Appendix 2 - Statistical

Summary

The statistical appendix referenced throughout this book uses an intuitive numbering system. Note that, in order to remove the need for time-consuming cross-referencing, data is occasionally duplicated.

Key

The acronym ‘EA’ refers to the East Anglian corpus; ‘NAT’ to the National Speakers corpus; ‘CONSTIT’ to the Constituencies corpus. ‘CON’, ‘LIB’, ‘LU’ and ‘LAB’ are also used for parties.

Notes

• 1. All scores provided in the figures (unless specifically indicated) are for lemmas, not just the words shown. See p. iv above.

• 2. All figures shown are based on corpora of exactly 50,000 words per party, per election. All readings are thus directly comparable with each other throughout. When both parties’ scores are aggregated, the weighting is 100,000 words.

• 3. Figures often include supplementary columns such as ‘Mean 1880–1900’ or ‘Mean 1880–1910’. These are simply averaged scores across all the elections in these years, and are included to provide comparative context.

• 4. Total scores for all the words on a figure are included at the foot of each figure. These are not necessarily the most important readings, but are often illustrative of the wider picture.

• 5. Whole numbers are used throughout, with decimals rounded up or down.