Menu
I would like to visualize a rank change (i.e. change of the relative order) of US states using a Sankey flow graph. I'm using the networkd3 package and came up with the following:
library(dplyr)
library(networkD3)
df <- data_frame(origins=state.name[1:10], destinations=state.name[1:10])
lab <- c(df$origins, df$destinations)
nodes <- data.frame(node=c(0:9), name=lab)
links <- data.frame(source=c(0:9), target=c(10:19), value=rep(1,10))
sankeyNetwork(Links = links, Nodes = nodes, Source = 'source',
Target = 'target', Value = 'value', NodeID = 'name')
This snippet produces the following graph:
Sankey flow graph of US states
I can change the relative order by hand now. However, I wonder whether it is possible to fix the order on the right-hand side and put e.g. Alabama on rank 3, California on rank 1, etc ...
841
You can simply drag and drop the bars in the desired sorting order.
The key to reading and interpreting Sankey Diagrams is remembering that the width is proportional to the quantity represented. In the example below, the audience quickly sees that largest destination for water is terrestrial evaporation, among other features of the hydrologic cycle.
Summary: As a visualization, Sankey diagrams are generally hard for the user to understand. If you must use them, two principles I recommend are to: 1) Keep it simple - no more than 2-3 nodes, 7-8 flows; 2) Interactivity - if you insist on a complex diagram, it should interactive and used for data exploraiton.
A Sankey chart requires three data columns or rows from your DataSet—one for the "From" column, one for the "To" column, and one for the values corresponding to each pairing.
If you set iterations = 0 in sankeyNetwork(), you will effectively disable the algorithm which automatically determines the node placement (which is the primary purpose of the sankeyNetwork() function), and the nodes will be placed in the order that they appear in the Nodes dataframe.
library(networkD3)
states <- state.name[1:10]
ranks <- sample.int(10)
nodes <- data.frame(name = c(states, states[ranks]))
links <- data.frame(source = 1:10 - 1, target = order(ranks) + 10 - 1, value = 1)
sankeyNetwork(Links = links, Nodes = nodes, Source = 'source',
Target = 'target', Value = 'value', NodeID = 'name',
iterations = 0)
129
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With
