Education-line Home Page

Invisible witnesses? How scientists, technologists, engineers and mathematicians are represented on UK television

Elizabeth Whitelegg*, Richard Holliman, Joachim Allgaier, Eileen Scanlon and Barbara Hodgson

Centre for Research in Education and Educational Technology, The Open University, UK 

Paper presented at the British Educational Research Association Annual Conference, University of Warwick, 6-9 September 2006


It is widely believed that sex-role stereotypes of scientists, engineers, technologists and mathematicians (STEM) have the potential to influence viewers’ perceptions of these subjects, and therefore their future course and career choices. Less, however, is known about current portrayals of STEM on UK television, and their influence on children and young people. This paper reports the initial findings from the Invisible Witnesses? Project. Commissioned by the UK Resources Centre for Women in Science, Engineering and Technology this project is investigating gendered representations of STEM on UK television, and the impact of these portrayals on young people's perceptions of STEM. In this paper we provide an outline of our data collection methods, and the quantitative and qualitative methods we have employed to analyse television output for two one-week samples. In presenting some early findings from our analyses we provide a quantitative overview of the first one-week sample of STEM output. We also introduce illustrative quantitative and qualitative findings from one of the genre - news and current affairs – in terms of the gender distribution, roles of the speakers, and emerging themes. The implications of these early findings are briefly discussed alongside suggestions for further work.

The authors of this paper gratefully acknowledge funding for this project from UK Resources Centre for Women in SET and the European Social Fund under the EQUAL Community Initiative programme.


A recent research review into the participation of girls in physics in the classroom (Murphy and Whitelegg, 2006) synthesised the evidence from over 15 years’ worth of research into the ‘girls’ and physics problem’, as it has come to be known, and described the multi-faceted nature of the issue for girls’ participation in physics, and within science more generally. The review revealed that one of the key factors determining students’ attitudes to physics is concerned with students’ self-concept – how students see themselves in relation to the subject both now and in the future. The development of self-concept is closely linked to development of identity. The project described in this paper looks beyond the school environment to investigate another sphere of influence on the development of children’s and young peoples’ identities, that of the mass media, specifically television. The Invisible Witnesses? Project(1) (Holliman, Whitelegg et al., 2006) examines images of female and male scientists, technologists, engineers and mathematicians on television with a view to informing further work on the influence of these images on children and young peoples’ evolving identities and the development of their attitudes to science, technology, engineering and mathematics (STEM).


Exposure to sex-role stereotypes of scientists (and technologists, engineers and mathematicians) in media portrayals can be one powerful factor that influences the formation of gendered identities (Long et al., 2001). The development of an identity as a scientist, technologist, engineer or mathematician affects educational subject choices which can subsequently open up career opportunities (whilst potentially closing down others) that can lead to enhanced life chances and improved salary prospects. Girls’ lower participation in STEM subjects, particularly physics, beginning at school and transmitted down the educational and employment pipeline, is well documented (Rees, 2001; Murphy and Whitelegg, 2006). So girls are potentially loosing out on pathways to better paid career opportunities by not choosing to study STEM at school. It follows that the UK economy may also be disadvantaged in the future because employers are recruiting from a smaller pool of (mainly male) candidates from which to select the nations’ future scientists, technologists, engineers and mathematicians. The STEM workforce is therefore likely to remain unbalanced in terms of its overall gender distribution, in favour of males, and so be lacking in diversity.

The results from previous ‘Draw-a-scientist’ tests (e.g. Chambers, 1983) have shown that the dominant stereotypical view of a scientist held by young children is that of an elderly white male with unruly white hair, wearing a white lab coat and glasses; the archetypal image of Albert Einstein. However, there are some more recent indications that new images of women scientists are becoming available, in some feature films at least, as suggested below:

The woman scientist tends to differ greatly from her male colleagues in her outer appearance: she is remarkably beautiful and compared with her qualifications unbelievably young. She had a model's body - thin athletic, perfect, is dressed provocatively and is sometimes "distorted" by wearing glasses. (Flicker, 2003 p. 316)

In a more recent implementation of the ‘Draw-a-scientist’ test (Frayling, 2005; Frayling, 2006), approximately half the girls tested drew female scientists (this had increased from 1.4% in Chambers’ tests conducted over 20 years ago). According to Frayling:

This may relate to cartoons produced in America and Japan featuring a feisty action-girl as a member of the intrepid team. Or to Hollywood films such as ‘Contact’, with female scientist-heroes. In the Nineties, quite a few young and beautiful scientists appeared on the big screen. Unlike their male counterparts, they terrific hair which they had to shake out; they were usually in their twenties (but already a "leader in her field"); and if they wore glasses, they removed them for much of the film. (Frayling, 2006, p.27)

Of course, there is now a global marketplace for broadcast output, not least in terms of feature films, cartoons and animations, and comedy. It is therefore not surprising that our research investigating gendered representations of STEM on UK television has uncovered several instances that support the phenomenon that Frayling describes.

So new sorts of gendered images of scientists, technologists, engineers and mathematicians are becoming available to children and young people via mass media, alongside evidence to suggest that perceptions of STEM may also be changing. But how often are these new portrayals available compared to the well established stereotypical images described above, and do the new portrayals present authentic role models that are useful for girls (and boys)? Do they represent another stereotypical image that fails to adequately represent what it means to be an authentic scientist, technologist, engineer or mathematician in the 21st Century? Will they have a positive effect on young people’s self-concept, particularly in terms of gender, potentially increasing recruitment to STEM courses and careers?

There is some evidence from the recent ‘Women and Work Report’ that images of women scientists, shown on television, can have positive effects on course and career choices in certain circumstances:

Girls at Quinitin Kynaston School get ideas and information from television and films. For example, one girl became interested in neurology after watching operations on the Discovery Channel; another was inspired by female TV news journalists. The girls feel that visible female role models are key to inspiring non-traditional career choices.

Two-thirds of the approximately 3,000 students on forensic science courses in the UK are female, making it one of the most science-based degrees for women. Research by the Sector-Skills Council for Science, Engineering and Manufacturing (Semta, 2005) largely attributes the increase in female students to the portrayal of a female forensic pathologist by Amanda Burton in the BBC TV drama Silent Witness. (Women and Work Commission, 2006 p. 26)

Of course, this evidence assumes a simplistic "cause and effect" relationship between television viewing of gendered role models and course/career choice. To be more confident about the influence of gendered portrayals of STEM on UK television requires a more systematic approach.

The aim of the project reported here is to investigate how STEM is represented on the five terrestrial television channels in the UK broadcasting using analogue (and digital) signals, and to analyse representations of scientists, technologists, engineers and mathematicians from a gender perspective. This is Phase 1 of the project. In so doing we will analyse a range of genres in two one-week samples, focusing on programmes for children and young people and using content analysis. This is defined as:

[...] an accepted method of textual investigation, particularly in the field of mass communications. It involves establishing categories and then counting the number of instances when those categories are used in a particular item of text, for instance in a newspaper report. (Silverman, 1993 p. 59)

In phase 2 of the project we will investigate how these portrayals influence young viewers’ perceptions of STEM, and whether they have the potential to influence future course and career choice. Embedded in this phase of the project is the concept of agency. From the perspective of the active mind, what the viewer already knows about STEM, or rather the culture and processes of these practices and of the practitioners will be influenced by what happens within and outside the classroom. These informal and formal learning experiences are likely to influence what the viewer takes notice of in the programme (or lesson) and how that influences the viewer (learner). Whether they accept or reject the image of STEM presented in the programme (or lesson) will be affected by their pre-existing feelings, attitudes, experiences and understanding of STEM. This indicates that mass media influences, particularly those viewed at a very young age can have an important role in shaping a girls’ and boys’ identity, their understanding of STEM, and other factors that influence their engagement with it.

This paper discusses initial findings from Phase 1 - Sample 1 of the project.


We recorded a selection of programmes from the five terrestrial television channels broadcasting using analogue signals (i.e. BBC1, BBC2, ITV1, Channel 4 and Channel 5) in two ‘snapshot’ weeks. Table 1 shows the audience share for these channels for the two one-week sample periods.

Table 1: Audience share for the five channels in the two week-long samples. (Broadcasters’ Audience Research Board (BARB) Ltd)(2)


Audience Share (%)

Sample 1
Week ending 16 October 2005

Audience Share (%)

Sample 2
Week ending 19 March 2006










Channel 4



Channel 5



Sub total






The two samples ran for seven days each and were six months apart - from 6 to 12 October 2005, inclusive and 10 to 19 March 2006, inclusive. The two samples include examples of breakfast, daytime, children’s, primetime and late evening television and of factual and fictional programming, including: news reporting, documentaries, schools programmes, cartoons, situation comedies, game shows and dramas.

In Sample 1 (week ending 16 October 2005) 302 programmes were recorded, resulting in just over 162 hours of recorded television. In sample 2 (week ending 12 March 2006)(4) 364 programmes were recorded, resulting on 211 hours and 50 minutes of recorded television. In total we recorded 666 programmes and a total of 373 hours 55 minutes of programming over the two samples; about 22% of the available programmes (Table 2)(5). (We note that the BBC shifted its children’s programmes – broadcast within CBeebies and CBBC - from BBC1 to BBC2 between the two sample periods. This accounts for the differences between these channels in Sample 1 and 2.)

Table 2: Programmes recorded from the five channels in the two week-long samples.




Channel 4

Channel 5


Sample 1
Number of programmes







Sample 1
Duration (hours/minutes)(6)







Sample 2
Number of programmes







Sample 2
Duration hours/minutes







To inform the pre-selection process we investigated the television listings for programmes that might include STEM content, e.g. the BBC2 science documentary programme Horizon. We also recorded programmes where descriptions were inconclusive but where STEM content might be reasonably expected to be broadcast, e.g. news and current affairs programmes and game shows. Finally, we recorded programmes where the television listings did not provide detailed information and where the research team was unsure of the specific content, e.g. children’s programmes. Overall, therefore, our sample should be considered representative of the types of STEM programmes broadcast using analogue signals in the UK during the sample periods.

Once collected, the programmes in the two week-long samples were analysed to confirm whether they included STEM content, with a view to conducting further detailed quantitative and qualitative analysis. To achieve this further selection we employed the following operational definitions of science, engineering and technology. These definitions were adapted from earlier work (The ENSCOT Team; Holliman, Trench et al., 2002).


These programmes should include significant explicit scientific content, namely a reference or references to scientific findings, scientific research, scientific procedure, science as an intellectual activity, or scientists.


These programmes should include significant explicit technology content, namely a reference or references to technological design, technology research, technological procedures, technology as an intellectual activity, or technologists.


These programmes should include significant explicit engineering content, namely a reference or references to engineering design, engineering research, engineering procedures, engineering as an intellectual activity, or engineers.


These programmes should include significant explicit mathematical content, namely a reference or references to mathematical concepts and formula, mathematics research, mathematics as an intellectual activity, or mathematicians.

Following an initial period where these operational definitions were discussed by the project team, the same researcher viewed each of the programmes recorded for the two samples, selecting those with STEM content, and entering the details of these programmes into a database, recording the: channel, name of the programme, when it was broadcast, whether it portrayed science, technology, engineering or mathematics, and the genre (e.g. whether it was science fiction, comedy, docudrama, etc.). The STEM extracts were also re-recorded for further quantitative and qualitative analyses.

In conducting our initial analyses we have used a flexible unit of analysis, which we have called an extract. We define an extract as ranging from complete a programme where this was deemed to contain STEM content its entirety through to discreet news items within a bulletin. Adopting this flexible unit of analysis allowed us to code individual STEM extracts within the same news bulletin, but also to record an entire programme. Further to this, the same definitions of STEM were applied to factual and fictional representations, and to adults, children and non-humans. Hence, the same operationalised definitions were applied to cartoon characters, children's characters and scientists portrayed in factual programmes, such as news and current affairs.

We note that there are challenges in deciding what counts as STEM, e.g. when programmes examine areas such as biomedicine, architecture, and/or information and communications technology. This was also the case when a programme was aimed at pre-school children, where defining what counted as STEM was often harder to judge because STEM professionals were largely absent, and the techniques and/or concepts being portrayed were more simplistic. When the researcher was unsure about whether to select a programme, or programme extract as STEM, this was viewed and discussed with the project team so that a consistent application of the operational definitions could be maintained.

Once the initial selection of extracts that portrayed STEM was completed, we conducted a quantitative analysis of the overall sample. We used these data to produce a further sub-set of STEM content for further analysis, focussing on extracts broadcast for children and young people, e.g. programmes broadcast as part of CBeebies, CBBC, CITV, Channel 4 Learning, or Channel 5’s Milkshake!

To inform the analysis of televisual content for this sub-set of extracts, fully annotated transcripts (Hansen, Cottle, et al., 1998) have been produced, coding all recorded speech in terms of gender and the role of the actor, then quantifying these data. Further to this, we have produced narrative summaries of these extracts, noting emerging themes. These themes will be explored in more detail in the later stages of the project.


Of the 302 programmes recorded for Sample 1, 156 extracts were coded as containing some STEM content; a further 57 were coded as children and young people (Table 3).

Table 3: Breakdown of STEM content from sample 1 (n=156), and for the sub-set of extracts: children and young people (n=57)


Sample 1


Children & young people














Table 3 also shows the breakdown of programmes coded as science, technology, engineering and mathematics. Science clearly dominates this distribution in the overall sample (73%) and the sub-set for children and young people (61%), respectively. The difference is partly explained by the larger proportion of engineering extracts in the sub-set, which was the result of Channel 4’s ‘Engineering at the Cutting Edge’. Five programmes in this series, which is linked to a GCSE, were broadcast during the sample week.

Further analysis of the annotated transcripts for the seven news and current affairs extracts for children and young people illustrates the following gender distribution for the speaking actors (Table 4).

Table 4: Gender distribution of speaking actors in news & current affairs (based on seven extracts)




Newsround (6 extracts)



Blue Peter (1 extract)



All news & current affairs



Table 4 illustrates that overall the gender distribution in the seven news and current affairs extracts was skewed in favour of male speakers. This distribution is particularly apparent in the six Newsround extracts, despite the fact that this show has a regular team of young, slim and attractive female and male presenters and reporters. In contrast, the distribution in the Blue Peter extract, which is almost as long as the Newsround extracts put together, is skewed in favour of the female presenters. (It is worth noting that Blue Peter also has a regular team of four young, slim and attractive presenters; two female and two male.)

Further analysis shows that the speaking roles in these extracts are almost exclusively newsreaders and reporters in the case of Newsround, and presenters in the case of Blue Peter. Only one (male) scientist appears in one of the Newsround (6 October 2005) extracts. Contributing just over 1% of the words spoken in the seven extracts, Professor Huosheng Hu briefly discusses his invention - a robot fish - and its potential for use as a piece of military technology.

The two children who are interviewed in the same extract state that:

Male - child

I think that this is very hard to make, because you have to need all the circuits and the wires and the batteries together to make a circuit, and because they’re not like normal fish.

Female - child

I think these fish are prettier than normal fish because they have nice scales, and they are shiny, and their eyes are big.

Notably, the male child discusses how the fish have been made, whilst the female child talks about how attractive the fish look. As such, they could be said to be (re)constructing gendered patterns of behaviour.


The results document considerably more STEM being shown on television in the first sample week than we expected. We therefore produced a sub-set of programmes for children and young people to subject to further systematic quantitative and qualitative analysis by systematically producing and analysing annotated transcripts for these 57 extracts in terms of all the speaking actors and their roles, and emerging themes.

We found a particular science emphasis in the Sample 1 STEM extracts (when compared to technology, engineering and mathematics), but with differences in the distribution of the science sub-categories, with much less biomedical material in the sub-set for children and young people. Early indications also suggest that there are few scientists, technologists, engineers and mathematicians portrayed in the news and current affairs extracts, also those for pre-school children.

We have presented findings that show that overall the gender distribution in the news and current affairs extracts is skewed in favour of male speakers, whilst both regularly feature young, slim, attractive presenters. Further analysis of the extracts suggests a more complicated pattern, however, with male newsreaders and reporters dominating Newsround, while female presenters dominate the Blue Peter extract. As it stands these results are inconclusive. We will compare these findings with those derived from the second sample to consider the implications for this genre.

The illustrative example given in this paper from Newsround is included both to demonstrate our data collection and analytical methods, and to demonstrate evidence of gendered patterns of behaviour, in this case in terms of the child interviewees. However, we note that this extract is atypical of the seven extracts, not least because it is the only one that features a (male) professorial scientist and/or interviewees. In the remaining 6 extracts, male and female newsreaders, reporters and presenters discuss STEM.

In conclusion, we have found evidence of gendered representations within STEM extracts in Sample 1. There is an emphasis on young, slim, attractive presenters in news and current affairs programming for children and young people with some indicative evidence to suggest that male speakers may dominate this genre when discussing STEM. In the next stage of our project we will complete our analysis of Sample 1 and undertake an equivalent systematic analysis of the images of STEM in Sample 2, to look for evidence of consistent patterns. Together these analyses will provide the basis for an investigation into whether and how these images influence girls’ subject choices at school, with a view to informing ongoing debates about the influence of popular culture on girls’ identities and subsequently on further and higher education choices and uptake of careers in STEM by girls (and boys) and women (and men).


1. The Invisible Witnesses? Project is funded by the UK Resource Centre for Women in SET – see; last accessed 2 April 2007. For a brief outline of the project, see; last accessed 2 April 2007.

2. These figures include those viewers who chose to watch these channels through satellite, cable and freeview. This is because BBC1, BBC2, ITV1, Channel 4 and Channel 5 were broadcast using both analogue and digital signals during the sample periods.

3.  The category “others” includes channels that are only available to viewers through satellite, cable and freeview.

4.  This was National Science, Engineering and Technology (SET) Week.

 5. 11 programmes - six broadcast for young children and five evening news bulletins – were not included in the final sample, either because the video recording equipment failed, or because previous scheduled programmes overran. However, we note that none of these programmes were described in the television listings as specifically including STEM content.

6.  Rounded up to nearest 5 minutes.


Broadcasters’ Audience Research Board (BARB) Ltd. Retrieved 2 April 2007, from

Chambers, D. W. (1983). "Stereotypic images of the Scientist: the Draw a Scientist Test." Science Education 67: 255-265.

Flicker, E. (2003). "Between brain and breasts-women scientists in fiction film: on the marginalization and sexualization of scientific competence." Public Understanding of Science 12: 307-318.

Frayling, C. (2005). Mad, Bad and Dangerous? The Scientist and the Cinema. London, Reaktion Books.

Frayling, C. (2006, May 9). "All boffins and bonkers: Scientists need Indiana Jones image to make their subjects more popular in the classroom." Daily Telegraph, page 27.

Hansen, A., Cottle, S., Negrine, R. and Newbold, C. (1998). Mass communication research methods, MacMillan, Basingstoke.

Holliman, R., Trench, B. et al. (2002). Science in the news: a cross-cultural study of newspapers in five European countries. Proceedings of the 7th International PCST Conference Science Communication in a Diverse World, Cape Town, South Africa.

Holliman, R., Whitelegg, E. et al. (2006). Invisible witnesses? - Representations of women scientists, engineers and technologists on UK terrestrial television. Proceedings of the 9th International Public Communication of Science and Technology Conference - Scientific culture for global citizenship, Seoul, South Korea, available online at:; last accessed 2 April 2007.

Long, M., et al. (2001). "Gender and racial counter-stereotypes in science education television: a content analysis." Public Understanding of Science 10: 255-269.

Murphy, P. and E. Whitelegg (2006). Girls in the Physics Classroom: A Review of Research on the Participation of Girls in Physics. London, Institute of Physics: 1-61. (Available at

Ormerod, M. B. and D. Duckworth (1975). Pupils' Attitudes to Science: A Review of Research. Windsor, Berks, National Federation for Educational Research.

Rees, T. (2001). "Mainstreaming gender equality in science in the European Union: The 'ETAN Report'." Gender and Education 13(3): 243.

Semta (2005). Forensic science: Implications for Higher Education 2004, Sector Skills Council for Science, Engineering and Manufacturing.

Silverman, D. (1993). Interpreting Qualitative data: Methods for analysing talk, text and interaction. London, Sage.

The ENSCOT Team (2001). "ENSCOT: The European Network of Science Communication Teachers." Public Understanding of Science 12(2): 167-181.

Women and Work Commission (2006). Shaping a fairer future. London, Department of Trade and Industry.

This document was added to the Education-Line database on 08 June 2007