Using stop words in text mining: Immigration and the election campaigns.

In order to understand immigration sentiment and its relationship to other concepts in the Italian general election campaign of 2018 and the European election campaign of 2019, we collected in two corpora all the tweets in the Italian language containing the word “immigration” in the period preceding the vote. Both corpora underwent a sentiment analysis and a stop word analysis using two textual software packages: Linguistic Inquiry and Word Count (LIWC) (Pennebaker et al., 2015) and WORDij. LIWC was originally designed by James Pennebaker to understand how some patients recover from traumatic experiences by writing about those experiences and the emotions associated with it then and afterwards. LIWC consists of a dictionary of words which assesses the percent that they occur in a given text. LIWC analysis provides a measure of positive and negative emotion in the immigration text over time. WORDij is a text analysis program that can compute a Z-Score or the relative proportional test of difference between words and words pairs in two sets of texts. Using an include list of stop words we can determine how these relational words change over time with an emotional valence and Z-score to assess the immigration political debate over time. The paper represents a focus on stop-words, which have been an aspect of textual analysis that is often dismissed yet can be very important to our understanding of relational power