Election study data in Britain has enabled scholars to analyse gendered associations in political attitudes and behaviour in Britain over the decades, in particular in relation to voting preferences. It however has also come with a number of limitations. From the first study in 1963 throughout much of the rest of the 20th century, data was mainly collected during election years. This prevented researchers from establishing the extent to which gendered patterns remained the same or altered in between. Moreover, without boosted sample sizes, there has not been sufficient power to carry out important sub-group analyse such as differences between working women and those primarily engaged in home duties.

In this paper, we provide comprehensive individual-level analyses of changing patterns in gendered political attitudes in Britain. This is done through using hundreds of historical Gallup polls from 1955 to 1991 that have just been made available to researchers for the first time, and merging these with contemporary sources up to the present time. We make a number of important advances. As our data begins in 1955, our analyses go back eight years further than available election study data. With at least one survey per month for most of this period, we also have sufficient sample sizes to investigate important relationships among subsamples and to examine between-election patterns that have to-date not been possible. Together with our range of political variables - including vote intentions, government approval, and leader evaluations – this allows us to delve deeper than ever before.