Facebook pages for parties in 2019 EU parliamentary elections

This dataset contains information about the parties from UK, Italy, Germany, Spain, and Poland that ran ads on Facebook during the 2019 EU parliamentary elections. For each party, we manually associated the corresponding Facebook page that ran the ads. We also report the PopuList tags as per https://popu-list.org.

"2019 EU elections parties.csv" contains aggregate data:

State: the country of the party

Party: acronym of the party

Name: extended name of the party

Percentage: share of votes in the election

Facebook page_name: name of the Facebook page that ran the ads

Facebook page_id: id of the Facebook page that ran the ads

populist: binary label indicating whether the party is populist according to PopuList

farright: binary label indicating whether the party is far right according to PopuList

farleft: binary label indicating whether the party is far left according to PopuList

eurosceptic: binary label indicating whether the party is eurosceptic according to PopuList

found on populist: binary label indicating whether the party is on any list on PopuList

seat: whether the party got a seat in any previous national election (selection criterion from PopuList)

"FB ads impressions cost populist demographic votes" contains detailed data:

ad_id: Facebook ad id

page_id: id of the Facebook page who published the ad

page_name: name of the Facebook page who published the ad

ad_creative_body: text of the ad

ad_creative_*: other properties of the ad (caption, links, title)

ad_delivery_start_time,ad_delivery_stop_time: time of beginning and end of the ad campaign

funding_entity: declared party who financed the ad

impressions_min, impressions_max: range of impressions for the ad

spend_min, spend_max: range of expenditure for the ad

ad_url: URL of the ad

state: state of the party

impression_mean, spend_mean: average number of impressions and expenditure

imp_per_day: number of impressions per day

populist, farright, farleft, eurosceptic, popu-list: binary tags from PopuList (refer to the description above)

female_*, male_*, unknown_*: fraction of impressions for each demographic bucket (defined as gender_age)

party: name of the party

percentage: share of votes in the election