Analyzing the Bührer family data set using an advanced mosaic plot - Thomas Roth

1. Analyzing the Bührer family
data set using an advanced
mosaic plot
Thomas Roth (IT architect, natural scientist and
hobbyist data cruncher & visualiser)
Zurich R User meetup, August 3rd 2016

2. Mosaic plot: Who’s on the map?
Map #
Bührer Status
EmigrationStatus
Map 1
Undetermined
N MD
Re−imigrated from third country
Emigrated to third country
Imigrated to CH
Emigrated from US to third country
Re−imigrated from US
Presumed US
Assumed US
US
Re−emigrated to third country
Third country emigrated to US
Presumed third country
Third country
Emigrated to US
Presumed Swiss
Swiss
Map 2
N M D
Not on Map
N M D

3. Mosaic plot code (vcd package)
library(vcd)
# Read data (source: SQL group by statement including all factors and count) and create contingency table
data <- read.csv(”[..] Data.csv")
art <- xtabs(count ~ map + emigrationstatus + buehrerstatus, data = data)
# Reorder factors and define color convention (visual tweaking)
art <- art[ c("Map 1", "Map 2", "No Map"), c("Swiss”, [..]), c("Named","Married","Descendant")]
color = c("#990000", "black", "grey”)
# Create mosaic plot
mosaic(art, direction = c("v", "h", "v"),
zero_size = 0.5, zero_split = TRUE,
margins = c(left = 12, right = 2, top = 2, bottom = 2),
shade = TRUE, gp = gpar(fill = color, col = 0),
abbreviate = c(buehrerstatus = 1),
labeling_args = list(set_varnames =
c(map = "Map #", emigrationstatus = "Emigration Status", buehrerstatus = "Bührer Status"),
set_labels = list(map = c("Map 1", "Map 2", "Not on Map")),
rot_labels = c(left = 0, bottom = 0, right = 0),
offset_varnames = c(left = 12, right = 1), offset_labels = c(left = 12, right = 0),
pos_labels = "center", just_labels = "left", rep = TRUE),
legend = FALSE )

4. More information
Visit the project’s microblog for more information on the
data, methodology, tools and visualisations used for
Mapping the emigration of the Bührers from Switzerland
to the United States
https://familytreemeetsgis.nogi.ch