Replace X-Axis Values in R (Example) | How to Change & Customize Ticks

In this R tutorial you’ll learn how to modify x-axis values manually.

The content of the post looks as follows:

Let’s get started.

Creating Example Plot

Let’s first create an exemplifying graphic in R:

plot(1:5)                       # Create plot in Base R

default axis r

Figure 1: Plot with Default X-Axis Values.

Figure 1 shows the output of the previous R code: A scatterplot with an x-axis ranging from 1 to 5.

Example: Changing X-Axis Values to Letters

Let’s assume that we want to modify our x-axis tick marks so that the currently shown numbers are replaced by upper case letters. Then, we first have to create a graph without any axis values:

plot(1:5,                       # Create plot without x-axis
     xaxt = "n",
     xlab = "My X-Axis")

without x axis plot r

Figure 2: Plot without X-Axis Values.

Figure 2 shows our plot without axis values.

Now, we can add custom axis values by using the axis function as shown below:

axis(1,                         # Define x-axis manually
     at = 1:5,
     labels = LETTERS[1:5])

custom axis r

Figure 3: Plot with Custom X-Axis Values.

Figure 3 shows our final result: A scatterplot with custom axis values.

Video & Further Resources

Do you need more explanations on the R codes of this article? Then I can recommend having a look at the following video instruction which I have published on my YouTube channel. I show the R programming syntax of the present tutorial in the video.

Please accept YouTube cookies to play this video. By accepting you will be accessing content from YouTube, a service provided by an external third party.

YouTube privacy policy

If you accept this notice, your choice will be saved and the page will refresh.

Furthermore, you might have a look at the related articles that I have published on my website. I have released numerous tutorials already.

In summary: You have learned in this tutorial how to change the values on our axis scale in the R programming language.

In this tutorial, I have illustrated how to do this based on an exemplifying scatterplot. However, please note that it would also be possible to change the axis values of other types of graphs such as a histogram, boxplot, barchart, line plot, or a density plot by using the same kind of R code.

If you have further questions, please let me know in the comments section below.

18 Comments. Leave new

Scottie Teichmer

February 3, 2022 12:54 am

How can I customize ticks to string values? E.g. “Control”, “Trial 1”, “Trial 2”.

Thank you for any help.

Joachim

February 7, 2022 12:11 pm

Hey,

You can add basically any text you want by using the label argument in the axis function. Have a look at the following example code:

plot(1:3,                       # Create plot without x-axis
     xaxt = "n",
     xlab = "My X-Axis")
axis(1,                         # Define x-axis manually
     at = 1:3,
     labels = c("Control",
                "Trial 1",
                "Trial 2"))

Regards,
Joachim

Apsana

May 16, 2022 5:30 pm

How can I customize range in the axis, for instance point A= 5-10, B=10-15 and so on

Joachim

May 17, 2022 6:30 am

Hey Apsana,

Are you looking for this?

plot(1:3,                       # Create plot without x-axis
     xaxt = "n",
     xlab = "My X-Axis")
axis(1,                         # Define x-axis manually
     at = 1:3,
     labels = c("5-10",
                "11-15",
                "16-2"))

Regards,
Joachim

Enrico SOzza
August 27, 2022 9:27 am

Is there a way to manually change axis ticks values also with ggplot2?

Many Thanks,
Enrico

Reply
- Joachim
  August 29, 2022 9:30 am
  
  Hey Enrico,
  
  Are you looking for this?
  
  Regards,
  Joachim
  
  Reply
  - Enrico Sozza
    August 30, 2022 10:03 am
    
    Hello Joachim,
    
    thank you for your help it’s a very easy solution! However in my case I am using plot (to plot ggeffects object from ggeffects package), and apparently it doesn’t work. Let me explain you my situation:
    
    p5 <- ggemmeans(B9, "variable") #where B9 is a GLMM S4 object
    #p5 is now a data.frame with x (level 0 or 1) in the first column, the predicted value in the 2nd column and the Standard error in column n°3
    
    plot(p5, ci.style="errorbar", add.data=FALSE, dodge = 1) +
    xlab("")+ ylab("Predicted Probability of Hit") + ggtitle ("Effect of Listening Time") +
    theme(plot.title = element_text(hjust = 0.5), text=element_text(family="Calibri", size=12,color="grey20"),
    title = element_text(family="Calibri", size=12,hjust=0.5),
    axis.title.y = element_text(family="Calibri", size=12,vjust=2),
    axis.text.x = element_text(family="Calibri",vjust=0.5, size=10),
    axis.text.y = element_text(family="Calibri",vjust=0.5, size=10),
    axis.ticks.y = element_line(c("60","70"), color="grey35", size=0.2))
    
    In this way on my x axis I get 0 and 1 but I want to change it with 0=10 seconds and 1=50 seconds..
    levels(p5$x) <- c("10 seconds","50 seconds") doesn't work.
    scale_x_discrete() doesn't work cause apparently there is an already existing mapping of the aesthetic (probably inherent to "plot" from ggeffects?)
    
    I don't really know how to solve this issue.. I could just copy the value but then it's tricky for the error bars.
    
    Thank you very much for your help,
    Regards
    Enrico
    
    Reply
    - Joachim
      August 30, 2022 2:03 pm
      Hey Enrico,
      
      Thank you for the clarifications. Could you please execute the following line of code and share the output? This would make it much easier for me to reproduce your data.
      
      head(p5)
      
      Regards,
      Joachim
      Reply
Enrico Sozza
August 30, 2022 3:00 pm

Hello again and thanks for your help.

Here you are the output:

variable | Predicted | 95% CI
———————————————-
0 | 0.59 | [0.54, 0.63]
1 | 0.75 | [0.70, 0.78]

Note that if I run “levels(p5$x) <- c("10 seconds","50 seconds")",
the output would be the following:

variable | Predicted | 95% CI
———————————————-
10 seconds | 0.59 | [0.54, 0.63]
50 seconds | 0.75 | [0.70, 0.78]

but running the above reported code "plot(…..)" will return a plot with 0 and 1 on the x-axis..

It shouldn't since I labelled it, but the issue might be related to how "plot()" function from ggeffects read the argument data.. I am really lost on it, I hope there's a way to overcome this and that im now wasting my and your time.

Thank you again,
Best
Enrico

Reply
- Joachim
  August 31, 2022 6:27 am
  
  Hey Enrico,
  
  I have tried to reproduce your situation. However, could you please explain why you are using the plot() function (i.e. plot(p5, ci.style=”errorbar”, add.data=FALSE, dodge = 1) ) to draw your data? The plot function creates Base R plots instead of ggplot2 plots, and this line already returns warning messages when I run your code.
  
  Regards,
  Joachim
  
  Reply
Enrico Sozza
August 31, 2022 10:14 am

Hi Joachim,

so “plot()” could either be a generic X-Y plotting function, a default scatterplot function or ggeffects-objects plot function. In my case is the last one (type F1 on “plot()” after installing and loading ggeffects and you’ll see all the informations), and it’s a function generated for plotting ggeffects objects (in my case –> ggmeans()). The ggeffects-object is a ggplot2-type object (as you can see I used ggtitle() and other adding codes for ggplot2). I could use ggplot() + geom_col() but I would be struggling with geom_errorbar() (I wouldn’t know how to reproduce the same graph with normal ggplot2 functions): for that reason plot() was made, as a straightforward solution for those objects. If you have any suggestion on how to proceed avoiding “plot()” from ggeffects, I would be very thankful for receiving it.

Thank you very much again,
Best
Enrico

Reply
- Joachim
  September 2, 2022 8:43 am
  Hey Enrico,
  
  Unfortunately, I’m still not able to reproduce your situation. I have tried the following code, but the plot function still doesn’t work as you are describing:
  set.seed(67243) p5 <- data.frame(variable = rbinom(100, 1, 0.5), predicted = runif(100)) p5 library("ggeffects") library("ggplot2") plot(p5, ci.style="errorbar", add.data=FALSE, dodge = 1) # + # xlab("")+ ylab("Predicted Probability of Hit") + ggtitle ("Effect of Listening Time") + # theme(plot.title = element_text(hjust = 0.5), text=element_text(family="Calibri", size=12,color="grey20"), # title = element_text(family="Calibri", size=12,hjust=0.5), # axis.title.y = element_text(family="Calibri", size=12,vjust=2), # axis.text.x = element_text(family="Calibri",vjust=0.5, size=10), # axis.text.y = element_text(family="Calibri",vjust=0.5, size=10), # axis.ticks.y = element_line(c("60","70"), color="grey35", size=0.2))
  I do not have any experience with the ggeffects package, so maybe it might be better to ask this question to the Statistics Globe Facebook group. Perhaps somebody there knows how to resolve this problem: https://www.facebook.com/groups/statisticsglobe
  
  Regards,
  Joachim
  Reply
Enrico Sozza
September 2, 2022 2:26 pm

Hello again Joachim and thank you for your time.

I think the problem in your codes is that the p5 object is not a ggeffects object (namely an object obtained through a function inside ggeffects such as “ggmean”). Unfortunately I don’t have Facebook but thank you for the hint. I decided to change the coding of the variable’s level in the dataset before running the regression model so that at the end I will hopefully obtain what I need.

Thank you very much again,
Best
Enrico

Reply
- Joachim
  September 5, 2022 10:56 am
  
  Hey Enrico,
  
  Thank you for the further explanations. I’m sorry that I couldn’t help you more, and I hope you’ll find a solution soon.
  
  Regards,
  Joachim
  
  Reply
Louise
November 17, 2022 11:21 pm

How do you manually add standard error bars when you don’t have the replicate data? What coding should you use? Geom_errorbar()??

Reply
- Joachim
  November 18, 2022 8:12 am
  
  Hey Louise,
  
  Are you looking for this?
  
  Regards,
  Joachim
  
  Reply
  - Louise
    November 18, 2022 9:18 am
    
    Hello, I am trying to add the error bars to a bar chat
    
    Reply
    - Joachim
      November 18, 2022 8:23 pm
      
      Ah OK, then please have a look here.
      
      Regards,
      Joachim
      
      Reply