UK Hansard 1935-2014

UK Hansard 1935-2014

The "uk_hansard_1935_2014_BvW_2022.tsv" is a metadata enriched version of the Hansard corpus. It is one of the outputs from Betto van Waarden's Marie Skłodowska-Curie project "Presenting Parliament: Parliamentarians’ visions of the communication and role of parliament within the mediated democracies of Britain, Belgium and the Netherlands, 1844-1995 ".

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 897761.

The project has been executed in collaboration with Mathias Johansson from the DigitalHistory@Lund platform, Lund University

Base corpora

The attached file is an agglomeration of two preexisting versions of the UK Hansard corpus: Political Mashup, provided to us by Kaspar Beelen and a .tsv file we received from Ludovic Rheault which itself is based on the Political Mashup corpus.

This corpus is based on the Rheault version and is reproduced with his written permission. The original corpus contained the following columns:

• cabinet
• function
• parliament
• party
• party_in_power
• speaker_id
• speech_id
• speech_text
• topic
• year

To which we have added the following columns:

• date
• times_in_house
• seniority
• district
• district_class

Changed the 'year' column to 'date' as it contained the ISO8061-formatted date, leaving only the year in the 'year' column

Career

In the Political Mashup data there is information about which offices each speaker has held and using this data we have calculated how many terms a speaker has held office at the time of a speech times_in_house. From this we derived a column we call seniority: speakers that have been in office at most one time before the current parliament/round is classified as a junior - everyone else as a senior.

Districts

By cross referencing the speaker_ids against the Political Mashup we have extracted which district each speaker was representing. We have then mapped these districts against a classification of UK districts (Baker, 2018, accessed on 2021-07-12) that uses six classes which we reduced to three classes for simplicity's sake:

Mapping districts to one of the three classes was mostly done automatically by matching district names against the list, and districts that have split or merged over time were processed manually. Still, not all districts were resolved satisfactorily leaving 43,541 speeches without a district classification resulting in a coverage of 98.7%.