Gender and Voting Behaviour at the United Kingdom's 2016 Referendum on EU Membership

UoM administered thesis: Phd


This thesis examines the role of gender in voting behaviour at the UK’s 2016 Referendum on EU Membership. Gender was important in shaping vote choices at the referendum despite, as this thesis shows, the lack of an aggregate-level gender gap in vote choice or substantial differences in men’s and women’s preferences and priorities on the key issues of the referendum debate. I argue that the processes of social change which produced divisions between young and old at the referendum are gendered and derive a theoretical expectation that referendum vote choice should differ by gender in some age groups. I find partial support for this expectation: there was a gender gap in vote choice at the referendum, but only among 18-25 year olds. I find that this gender gap is associated with differences in socioeconomics, values, and attitudes to the EU. However, men and women were remarkably similar in their vote choices despite
the socioeconomic divisions between them, particularly when such divisions are considered by age. The second half of the thesis is thus devoted to explaining this surprising similarity between men and women by examining social and geographic contextual effects at the referendum. I find that context brings men
and women together. For instance, women or men with a partner who voted Leave, or who previously supported UKIP, are found to be more likely to have voted Leave than single individuals. This study of context also shows some small but interesting differences in the contextual factors which affected men’s and women’s vote choices at the referendum. Women are more likely to discuss politics with their partners and family, and this type of political discussant has the strongest effects on vote choice. Thus, women may have been more influenced by social ties and networks than men when it came to casting their referendum vote. As younger women and men are more likely to be single, I suggest that this lack of a partner’s political influence is a possible reason why gender gaps appear in the youngest age group only. I also find that women’s voting behaviour at the referendum is associated with the level of change in immigration into their local area, with women who live in areas which experienced no or little change to the level of immigration being less likely to have voted Leave, but that this effect is not significant for men. Thus, to understand why women voted to Leave the EU, we must consider their context as well as their socioeconomic attributes. This thesis raises the importance of considering how gender affects voting behaviour even when there is no obvious gender gap in vote choice. It shows not just how gender interacts with a person’s other individual characteristics such as their age or generation, but how their gender interacts with their social context. Geographic and social contextual effects likely reduce differences between women and men in their political preferences, despite large gender differences in socioeconomic position, characteristics, and experiences. This should be integrated into future research into gender and voting behaviour.


Original languageEnglish
Awarding Institution
Award date19 Apr 2022