How can I indicate significance in boxplots?

87 次查看(过去 30 天)
Hi guys,
I would like to compare data of several groups using boxplots. Therefore, I put 4 boxes beside each other. Now I am looking for a possibility to display asterisks over the boxplots (* for p<0.05, * * for p<0.01, ...) and also brackets to make clear which groups are significantly different at which level. To see what I am hoping to be able to do you can go on that page: http://www.frontiersin.org/files/Articles/1379/fnsys-04-00033-r1/image_m/fnsys-04-00033-g004.jpg (this is not my figure, but I would like to do something similar).
Do you have any idea which commands or toolboxes could be helpful in this case?
Thanks a lot!

采纳的回答

Star Strider
Star Strider 2015-2-12
I did this recently with Creating sigstar in bar graph.
For your box plot, experiment with this to get the result you want:
d = randi(10, 20, 3);
figure(1)
boxplot(d)
yt = get(gca, 'YTick');
axis([xlim 0 ceil(max(yt)*1.2)])
xt = get(gca, 'XTick');
hold on
plot(xt([2 3]), [1 1]*max(yt)*1.1, '-k', mean(xt([2 3])), max(yt)*1.15, '*k')
hold off
The ‘xt’ and ‘yt’ variables get the x-tick and y-tick values, and are then used to do the ‘sigline’ and asterisk calculations on. The axis call simply creates room at the top of the plot box for the line and asterisk.
  4 个评论
Jake Ogata Bernstein
One thing I also discovered is you may have to adjust the y-axis range in the axis call. For example, I'm working with data that range from -23 to +18, so I rewrote the axis call as
axis = ([xlim -25 ceil(max(yt)*1.2)])
Himanshu Saxena
Himanshu Saxena 2021-7-12
It seems these commands do not work for boxchart command. Is there any other way to do the same for boxchart?

请先登录,再进行评论。

更多回答(1 个)

Angela Braig
Angela Braig 2015-2-19
Hi again!
Thanks a lot for your answer! It was really helpful. I had to insert set(gca, 'xtick', 1:3); because my boxplots did not have xticks.
Great!
Best regards

类别

Help CenterFile Exchange 中查找有关 Creating and Concatenating Matrices 的更多信息

产品

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by