It does not rain more in the UK : A corpus analysis comparing the optimism variable in Swedish and UK-based listed companies’ sustainability reports

Abstract: Abstract This thesis sets out to compare the level of optimism, quantified as the optimism variable, in UK and Swedish-based corporate Environmental, social, and corporate governance (“ESG”) disclosures published in English. The comparison is conducted by creating two corpora, one UK-based and one Swedish-based. The corpora are made up by 21 ESG disclosures per country, all collected from financial service companies and published between 2019 and 2022. The two corpora are then semantically analysed, using the text-mining software DICTION. DICTION compares the language and lexemes used in each ESG disclosure against a dictionary with pre-assigned values per word and strings of words, creating an optimism score for each ESG disclosure. These scorings are aggregated per country, creating a mean and a standard deviation profile for the UK-based and the Swedish-based companies, respectively.  By applying a T-test, comparing the means of the two populations regarding the optimism variable it can be concluded with 95 % confidence that there is no difference. It thus seems that the Swedish-based and the UK-based are similar in their level of language optimism.    Keywords Corpus analysis, semantics and Computer assisted discourse analysis.