YouTube communication of Italian political parties

If we analysed the verbal communication in videos published by official YouTube accounts for the major Italian political parties, what differences would we find? I used the APIs YouTube Data API v3 to find out.

🔹 Hypotheses H1: Parties with common ideologies also show common keywords, emotional words and syntax. 🔹Data I created my datasets using the YouTube Data API v3, along with the R Packages Tuber and youtubecaptions. The Tuber package allowed me to manage the API directly from R: the function list_channel_videos provided me with metadata for each video published on the YouTube channels, but I mostly relied on the metadata containing publishing date and video ID. The youtubecaptions package allowed me to obtain captions for each video using the URLs, which I produced using the video IDs previously scraped. 🔹Methods For each party's channel, I tokenized the captions to obtain overall word counts. I then calculated in-group relative frequencies to compare word choices between parties. 🔹Expected results I expect allied parties to share common keywords and emotional words, as they may share values and ideologies. I also expect in-group differences over time, as some parties have had drastic changes over time (to name one example, Lega's communication after Lega Nord has become Lega - the YouTube account hasn't been changed, just renamed).