Are more women leaving the profession in the UK because they have a better understanding of the state of play? Or are the numbers misleading? Gill Matthewson investigates.

An alarming set of UK government data seems to say that women have wholesale left the profession in recent years! This has led Christine Murray, the driving force behind the excellent AR/AJ Women in Architecture awards and the founder of the AJ Women in Architecture survey, to wonder if maybe all her work has not resulted in better conditions and levels of appreciation but women getting the hell out of Dodge.1

I’m a data sceptic. I do not trust data until I know where it’s from and what has been done to it. The information Murray refers to comes from the UK’s Department for Digital, Culture, Media and Sport (DCMS): the DCMS Sector Economic Estimates 2017: Employment Report. So far so good. One might expect data from government sources to be reliable. However, closer scrutiny reveals a number of problems. Here is the data for is for those of you who can’t be bothered digging through the spreadsheets.

DCMS Sectors Economic Estimates – Employment, Architecture, July 2018 (from Table 24)

2012 2013 2014 2015 2016 2017
Male 64,000 67,000 74,000 59,000 63,000 72,000
Female 26,000 26,000 27,000 31,000 35,000 31,000
Total 89,000 94,000 101,000 90,000 80,000 104,000

Is anyone else looking at those figures and thinking, WTF? Between 2014 and 2015, did 15,000 men (one in five of those employed in 2014) really leave employment in architecture, while 4,000 women came on board at the same time? Of course, the architectural workforce population does fluctuate with the economy, but economic fluctuations do not usually favour women’s numbers increasing. (For example, in the USA, the GFC hit more women than men.) And from 2016 to 2017, were 9,000 more men employed and did 4,000 women leave? Really? These last figures are the ones that triggered Murray’s musings on Dezeen under the heading “By highlighting the problems faced by women in architecture, are we making it worse?

Such huge fluctuations don’t  correspond to what is known from other data sources. For example, the data on registered architects in the UK shows a steady increase in the numbers of women. There is a bit of a dip for men’s numbers, then a rise, but nothing like the fluctuation shown in the DCMS data.

Registered Architects, UK ARB

2012 2013 2014 2015 2016 2017
Male 26,755 26,728 26,300 27,508 28,505 29,191
Female 7,319 7,538 8,857 9,170 10,006 10,796
Total 34,074 34,266 35,157 36,678 38,511 39,987

The numbers in the DCMS table are also much higher than the numbers of registered architects – it suggests that only one third are registered. We do know that the architectural workforce is larger than the numbers of registered architects, but these figures also seem out of kilter. For example, comparing data from the Australian Census to Australian registration figures, I estimate that around a third of the architectural workforce is unregistered in Australia. Anecdotal evidence suggests that registration is more strongly supported in the UK than Australia, so the suggestion that such a large proportion of the UK architectural workforce is unregistered seems suspicious.

So, what is going on? One clue is the word ‘estimates’ in the DCMS data table title. Another is the fact that the figures are rounded to the nearest 1,000. That’s quite a big rounding for an architectural population that is not high to begin with. The data is based on the Annual Population Survey (APS), which in turn draws on Labour Force Surveys. 2 The APS surveys around 320,000 people. The UK population was around 66 million in 2017, so that’s a survey of roughly 0.5% of the UK population that is sampled to be geographically representative. But, according to the APS website: “The sample design provides no guarantee of adequate coverage of any industry, as the survey is not industrially stratified.” If it were then, proportional, maybe 194 registered architects might be scooped up in that survey and an unknown number of unregistered. But just a handful more or less of women, of men, of any architectural worker, and the estimates for architects would swoop up and down. As indeed they do. Reviewing the longer trends is probably a more useful way to consider this DCMS data than the year-to-year fluctuations. I strongly suspect the 2018 figures will flip back again.

This is not to say that Murray’s musing around the need to celebrate women architects and the creepiness of ‘benevolent sexism’ aren’t worth reading. They are. But the premise that women are leaving in droves because of one year’s fluctuation in one very much approximate data source is not really sustainable.

I don’t think we should be beating ourselves up about putting women off. Yes, there might be some unintended consequences of the swell of activity and campaigns in recent years, but the word we hear on the street is that the gender equity research and advocacy across the world is making women architects feel less alone and giving them ammunition for fighting. My favourite story is from a director of a large practice who tells of young women turning up to their performance reviews with the Parlour Census Report tucked under their arms. None of that gender pay gap for them! Forewarned is forearmed!


  1. Christine Murray, “By highlighting the problems faced by women in architecture, are we making it worse?” Dezeen, 8 March 2019.
  2. DCMS Sectors Economic Estimates 2017: Employment Report, page 2,