Scatterplot

Problem

You want to make a scatterplot.

Solution

Suppose this is your data:

  1. set.seed(955)
  2. # Make some noisily increasing data
  3. dat <- data.frame(xvar = 1:20 + rnorm(20,sd=3),
  4.                   yvar = 1:20 + rnorm(20,sd=3),
  5.                   zvar = 1:20 + rnorm(20,sd=3))
  7. head(dat)
  8. #>        xvar         yvar        zvar
  9. #> 1 -4.252354  3.473157275 -2.97806724
  10. #> 2  1.702318  0.005939612 -1.16183118
  11. #> 3  4.323054 -0.094252427  4.85516658
  12. #> 4  1.780628  2.072808278  4.65078709
  13. #> 5 11.537348  1.215440358 -0.06613962
  14. #> 6  6.672130  3.608111411  6.24349897

Basic scatterplots

  1. # Plot the points using the vectors xvar and yvar
  2. plot(dat$xvar, dat$yvar)
  4. # Same as previous, but with formula interface
  5. plot(yvar ~ xvar, dat)
  7. # Add a regression line
  8. fitline <- lm(dat$yvar ~ dat$xvar)
  9. abline(fitline)

plot of chunk unnamed-chunk-3plot of chunk unnamed-chunk-3

Scatterplot matrices

It is also possible to make a matrix of scatterplots if you would like to compare several variables.

See this for a way to make a scatterplot matrix with r values.

  1. # A scatterplot matrix
  2. plot(dat[,1:3])
  4. # Another way of making a scatterplot matrix, with regression lines
  5. # and histogram/boxplot/density/qqplot/none along the diagonal
  6. library(car)
  7. scatterplotMatrix(dat[,1:3],
  8.                    diagonal="histogram",
  9.                    smooth=FALSE)

plot of chunk unnamed-chunk-4plot of chunk unnamed-chunk-4

To calculate the corresponding correlation matrix, see ../../Statistical analysis/Regression and correlation.

To visualize the correlation matrix, see ../Correlation matrix.