Royal Holloway Social Purpose CDT

Department of Economics

Dr Gabriel Facchini and Professor Arnaud Chevalier

Learning, School Starting Age and Children's Mental Health

Mental ill health among children and young people has risen sharply, and doing badly at school is one of its strongest correlates. What is not known is which way the causation runs. Struggling in the classroom may damage wellbeing; poor mental health certainly damages attainment. The distinction matters for policy: if what children learn genuinely protects their mental health, then teaching quality and school readiness become instruments of public health, not only of education.

The project separates cause from correlation by exploiting changes in the English school system — past reforms to how children are taught, and the rules that decide when they start school. Changes of this kind shift children's learning for reasons that have nothing to do with their mental health, which is what makes it possible to ask what learning itself does. We have particular sources of variation in mind and will discuss them with the student, who will also have scope to propose their own.

Three questions follow. Does better early attainment reduce later mental ill health? Are any effects short-lived or lasting? And who is affected — by deprivation, sex and special educational need?

The student will work with ECHILD (Education and Child Health Insights from Linked Data), which joins the school records of every child in state education in England to their hospital, mental health service and mortality records — millions of children who can be followed from the day they start school into adulthood. Training in secure-data research and the necessary accreditation form part of the first year.

We are looking for a student with a strong quantitative background in economics or a related social science, comfortable with Stata, R or Python, and interested in applied microeconometrics and large administrative datasets. Prior experience with linked data is welcome but not required. Informal enquiries are encouraged.

Business School

Dr Gul Berna Ozcan

AI and SMEs; Political Economy of Business; Migrant & Refugee Entrepreneurship; New Business Models & Precarious Work

AI and SMEs: In what way AI adoption affects organisations and workers; how supply chains operate; how AI shapes managerial power and labour within and beyond the firm.
Political Economy of Business: How businesses engage with political leaders and parties; in what way firms benefit or lose market value due to their political ties; empirical contexts can be geographical or industry-specific through qualitative or quantitative research.
Migrant & Refugee Entrepreneurship: How migrants and refugees create and grow businesses; the role of institutions, networks and resources in shaping entrepreneurial outcomes; implications for integration, inclusion and economic development. This can be analysed within a cross-country or regional context.
New Business Models & Precarious Work: How digital platforms and emerging business models reshape employment relations; the causes and consequences of insecure work; implications for worker welfare, regulation and organisational performance.

Department of Physical Sciences and Engineering

Professor Martin King

The reflectivity of sea ice as a function of its evolution and light absorbing impurities: Radiative forcing for climate change and BRDF for Earth Observation

Sea ice plays a vital role in the Earth system, regulating how much solar energy is reflected back to space and how much is absorbed by the polar environment. However, impurities such as black carbon, volcanic ash and mineral dust can darken sea ice, reducing its reflectivity (albedo) and potentially accelerating polar change. Understanding these processes is important for climate prediction, Earth Observation and monitoring our rapidly changing polar regions.

This interdisciplinary Social Purpose PhD will investigate how the reflectivity of sea ice changes as it evolves and as light-absorbing impurities accumulate. You will explore how impurity concentration, temperature, thickness and the changing structure of sea ice influence the way light is reflected, absorbed and transmitted.

The project combines laboratory experiments, field observations and modelling. You will produce sea ice under controlled conditions in refrigerated tanks, making optical measurements alongside measurements of ice temperature, brine content and microstructure. There may also be opportunities for field observations in Svalbard, studying natural sea ice and snow-covered sea ice through the seasonal transition from winter to summer, including melt ponds.

Your observations will be used to develop and validate radiative-transfer models that quantify changes in albedo, energy absorption and light reaching the ocean beneath the ice. The work will also contribute to improved interpretation of satellite and Earth Observation data.

A key feature of this PhD is its 50/50 academic–industry partnership with Arctic Research and Development (ARD). You will spend significant time working alongside ARD, gaining experience of applying scientific research to real-world Arctic challenges. The partnership provides access to specialist laboratories, engineering expertise, technology development and potentially field deployments in Svalbard, giving you the opportunity to work at the interface of science, engineering and Arctic technology.

You will gain transferable skills in experimental design, optical measurements, fieldwork, programming, data analysis, modelling and interdisciplinary collaboration.

Applicants from Physics, Engineering, Chemistry or Earth Sciences are encouraged to apply. Previous experience in programming, modelling, laboratory work, spectroscopy or fieldwork is advantageous but not essential.

Department of Psychology 

Dr Nicholas Furl

Individual differences in eyewitness face identification: towards more reliable identification procedures

People vary substantially in their ability to recognise and identify unfamiliar faces. This project will investigate the cognitive sources of these individual differences and ask how face-identification procedures can be designed to remain reliable across people with very different abilities.

Using controlled behavioural experiments, the project will examine performance in police-style identity line-ups alongside other unfamiliar-face identification and matching tasks. It will investigate the extent to which performance depends on general cognitive ability, general visual and object-recognition abilities, and abilities specific to face perception and memory. This approach will also allow the project to ask whether apparently distinct face-processing abilities make different contributions to successful identification.

A further focus will be confidence and metacognition. Do people know how good they are at recognising faces? Does confidence provide a reliable indication of accuracy across individuals with different underlying abilities? Understanding these relationships may help establish when confidence is informative and how identification procedures might be designed to minimise predictable errors across a heterogeneous witness population.

The project will combine experimental psychology with psychometrics and latent-variable modelling. Depending on the student’s interests, there will also be optional opportunities to use modern computer-vision or deep-learning models to compare human and machine face representations and investigate which properties of faces and identification displays predict errors.

The broader aim is to improve understanding of the sources of face-identification errors and contribute to the development and interpretation of fairer and more reliable identification procedures. Related proposals in face recognition, eyewitness identification or individual differences would also be welcome.