How to plot multiple columns of a data frame to see where data exists in each column?

Question

I have following data frame with me:

    Index | ColA | ColB | ColC | ColD
    1     | NA   | NA   | 0    | NA
    2     | NA   | 0    | 1    | 0
    3     | NA   | NA   | 2    | 1
    4     | 1    | 0    | 2    | 2
    5     | NA   | NA   | 2    | NA
    6     | NA   | 1    | 1    | 1
    7     | 0    | 1    | 0    | 2
    8     | NA   | 2    | 0    | 2
    9     | NA   | 0    | NA   | 1
   10     | 2    | 1    | 0    | 0

Now, I want to plot this data using R where X-axis is Index column and Y-axis represents the remaining column names (ColA, ColB, ColC and ColD). Each x-y point in the plot should represent whether there exist a NA or Non-NA. Similar to this (for above mentioned data frame):

ColD    - - -   - - - - -
ColC  - - - - - - - -   -
ColB    -   -   - - - - -
ColA        -     -     -
      1 2 3 4 5 6 7 8 9 10

Thanks in advance for the help!

Answer 1

Here is the plot using ggplot:

Data

df <- structure(list(Index = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10),
                     ColA = c(NA, NA, NA, 1, NA, NA, 0, NA, NA, 2),
                     ColB = c(NA, 0, NA, 0, NA, 1, 1, 2, 0, 1),
                     ColC = c(0, 1, 2, 2, 2, 1, 0, 0, NA, 0), 
                     ColD = c(NA, 0, 1, 2, NA, 1, 2, 2, 1, 0)), 
                     .Names = c("Index", "ColA", "ColB", "ColC", "ColD"),
                     row.names = c(NA, -10L), class = "data.frame")                                                                       0, 1, 2, NA, 1, 2, 2, 1, 0)), .Names = c("Index", "ColA", "ColB", "ColC", "ColD"), row.names = c(NA, -10L), class = "data.frame")

Plot

library(ggplot2)
library(reshape2)
ggplot(melt(df, "Index"), aes(x=as.factor(Index), y=variable, alpha=!is.na(value))) +
  geom_point() +
  labs(x="Index", y="Variable") +
  scale_alpha_discrete("", breaks=c(TRUE, FALSE), labels=c("Not NA", "NA"))

Answer 2

Here is one method using plot.

# get values of x axis from data as a vector
xVals <- as.integer(!is.na(df)) * 1:10
# get values of y axis 
yVals <- rep(1:4, each=10)
# add appropriate NAs
is.na(xVals) <- xVals == 0
is.na(y) <- is.na(xVals)

# plot the results
plot(xVals, yVals)

data

set.seed(1234)
df <- data.frame(ColA=sample(c(0:2,NA), size=10, replace=T, prob=c(.2,.2,.2,.4)),
                 ColB=sample(c(0:2,NA), size=10, replace=T, prob=c(.2,.2,.2,.4)),
                 ColC=sample(c(0:2,NA), size=10, replace=T, prob=c(.2,.2,.2,.4)),
                 ColD=sample(c(0:2,NA), size=10, replace=T, prob=c(.2,.2,.2,.4)))