Fig. 2
From: Semi-supervised graph labelling reveals increasing partisanship in the United States Congress
Top: F1 Score for the House from 1935-37 (74th House) to 2017-19 (115th House). Bottom: Difference between the smallest standardised P(Yi=Democrat) among all true Democrats and the largest standardised P(Yi=Democrat) among all true Republicans for the House from 1935-37 (74th House) to 2017-19 (115th House). While there is some variability over time, F1 scores are relatively high for all Houses (minimum F1 score = 0.8776), indicating very strong performance of the GLaSS method in labelling members of the House as Democrat or Republican. In particular, every House from the 108th (2003-05) onwards has an F1 score of 1, implying that the GLaSS method was able to perfectly identify the party affiliation of every member in those Houses. The plot of standardised differences shows the magnitude of overlap (values below the horizontal line at 0) or separation (values above the horizontal line at 0) between Democrats and Republicans, according to absorption probabilities calculated by the GLaSS method. F1 score appears to decrease with increasing magnitude of overlap, while also showing that the two parties have grown increasingly far apart since they first separated entirely in the 108th House