This lecture brought together four perspectives on diversity in computing, structured around ethnicity, gender, and socioeconomic disadvantage. The day included contributions from Lynda Chinaka (Roehampton) and Miranda Simmons (City of London School for Girls), an afternoon session from Dr Pete Kemp (King’s College London), and was framed and closed by Prof Miles Berry.

Framing the Day (Miles Berry)

Miles opened by grounding the day in the entitlement principle at the heart of the national curriculum: computing is for every child, regardless of gender, ethnicity, or family income. He noted that the Core Content Framework — the government’s prescribed curriculum for teacher training — makes no explicit mention of ethnicity or gender in computing, though disability does appear. This absence is worth sitting with. He also pointed out that socioeconomic disadvantage is not a protected characteristic in law, meaning the legal context for addressing it differs from that for race or sex discrimination.

Culturally Responsive Computing (Lynda Chinaka)

Lynda drew on her research with the Raspberry Pi Foundation to introduce culturally relevant and responsive pedagogy. The intellectual foundations come from Gloria Ladson-Billings, who repositioned the question from “what’s wrong with these students?” to “how do we build on what learners bring?” The key idea is that teaching should connect with pupils’ cultural identities, histories, and contexts as a genuine pedagogical strategy, not a token gesture.

Lynda shared a framework of ten “areas of opportunity” for teachers, covering: getting to know learners as individuals, recognising teachers’ own biases, reviewing lesson content and context, ensuring accessibility, giving pupils agency, and thinking about school policy. A practical entry point she used was an identity-mapping activity, in which pupils note aspects of their cultural background that can then shape computing tasks. Concrete examples included using Noor Inayat Khan’s story as a route into cryptography with Year 1; a Commonwealth soldiers project connecting history with digital artefact creation; and a podcast unit where children chose their own topic. The principle throughout was preserving the integrity of computing curriculum objectives while changing the context to something meaningful to the learners in the room.

Gender and Computing (Miranda Simmons)

Miranda, teaching at an all-girls school, presented data on the gender gap and its origins. Around 21% of the computing workforce is female — a figure that maps onto GCSE and degree uptake, suggesting the problem begins at school. Girls in single-sex schools are three times more likely to choose GCSE computer science than girls in mixed schools, pointing to environment and confidence rather than ability.

Miranda drew on her own experience to illustrate how unconscious bias operates: being told by well-meaning male colleagues to “just do the PowerPoint” carries a message even when no malice is intended. She described the anxiety that programming generates: code either runs or it doesn’t, which can make partial progress feel like failure. Her practical responses included: showing pupils mark schemes so they see partial credit; using role models who are current and relatable rather than historical; hosting alumni events so girls can map a realistic path from their own school into the industry; and holding high expectations consistently. She also noted that three topics girls tend to find more engaging — digital media, projects, and presentations — were largely removed from examined content when ICT shifted to computer science, and that curriculum reform may offer a chance to address this.

Socioeconomic Disadvantage (Pete Kemp)

Pete presented data from his research using the National Pupil Database on the relationship between computing and socioeconomic disadvantage. The headline finding: since 2018, growth in GCSE computer science uptake has occurred almost entirely in private and grammar schools. Comprehensive schools have seen essentially no increase. Large numbers of pupils — disproportionately girls and those eligible for free school meals — reach 14 and receive no further computing education at all.

Pete also challenged simple assumptions about the digital divide. Poorer children often have more access to smartphones than their wealthier peers at young ages, but less access to laptops and less parental time to support meaningful use of technology at home. The divide is not simply about devices. His conclusion on solutions was blunt: improve curriculum and improve pedagogy. Curriculum change is partly outside individual teachers’ control; pedagogy is not. He pointed to the ongoing curriculum review as a real opportunity to broaden the subject.

Gender Revisited (Miles Berry)

Miles closed by returning to gender with cross-national data showing that countries rated low on gender equity indices — including some North African and Middle Eastern nations — produce far higher proportions of women in STEM degrees than highly equitable Scandinavian countries. Discussion of possible explanations touched on economic independence as a driver, cultural expectations around academic subjects, and the greater freedom women in equitable societies have to pursue non-STEM paths.

He also presented data showing that girls who choose GCSE computer science outperform boys, but when results are adjusted for overall academic performance, computer science is one subject where girls underperform relative to their own average — alongside maths and physics. On action, Miles was direct: sexist comments in computing classrooms should carry the same zero tolerance as racist ones. School culture matters as much as individual lessons. He also noted the influence of mothers on girls’ subject choices, and the damage done when negative parental messages about computing go unchallenged.

Summary for Mentors

The day as a whole argued that diversity in computing is not primarily a matter of individual teacher effort, though teacher choices matter. It is structural: shaped by curriculum design, school culture, parental attitudes, socioeconomic inequality, and accumulated stereotype. The message for trainees, and for the mentors who support them in school, is that addressing these issues requires attention at every level — in the examples and contexts chosen for lessons, in how confidence and failure are handled, in school culture, and in the design of the subject itself.

Based on the 16th Roehampton Computing Education lecture, Diversity in Computing, 25 February 2026