Experience

Undertaken with Opcit Research for the European Parliament. This research sought to investigate the effect of various electoral systems used in the European Parliament on the probability of women being elected to office. An original dataset was constructed using documentation from the European elections, which was then used to estimate multilevel regression models. These allowed the association of various characteristics of the voting systems and candidates on the probability of being elected to be estimated. It was found that the probability of being elected was highly dependent on the voting system used. Quotas did, however, have a positive impact on women’s chances of being elected. Research report (pdf)

This research involved exploratory, quantitative research into the association between level of secure income and a range of retirement outcomes, for retirees in England with some form of private pension income. The research was undertaken in the wake of the April 2015 changes to rules on ‘Defined Contribution’ (DC) pension savings, which broke with the previous regime – known as the ‘annuities deal’ – which required DC pension savers to convert their savings into a secure income at retirement. The study analysed data from Wave 6 (2012-13) of the English Longitudinal Study of Ageing (ELSA), which is a longitudinal, multidisciplinary social survey undertaken every two years, of a representative sample of the English population aged 50 and older. Summary (pdf), Research report (pdf), Policy report (pdf), Model estimates (xlsx), Launch event presentation (pdf)

The CCG is facing real terms budget cuts in the coming years, whilst the demands on health services increase due to population aging. This research involved segmenting the users of the CCG’s health services using latent class analysis, in order to identify types of older people using the health services, and the demands they placed on the system. Secondary Uses Service data consisting of finished consultant episodes, admitted patient care, outpatient appointments, and A&E admissions were analysed. The potential impact of using health services to extend the proportion of older people’s lifespan lived in good health was also modelled using high-order polynomial ordinary least squares regression.

A review of multilevel modelling studies was undertaken as one part of this wide-ranging systematic review of the evidence for health effects of schools and school environments. The focus of the multilevel review was to identify school-level factors that influence student health. Sophisticated searches were undertaken of academic databases, the public sector and grey literature, and noted academics in the field were contacted to aid study identification. Multilevel regression studies were identified and a thorough data extraction procedure was undertaken. The findings were synthesised and presented in detail. Research report (pdf), Journal article (pdf)

This was a large and wide-ranging piece of research for the DCMS Culture and Sport Evidence programme, investigating (a) the drivers of participation in culture and sport, (b) the impact of engagement in culture and sport on young people’s learning, and (c) the value to the economy of engagement in culture and sport. I undertook (a), the part of the research which identified the drivers of participation in culture and sport. A thorough descriptive analysis was undertaken of the factors which might influence engagement. Multilevel logistic regression models were estimated using data from the Taking Part survey, which collects data on many aspects of leisure, culture and sport in England, as well as an in-depth range of socio-demographic information on respondents. Recommendations were made to the DCMS. This research was the foundation for my PhD research.
Over-arching summary (pdf), Summary report (pdf), Technical report (pdf), Technical appendices (pdf)