ggplot2 pass variable to plot to a function

Question

I use ggplot to use multiple plots, so I built my own function.

plothist <- function(a) {
  ggplot(aes(x = a), data = data) + geom_histogram()
}

p1 <- plothist(data$fixed.acidity)
p2 <- plothist(data$volatile.acidity)
p3 <- plothist(data$citric.acid)
p4 <- plothist(data$residual.sugar)
p5 <- plothist(data$chlorides)
p6 <- plothist(data$free.sulfur.dioxide)
p7 <- plothist(data$total.sulfur.dioxide)
p8 <- plothist(data$density)
p9 <- plothist(data$pH)
p10 <- plothist(data$sulphates)
p11 <- plothist(data$alcohol)

x <- grid.arrange(p1, p2, p3, p4, p5, p6, p7, p8, p9, p10, p11,
                  top = "Histogram of independent variables")
x

the x-axis does not have the name of the variable, I only can see "a" on every plot, which makes the plots pretty useless. Can you help me how to display the actual variable there?

Nathan Werth · Accepted Answer · 2017-09-05 18:07:41Z

2

Use aes_string for this kind of programming:

library(ggplot2)

plothist <- function(data, column) {
  ggplot(data, aes_string(x = column)) + geom_histogram()
}

plothist(data, "fixed.acidity")

answered Sep 5, 2017 at 18:07

Nathan Werth

5,35320 silver badges25 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

pogibas Over a year ago

What if you want to mix string with non-string? One can use aes(get("STRING"), non-STRING), but what about aes_string("STRING", non-STRING)?

Nathan Werth Over a year ago

@PoGibas, I would strongly suggest avoiding non-standard evaluation in functions. If you really need to do this, use aes_. I'm not good at using it, because I've never found it necessary. Usually, adding columns inside the function and referring to them by name results in simpler code.

pogibas · Accepted Answer · 2017-09-05 17:59:45Z

2

Your function needs only minor edits:

plotHist <- function(inputData, columnToPlot) {
    # Load library as your function doesn't know
    # what is this ggplot function you want to use
    library(ggplot2)
    resultPlot <- ggplot(inputData, aes(get(columnToPlot))) + 
        geom_histogram() +
        labs(x = columnToPlot)
    return(resultPlot)
}
plotHist(mtcars, "cyl")

columnToPlot is name of a column you want to plot. Also you need to pass inputData argument as your given function doesn't know what dinputData is.

edited Sep 5, 2017 at 17:59

answered Sep 5, 2017 at 17:54

pogibas

28.5k21 gold badges92 silver badges120 bronze badges

Collectives™ on Stack Overflow

ggplot2 pass variable to plot to a function

2 Answers 2

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related