This figure plots the communities found among survey answers using the “walktrap” algorithm. Correlation coefficients are used as distance for the clustering. Positive correlations are shown in green and negative correlations in red. Plotted using R/igraph.

This is a plot of (almost) all answers to the survey as a heatmap. The columns are the questions and the rows respondents, but the image is separated in three vertical sections corresponding to three subsets of respondents. All respondents who gave at least 25 non-“other” answers are included (1728). The answers are reduced to numerical values using the procedure described in the paper, then plotted as colors from dark blue to bright green. Respondents are clustered based on Euclidean distance.

What is it supposed to show? It gives us a sense of how messy philosophy is and how diverse philosophers' views are because we don't see large clusters or patterns emerge despite our best efforts to group similar respondents together.