Cross-tabulation of clusters

Chapter 5.3 Cross-tabulation of groups from different dissimilarity matrices

Click here to get instructions…

• Please download and unzip the replication files for Chapter 5 ( Chapter05.zip).
• Read readme.html and run 5-0_ChapterSetup.R. This will create 5-0_ChapterSetup.RData in the sub folder data/R. This file contains the data required to produce the plots shown below.
• You also have to add the function legend_large_box to your environment in order to render the tweaked version of the legend described below. You find this file in the source folder of the unzipped Chapter 5 archive.
• We also recommend to load the libraries listed in Chapter 5’s LoadInstallPackages.R

# assuming you are working within .Rproj environment
library(here)

# install (if necessary) and load other required packages
source(here("source", "load_libraries.R"))

# load environment generated in "5-0_ChapterSetup.R"
load(here("data", "R", "5-0_ChapterSetup.RData"))

In chapter 5.3, we introduce one of the options to account for the parallel unfolding of temporal processes: the cross-tabulation of cluster solutions extracted separately from two (or more) pools of sequences representing the trajectories in different domains. We are now using the data.frame multidim, which contains both family formation and labour market sequences. The data come from a sub-sample of the German Family Panel - pairfam. For further information on the study and on how to access the full scientific use file see here.

Preparatory work for family formation trajectories

First, we run a Ward cluster analysis based on the dissimilarity matrix mc.fam.year.om:

fam.ward<-hclust(as.dist(mc.fam.year.om), 
                       method="ward.D", 
                       members=multidim$weight40)

… to be used as initialization of the PAM clustering

fam.pam <- wcKMedRange(mc.fam.year.om, 
                            weights = multidim$weight40, 
                            kvals = 2:10,
                            initialclust = fam.ward)

We now extract 5 clusters…

fam.pam.5cl <- fam.pam$clustering$cluster5

…attach the cluster info to the main data.frame multidim…

multidim$fam.pam.5cl<-fam.pam.5cl

… and re-label clusters from 1 to 5 instead of medoid identifiers…

fam.pam.5cl.factor <- factor(fam.pam.5cl, 
                             levels = c(16, 460, 479, 892, 898), 
                             c("1", "2", "3", "4", "5"))

…to finally attach the factor info to the main data.frame multidim:

multidim$fam.pam.5cl.factor<-fam.pam.5cl.factor

Preparatory work for labor market trajectories

First, we run a Ward cluster analysis based on the dissimilarity matrix mc.act.year.om:

act.ward<-hclust(as.dist(mc.act.year.om), 
                 method="ward.D", 
                 members=multidim$weight40)

… to be used as initialization of the PAM clustering

act.pam <- wcKMedRange(mc.act.year.om, 
                       weights = multidim$weight40, 
                       kvals = 2:10,
                       initialclust = act.ward)

We now extract 5 clusters…

act.pam.5cl <- act.pam$clustering$cluster5

…attach the cluster info to the main data.frame multidim…

multidim$act.pam.5cl<-act.pam.5cl

… and re-label clusters from 1 to 5 instead of medoid identifiers…

act.pam.5cl.factor <- factor(act.pam.5cl, 
                             levels = c(6, 25, 78, 539, 709), 
                             c("1", "2", "3", "4", "5"))

…to finally attach the factor info to the main data.frame multidim

multidim$act.pam.5cl.factor<-act.pam.5cl.factor

Cross-tabulation for a 5-cluster solution on both channels

Tabulate the two vectors and store the results in an object that we name crosstab…

crosstab<-table(multidim$act.pam.5cl.factor, multidim$fam.pam.5cl.factor)

…to print it at our convenience:

crosstab

   
      1   2   3   4   5
  1 118  40  58  76  42
  2  74  32  44  43  42
  3  69  32  21  32  12
  4  38  13   7  19   5
  5 131   8   9  29  33