Introducing the htmlTable-package

Posted on by
How should we convey complex data? The image is is CC by Sacha Fernandez.

How should we convey complex data? The image is is CC by Sacha Fernandez.

My htmlTable-function has perhaps been one of my most successful projects. I developed it in order to get tables matching those available in top medical journals. As the function has grown I’ve decided to separate it from my Gmisc-package into a separate package, and at the time of writing this I’ve just released the 1.3 version. While htmlTable allows for creating plain tables without any fancy formatting (see usage vignette) it is primarily aimed at complex tables. In this post I’ll try to show you what you can do and how to tame some of the more advanced features.

Objective: visualize migration patterns between Swedish counties the last 15 years

In this example I will try to convey a table with 240 values without overwhelming the reader. The dataset is from Statistics Sweden (downloaded using pxweb) and comes with the htmlTable-package. Our first job is to reshape our tidy dataset into a more table viewing friendly format.

library(htmlTable)
data(SCB)

# The vignette includes the Uppsala county but this generates a 
# too wide table for the blog and we therefore need to drop these
SCB <- subset(SCB, region != "Uppsala county")

# The SCB has three other coulmns and one value column
library(reshape)
SCB$region <- relevel(SCB$region, "Sweden")
SCB <- cast(SCB, year ~ region + sex, value = "values")

# Set rownames to be year
rownames(SCB) <- SCB$year
SCB$year <- NULL

The next step is to calculate two new columns:

  • Δint = The change within each group since the start of the observation.
  • Δstd = The change in relation to the overall age change in Sweden.

To separete these layers of information we use stacked column spanners:

County
Men   Women
Age Δint. Δext.   Age Δint. Δext.

These are created through using cgroup with multiple rows:

mx <- NULL
for (n in names(SCB)){
  tmp <- paste0("Sweden_", strsplit(n, "_")[[1]][2])
  mx <- cbind(mx,
              cbind(SCB[[n]], 
                    SCB[[n]] - SCB[[n]][1],
                    SCB[[n]] - SCB[[tmp]]))
}
rownames(mx) <- rownames(SCB)
colnames(mx) <- rep(c("Age", 
                      "Δint",
                      "Δstd"), 
                    times = ncol(SCB))
mx <- mx[,c(-3, -6)]

# This automated generation of cgroup elements is 
# somewhat of an overkill
cgroup <- 
  unique(sapply(names(SCB), 
                function(x) strsplit(x, "_")[[1]][1], 
                USE.NAMES = FALSE))
n.cgroup <- 
  sapply(cgroup, 
         function(x) sum(grepl(paste0("^", x), names(SCB))), 
         USE.NAMES = FALSE)*3
n.cgroup[cgroup == "Sweden"] <-
  n.cgroup[cgroup == "Sweden"] - 2

cgroup <- 
  rbind(c(cgroup, rep(NA, ncol(SCB) - length(cgroup))),
        Hmisc::capitalize(
          sapply(names(SCB), 
                 function(x) strsplit(x, "_")[[1]][2],
                 USE.NAMES = FALSE)))
n.cgroup <- 
  rbind(c(n.cgroup, rep(NA, ncol(SCB) - length(n.cgroup))),
        c(2,2, rep(3, ncol(cgroup) - 2)))

print(cgroup)

##      [,1]     [,2]                [,3]               [,4]    [,5]  [,6]   
## [1,] "Sweden" "Norrbotten county" "Stockholm county" NA      NA    NA     
## [2,] "Men"    "Women"             "Men"              "Women" "Men" "Women"

print(n.cgroup)

##      [,1] [,2] [,3] [,4] [,5] [,6]
## [1,]    4    6    6   NA   NA   NA
## [2,]    2    2    3    3    3    3

Next step is to output the table after rounding to the correct number of decimals. The txtRound function helps with this, as it uses the sprintf function instead of the round the resulting strings have the correct number of decimals, i.e. 1.02 will by round become 1, in text we generally want to retain the last decimal, i.e. 1.02 be displayed as 1.0.

htmlTable(txtRound(mx, 1), 
          cgroup = cgroup,
          n.cgroup = n.cgroup,
          rgroup = c("First period", 
                     "Second period",
                     "Third period"),
          n.rgroup = rep(5, 3),
          tfoot = txtMergeLines("Δint correspnds to the change since start",
                                "Δstd corresponds to the change compared to national average"))
Sweden   Norrbotten county   Stockholm county
Men   Women   Men   Women   Men   Women
Age Δint   Age Δint   Age Δint Δstd   Age Δint Δstd   Age Δint Δstd   Age Δint Δstd
First period
  1999 38.9 0.0   41.5 0.0   39.7 0.0 0.8   41.9 0.0 0.4   37.3 0.0 -1.6   40.1 0.0 -1.4
  2000 39.0 0.1   41.6 0.1   40.0 0.3 1.0   42.2 0.3 0.6   37.4 0.1 -1.6   40.1 0.0 -1.5
  2001 39.2 0.3   41.7 0.2   40.2 0.5 1.0   42.5 0.6 0.8   37.5 0.2 -1.7   40.1 0.0 -1.6
  2002 39.3 0.4   41.8 0.3   40.5 0.8 1.2   42.8 0.9 1.0   37.6 0.3 -1.7   40.2 0.1 -1.6
  2003 39.4 0.5   41.9 0.4   40.7 1.0 1.3   43.0 1.1 1.1   37.7 0.4 -1.7   40.2 0.1 -1.7
Second period
  2004 39.6 0.7   42.0 0.5   40.9 1.2 1.3   43.1 1.2 1.1   37.8 0.5 -1.8   40.3 0.2 -1.7
  2005 39.7 0.8   42.0 0.5   41.1 1.4 1.4   43.4 1.5 1.4   37.9 0.6 -1.8   40.3 0.2 -1.7
  2006 39.8 0.9   42.1 0.6   41.3 1.6 1.5   43.5 1.6 1.4   37.9 0.6 -1.9   40.2 0.1 -1.9
  2007 39.8 0.9   42.1 0.6   41.5 1.8 1.7   43.8 1.9 1.7   37.8 0.5 -2.0   40.1 0.0 -2.0
  2008 39.9 1.0   42.1 0.6   41.7 2.0 1.8   44.0 2.1 1.9   37.8 0.5 -2.1   40.1 0.0 -2.0
Third period
  2009 39.9 1.0   42.1 0.6   41.9 2.2 2.0   44.2 2.3 2.1   37.8 0.5 -2.1   40.0 -0.1 -2.1
  2010 40.0 1.1   42.1 0.6   42.1 2.4 2.1   44.4 2.5 2.3   37.8 0.5 -2.2   40.0 -0.1 -2.1
  2011 40.1 1.2   42.2 0.7   42.3 2.6 2.2   44.5 2.6 2.3   37.9 0.6 -2.2   39.9 -0.2 -2.3
  2012 40.2 1.3   42.2 0.7   42.4 2.7 2.2   44.6 2.7 2.4   37.9 0.6 -2.3   39.9 -0.2 -2.3
  2013 40.2 1.3   42.2 0.7   42.4 2.7 2.2   44.7 2.8 2.5   38.0 0.7 -2.2   39.9 -0.2 -2.3
Δint correspnds to the change since start
Δstd corresponds to the change compared to national average

In order to increase the readability we may want to separate the Sweden columns from the county columns, one way is to use the align option with a |. Note that in 1.0 the function continues with the same alignment until the end, i.e. you no longer need count to have the exact right number of columns in your alignment argument.

htmlTable(txtRound(mx, 1), 
          align="rrrr|r",
          cgroup = cgroup,
          n.cgroup = n.cgroup,
          rgroup = c("First period", 
                     "Second period",
                     "Third period"),
          n.rgroup = rep(5, 3),
          tfoot = txtMergeLines("Δint correspnds to the change since start",
                                "Δstd corresponds to the change compared to national average"))
Sweden   Norrbotten county   Stockholm county
Men   Women   Men   Women   Men   Women
Age Δint   Age Δint   Age Δint Δstd   Age Δint Δstd   Age Δint Δstd   Age Δint Δstd
First period
  1999 38.9 0.0   41.5 0.0   39.7 0.0 0.8   41.9 0.0 0.4   37.3 0.0 -1.6   40.1 0.0 -1.4
  2000 39.0 0.1   41.6 0.1   40.0 0.3 1.0   42.2 0.3 0.6   37.4 0.1 -1.6   40.1 0.0 -1.5
  2001 39.2 0.3   41.7 0.2   40.2 0.5 1.0   42.5 0.6 0.8   37.5 0.2 -1.7   40.1 0.0 -1.6
  2002 39.3 0.4   41.8 0.3   40.5 0.8 1.2   42.8 0.9 1.0   37.6 0.3 -1.7   40.2 0.1 -1.6
  2003 39.4 0.5   41.9 0.4   40.7 1.0 1.3   43.0 1.1 1.1   37.7 0.4 -1.7   40.2 0.1 -1.7
Second period
  2004 39.6 0.7   42.0 0.5   40.9 1.2 1.3   43.1 1.2 1.1   37.8 0.5 -1.8   40.3 0.2 -1.7
  2005 39.7 0.8   42.0 0.5   41.1 1.4 1.4   43.4 1.5 1.4   37.9 0.6 -1.8   40.3 0.2 -1.7
  2006 39.8 0.9   42.1 0.6   41.3 1.6 1.5   43.5 1.6 1.4   37.9 0.6 -1.9   40.2 0.1 -1.9
  2007 39.8 0.9   42.1 0.6   41.5 1.8 1.7   43.8 1.9 1.7   37.8 0.5 -2.0   40.1 0.0 -2.0
  2008 39.9 1.0   42.1 0.6   41.7 2.0 1.8   44.0 2.1 1.9   37.8 0.5 -2.1   40.1 0.0 -2.0
Third period
  2009 39.9 1.0   42.1 0.6   41.9 2.2 2.0   44.2 2.3 2.1   37.8 0.5 -2.1   40.0 -0.1 -2.1
  2010 40.0 1.1   42.1 0.6   42.1 2.4 2.1   44.4 2.5 2.3   37.8 0.5 -2.2   40.0 -0.1 -2.1
  2011 40.1 1.2   42.2 0.7   42.3 2.6 2.2   44.5 2.6 2.3   37.9 0.6 -2.2   39.9 -0.2 -2.3
  2012 40.2 1.3   42.2 0.7   42.4 2.7 2.2   44.6 2.7 2.4   37.9 0.6 -2.3   39.9 -0.2 -2.3
  2013 40.2 1.3   42.2 0.7   42.4 2.7 2.2   44.7 2.8 2.5   38.0 0.7 -2.2   39.9 -0.2 -2.3
Δint correspnds to the change since start
Δstd corresponds to the change compared to national average

If we still feel that we want more separation it is always possible to add colors.

htmlTable(txtRound(mx, 1), 
          col.columns = c(rep("#E6E6F0", 4),
                          rep("none", ncol(mx) - 4)),
          align="rrrr|r",
          cgroup = cgroup,
          n.cgroup = n.cgroup,
          rgroup = c("First period", 
                     "Second period",
                     "Third period"),
          n.rgroup = rep(5, 3),
                    tfoot = txtMergeLines("Δint correspnds to the change since start",
                                "Δstd corresponds to the change compared to national average"))
Sweden   Norrbotten county   Stockholm county
Men   Women   Men   Women   Men   Women
Age Δint   Age Δint   Age Δint Δstd   Age Δint Δstd   Age Δint Δstd   Age Δint Δstd
First period
  1999 38.9 0.0   41.5 0.0   39.7 0.0 0.8   41.9 0.0 0.4   37.3 0.0 -1.6   40.1 0.0 -1.4
  2000 39.0 0.1   41.6 0.1   40.0 0.3 1.0   42.2 0.3 0.6   37.4 0.1 -1.6   40.1 0.0 -1.5
  2001 39.2 0.3   41.7 0.2   40.2 0.5 1.0   42.5 0.6 0.8   37.5 0.2 -1.7   40.1 0.0 -1.6
  2002 39.3 0.4   41.8 0.3   40.5 0.8 1.2   42.8 0.9 1.0   37.6 0.3 -1.7   40.2 0.1 -1.6
  2003 39.4 0.5   41.9 0.4   40.7 1.0 1.3   43.0 1.1 1.1   37.7 0.4 -1.7   40.2 0.1 -1.7
Second period
  2004 39.6 0.7   42.0 0.5   40.9 1.2 1.3   43.1 1.2 1.1   37.8 0.5 -1.8   40.3 0.2 -1.7
  2005 39.7 0.8   42.0 0.5   41.1 1.4 1.4   43.4 1.5 1.4   37.9 0.6 -1.8   40.3 0.2 -1.7
  2006 39.8 0.9   42.1 0.6   41.3 1.6 1.5   43.5 1.6 1.4   37.9 0.6 -1.9   40.2 0.1 -1.9
  2007 39.8 0.9   42.1 0.6   41.5 1.8 1.7   43.8 1.9 1.7   37.8 0.5 -2.0   40.1 0.0 -2.0
  2008 39.9 1.0   42.1 0.6   41.7 2.0 1.8   44.0 2.1 1.9   37.8 0.5 -2.1   40.1 0.0 -2.0
Third period
  2009 39.9 1.0   42.1 0.6   41.9 2.2 2.0   44.2 2.3 2.1   37.8 0.5 -2.1   40.0 -0.1 -2.1
  2010 40.0 1.1   42.1 0.6   42.1 2.4 2.1   44.4 2.5 2.3   37.8 0.5 -2.2   40.0 -0.1 -2.1
  2011 40.1 1.2   42.2 0.7   42.3 2.6 2.2   44.5 2.6 2.3   37.9 0.6 -2.2   39.9 -0.2 -2.3
  2012 40.2 1.3   42.2 0.7   42.4 2.7 2.2   44.6 2.7 2.4   37.9 0.6 -2.3   39.9 -0.2 -2.3
  2013 40.2 1.3   42.2 0.7   42.4 2.7 2.2   44.7 2.8 2.5   38.0 0.7 -2.2   39.9 -0.2 -2.3
Δint correspnds to the change since start
Δstd corresponds to the change compared to national average

If we add a color to the row group and restrict the rgroup spanner we may even have a more visual aid.

htmlTable(txtRound(mx, 1),
          col.rgroup = c("none", "#FFFFCC"),
          col.columns = c(rep("#EFEFF0", 4),
                          rep("none", ncol(mx) - 4)),
          align="rrrr|r",
          cgroup = cgroup,
          n.cgroup = n.cgroup,
          # I use the   - the no breaking space as I don't want to have a
          # row break in the row group. This adds a little space in the table
          # when used together with the cspan.rgroup=1.
          rgroup = c("1st period", 
                     "2nd period",
                     "3rd period"),
          n.rgroup = rep(5, 3),
          tfoot = txtMergeLines("Δint correspnds to the change since start",
                                "Δstd corresponds to the change compared to national average"),
          cspan.rgroup = 1)
Sweden   Norrbotten county   Stockholm county
Men   Women   Men   Women   Men   Women
Age Δint   Age Δint   Age Δint Δstd   Age Δint Δstd   Age Δint Δstd   Age Δint Δstd
1st period          
  1999 38.9 0.0   41.5 0.0   39.7 0.0 0.8   41.9 0.0 0.4   37.3 0.0 -1.6   40.1 0.0 -1.4
  2000 39.0 0.1   41.6 0.1   40.0 0.3 1.0   42.2 0.3 0.6   37.4 0.1 -1.6   40.1 0.0 -1.5
  2001 39.2 0.3   41.7 0.2   40.2 0.5 1.0   42.5 0.6 0.8   37.5 0.2 -1.7   40.1 0.0 -1.6
  2002 39.3 0.4   41.8 0.3   40.5 0.8 1.2   42.8 0.9 1.0   37.6 0.3 -1.7   40.2 0.1 -1.6
  2003 39.4 0.5   41.9 0.4   40.7 1.0 1.3   43.0 1.1 1.1   37.7 0.4 -1.7   40.2 0.1 -1.7
2nd period          
  2004 39.6 0.7   42.0 0.5   40.9 1.2 1.3   43.1 1.2 1.1   37.8 0.5 -1.8   40.3 0.2 -1.7
  2005 39.7 0.8   42.0 0.5   41.1 1.4 1.4   43.4 1.5 1.4   37.9 0.6 -1.8   40.3 0.2 -1.7
  2006 39.8 0.9   42.1 0.6   41.3 1.6 1.5   43.5 1.6 1.4   37.9 0.6 -1.9   40.2 0.1 -1.9
  2007 39.8 0.9   42.1 0.6   41.5 1.8 1.7   43.8 1.9 1.7   37.8 0.5 -2.0   40.1 0.0 -2.0
  2008 39.9 1.0   42.1 0.6   41.7 2.0 1.8   44.0 2.1 1.9   37.8 0.5 -2.1   40.1 0.0 -2.0
3rd period          
  2009 39.9 1.0   42.1 0.6   41.9 2.2 2.0   44.2 2.3 2.1   37.8 0.5 -2.1   40.0 -0.1 -2.1
  2010 40.0 1.1   42.1 0.6   42.1 2.4 2.1   44.4 2.5 2.3   37.8 0.5 -2.2   40.0 -0.1 -2.1
  2011 40.1 1.2   42.2 0.7   42.3 2.6 2.2   44.5 2.6 2.3   37.9 0.6 -2.2   39.9 -0.2 -2.3
  2012 40.2 1.3   42.2 0.7   42.4 2.7 2.2   44.6 2.7 2.4   37.9 0.6 -2.3   39.9 -0.2 -2.3
  2013 40.2 1.3   42.2 0.7   42.4 2.7 2.2   44.7 2.8 2.5   38.0 0.7 -2.2   39.9 -0.2 -2.3
Δint correspnds to the change since start
Δstd corresponds to the change compared to national average

If you want to further add to the visual hints you can use specific HTML-code and insert it into the cells. Here we will color the Δstd according to color.

cols_2_clr <- grep("Δstd", colnames(mx))
# We need a copy as the formatting causes the matrix to loos
# its numerical property
out_mx <- txtRound(mx, 1)

min_delta <- min(mx[,cols_2_clr])
span_delta <- max(mx[,cols_2_clr]) - min(mx[,cols_2_clr]) 
for (col in cols_2_clr){
  out_mx[, col] <- mapply(function(val, strength)
    paste0("",
           val, ""), 
    val = out_mx[,col], 
    strength = round((mx[,col] - min_delta)/span_delta*100 + 1),
    USE.NAMES = FALSE)
}

htmlTable(out_mx,
          caption = "Average age in Sweden counties over a period of
                     15 years. The Norbotten county is typically known
                     for having a negative migration pattern compared to
                     Stockholm.",
          pos.rowlabel = "bottom",
          rowlabel="Year", 
          col.rgroup = c("none", "#FFFFCC"),
          col.columns = c(rep("#EFEFF0", 4),
                          rep("none", ncol(mx) - 4)),
          align="rrrr|r",
          cgroup = cgroup,
          n.cgroup = n.cgroup,
          rgroup = c("1st period", 
                     "2nd period",
                     "3rd period"),
          n.rgroup = rep(5, 3),
          tfoot = txtMergeLines("Δint correspnds to the change since start",
                                "Δstd corresponds to the change compared to national average"),
          cspan.rgroup = 1)
Average age in Sweden counties over a period of 15 years. The Norbotten county is typically known for having a negative migration pattern compared to Stockholm.
Sweden   Norrbotten county   Stockholm county
Men   Women   Men   Women   Men   Women
Year Age Δint   Age Δint   Age Δint Δstd   Age Δint Δstd   Age Δint Δstd   Age Δint Δstd
1st period          
  1999 38.9 0.0   41.5 0.0   39.7 0.0 0.8   41.9 0.0 0.4   37.3 0.0 -1.6   40.1 0.0 -1.4
  2000 39.0 0.1   41.6 0.1   40.0 0.3 1.0   42.2 0.3 0.6   37.4 0.1 -1.6   40.1 0.0 -1.5
  2001 39.2 0.3   41.7 0.2   40.2 0.5 1.0   42.5 0.6 0.8   37.5 0.2 -1.7   40.1 0.0 -1.6
  2002 39.3 0.4   41.8 0.3   40.5 0.8 1.2   42.8 0.9 1.0   37.6 0.3 -1.7   40.2 0.1 -1.6
  2003 39.4 0.5   41.9 0.4   40.7 1.0 1.3   43.0 1.1 1.1   37.7 0.4 -1.7   40.2 0.1 -1.7
2nd period          
  2004 39.6 0.7   42.0 0.5   40.9 1.2 1.3   43.1 1.2 1.1   37.8 0.5 -1.8   40.3 0.2 -1.7
  2005 39.7 0.8   42.0 0.5   41.1 1.4 1.4   43.4 1.5 1.4   37.9 0.6 -1.8   40.3 0.2 -1.7
  2006 39.8 0.9   42.1 0.6   41.3 1.6 1.5   43.5 1.6 1.4   37.9 0.6 -1.9   40.2 0.1 -1.9
  2007 39.8 0.9   42.1 0.6   41.5 1.8 1.7   43.8 1.9 1.7   37.8 0.5 -2.0   40.1 0.0 -2.0
  2008 39.9 1.0   42.1 0.6   41.7 2.0 1.8   44.0 2.1 1.9   37.8 0.5 -2.1   40.1 0.0 -2.0
3rd period          
  2009 39.9 1.0   42.1 0.6   41.9 2.2 2.0   44.2 2.3 2.1   37.8 0.5 -2.1   40.0 -0.1 -2.1
  2010 40.0 1.1   42.1 0.6   42.1 2.4 2.1   44.4 2.5 2.3   37.8 0.5 -2.2   40.0 -0.1 -2.1
  2011 40.1 1.2   42.2 0.7   42.3 2.6 2.2   44.5 2.6 2.3   37.9 0.6 -2.2   39.9 -0.2 -2.3
  2012 40.2 1.3   42.2 0.7   42.4 2.7 2.2   44.6 2.7 2.4   37.9 0.6 -2.3   39.9 -0.2 -2.3
  2013 40.2 1.3   42.2 0.7   42.4 2.7 2.2   44.7 2.8 2.5   38.0 0.7 -2.2   39.9 -0.2 -2.3
Δint correspnds to the change since start
Δstd corresponds to the change compared to national average

Although a graph most likely does the visualization task better, tables are good at conveying detailed information. It is in my mind without doubt easier in the last table to find the pattern in the data.

Lastly I would like to thank Frank Harrel for the Hmisc::latex function that inspired me to start this. Also important sources of inspirations have been Stephen Few, ThinkUI, ACAPS, and LabWrite.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.