7 Tables

7.1 Displaying tables - htmlTable

htmlTable is a package that will produce the HTML code needed to display a table in a browser.

library(htmlTable)
htmlTable::htmlTable(esoph[1:5,]) # You can only see what the table looks like when Knitr generates a document

	agegp	alcgp	tobgp	ncases	ncontrols
1	25-34	0-39g/day	0-9g/day	0	40
2	25-34	0-39g/day	10-19	0	10
3	25-34	0-39g/day	20-29	0	6
4	25-34	0-39g/day	30+	0	5
5	25-34	40-79	0-9g/day	0	27

There are various options to change how a htmlTable will look:

htmlTable::htmlTable(esoph[1:5,], 
                     rnames = FALSE,
                     header = c("Age Group","Alcohol Group","Tobacco Group","ncases","ncontrol"),  # Change headings - specify all columns  
                     caption = "Oesophageal Cancer",  #To add a Title   
                     align = 'lllrr',       #To align column values 'r'=right, 'l'=left, c='centre(default)
                     align.header='lllcc',  #To align column headers
                     css.table = "width: 70%; table-layout: fixed; word-wrap: break-word;") # Add CSS formats to the table

Oesophageal Cancer
Age Group	Alcohol Group	Tobacco Group	ncases	ncontrol
25-34	0-39g/day	0-9g/day	0	40
25-34	0-39g/day	10-19	0	10
25-34	0-39g/day	20-29	0	6
25-34	0-39g/day	30+	0	5
25-34	40-79	0-9g/day	0	27

More css styling is avilable : https://www.w3schools.com/css/css_table.asp

7.2 Displaying Interactive tables - DT

DT is a package that will produce an interactive table in a browser.

library(DT)

## Warning: package 'DT' was built under R version 3.4.4

DT::datatable(mtcars[1:5,1:5])

There are various options and additional format functions to change how a DT table looks:

mtcars[,1:5] %>% mutate(cyl = as.factor(cyl)) %>%
  DT::datatable(colnames = c("Miles per gallon","Cylinders","disp","hp","drat")  # Rename columns
               ,rownames = FALSE   #To remove row numbers
               ,width = '80%'
               ,filter = 'top'     #Add filters above each column 
               ,options = list(pageLength = 4,     # Set no. rows on each page
                               dom = 'rtBp')) %>%  # Specify wheter search box etc. appears
  DT::formatStyle(0, target = 'row', lineHeight='8px') %>%  #Changes row height
  DT::formatRound(columns=c('mpg', 'drat'), digits=1)  #Rounds values to 1dp

You might need to convert some columns to factors in order for the filters to work.
For more detail : https://rstudio.github.io/DT/

7.3 Creating Tabulations

Before displaying a table you may need to coerce your data into the right shape. A basic frequency table can be created by using dplyr functions as then displayed with htmlTable:

mtcars %>% count(Cylinders=cyl, Gears=gear) %>% htmlTable::htmlTable(rnames=FALSE)

Cylinders	Gears	n
4	3	1
4	4	8
4	5	2
6	3	2
6	4	4
6	5	1
8	3	12
8	5	2

When grouping by multiple categories the results will be in a tidy format.
To produce the same reuslts as a 2-way tables, you can use the spread function

mtcars %>% count(Cylinders=cyl, Gears=gear) %>%
           spread(Gears, n) %>%  # The 'Gears' category will be along columns, 'n' values will fill the cells  
           htmlTable::htmlTable(rnames=FALSE)

Cylinders	3	4	5
4	1	8	2
6	2	4	1
8	12		2

To easily add a row totals use the adorn functions from the janitor package:

mtcars %>% count(Cylinders=cyl, Gears=gear) %>%
           spread(Gears, n) %>%   
           janitor::adorn_totals("row") %>%  
           htmlTable::htmlTable(rnames=FALSE, total=TRUE)

Cylinders	3	4	5
4	1	8	2
6	2	4	1
8	12		2
Total	15	12	5

They only way to nest multiple categories along the column dimension, is to create a combined field using the unite function.

mtcars %>% count(Cylinders=cyl, Gears=gear, am) %>%
           unite(gears_am, Gears, am, sep = " <br> ") %>% 
           spread(gears_am, n) %>%
           htmlTable::htmlTable(rnames=FALSE)

Cylinders	3 0	4 0	4 1	5 1
4	1	2	6	2
6	2	2	2	1
8	12			2

An alternative to the above is to reshape data with a cast function. The best version of this function is from the data.table package (it originally comes from the reshape package).

data.table::data.table(mtcars) %>%
 data.table::dcast(cyl + am ~ gear, sep = "<br>", fun.aggregate = c(length, median), value.var = c("mpg","wt")) %>%
  htmlTable::htmlTable(rnames=FALSE)

cyl	am	mpg length 3	mpg length 4	mpg length 5	wt length 3	wt length 4	wt length 5	mpg median 3	mpg median 4	mpg median 5	wt median 3	wt median 4	wt median 5
4	0	1	2	0	1	2	0	21.5	23.6		2.465	3.17
4	1	0	6	2	0	6	2		28.85	28.2		2.0675	1.8265
6	0	2	2	0	2	2	0	19.75	18.5		3.3375	3.44
6	1	0	2	1	0	2	1		21	19.7		2.7475	2.77
8	0	12	0	0	12	0	0	15.2			3.81
8	1	0	0	2	0	0	2			15.4			3.37

# Your data must be converted to a data.table format.
# The shape of the table is defined by a formula (eg. x1 + x2 ~ y) where variables on the left of the ~ represent row dimensions and those on the right represent column dimensions.
# You can calcuate one or more statistics (mean, max, median..) to appear in the cells, for one or more values. The function automatically aggregates (ie. no group-by is needed).