What does the UK’s scientific workforce look like and how closely does it mirror the overall workforce? These seemingly straightforward questions turn out to be anything but when the numbers are analysed. If we are to understand the diversity of those who work in science we need to know who works where and what their progression is like but, as a recent Royal Society report A picture of the UK scientific workforce has demonstrated, the answers are hard to extract from available data.
You might not expect it to be difficult to know who did or didn’t work in science, but in fact uniform definitions do not exist and different datasets analyse the numbers in different ways. A science teacher, for instance, may or may not be classed as part of the scientific workforce in the different existing datasets. Anomalies such as this pose all kinds of problems when attempting to analyse the effect socio-economic background, ethnicity, gender and disability have on career progression. The Royal Society report is the first attempt to do this and it clearly identifies the issues which surround the datasets that are available for analysis. As a scientist with a keen interest in diversity issues I hope the recommendations the report makes encouraging government departments, UCAS and other key bodies to work to agreed and consistent definitions are heard and followed. Otherwise it will continue to be hard to identify where the hurdles lie for different sections of the population, or what interventions might help them.
There are some headline conclusions that can be drawn from the data available, despite these caveats. It is clear that, for those occupations labelled as science, the proportion of people working in the top two socio-economic classes (SECs) at around 78% is much higher than in the non-scientific workforce (ca 32%); these classifications correspond to managerial and professional occupations. This classification does not tell us anything about socio-economic background and from the data available it is hard to learn much about social mobility. Using the 1970 cohort (data on all children born in one week in 1970) gives some limited insight into this. For instance it is found that the percentage of people entering the scientific workforce steadily declines with decreasing age of parents when they left full time education and that the average time to enter a science-based job depends on parental social class, the children of unskilled workers taking on average 4 years longer to gain a scientific job than the children of those working in the professions.
The pattern of employment amongst different ethnic groups is on the whole not very different from those of the population as a whole. However, in the top socio-economic class there is a high proportion of Chinese scientific workers compared with white (72.0% compared with 39.4%) and a significant dearth of Black or Black British workers, at only 29.3%. However, the absolute numbers of ethnic Chinese is small. In the case of disability, the scientific workforce has broadly the same proportion registering a disability as in the overall workforce, although slightly more can be found in SECs 2,3 and 4 and they are slightly under-represented in SECs 1, 5,6,7 and 8.
When it comes to gender, there is a slight majority (50.3%) of women in the scientific workforce, although this majority is lost and the percentage drops to only 39.6% when the Health sector is excluded. However, only 6.3% of women are in the top socio-economic class as opposed to 23.6% of men. In academia, the percentage of professors who are female varies widely between a high of 58.2% in Nursing and Paramedical Studies to a low of only 5.0% in Mechanical Aero and Production Engineering. These disciplinary variations are not surprising, but the magnitude of the disparity may come as a surprise.
Few of the findings are unexpected, but the importance of the report – which forms part of a larger BIS-funded project about diversity in the workforce – resides in it establishing a benchmark on which future studies can build. Furthermore by highlighting the problems of definitions and datasets, it is to be hoped that future work in this area will be made easier. Anecdote may suggest that science is too often a white, male middle-class occupation but we need to know if this perception is accurate. Only if we have all the facts will it be possible to design interventions to make science equally accessible to all.
Further details and data can be found on the Royal Society’s website where there is an interactive chart providing additional information.