nonetheless, it’s important to acknowledge the limitation.

Are some disciplines making more progress towards gender equity than others? In the next Data Corner, we’ll take a closer look at how the gender gap has changed in the Engineering disciplines.


[1] DESE Staff Time Series | Definitions and Notes: Data Notes

[2] The data notes for the student data state that “Students who have requested their gender to be recorded as neither male nor female are counted as female.” We could find no indication that this is different for the staff data.