CODEWITHSUNDEEP
rggplot2data-visualizationbar-chart
xhi4m
xhi4m

Reputation: 29

ggplot2 R : Percent stacked barchart with multiple variables

R version 4.0.5 (2021-03-31) Platform: x86_64-w64-mingw32/x64 (64-bit) Running under: Windows 10 x64 (build 19042)

I want to create a percent stacked barchart including 2 groups (regional, international) and the means of 4 different numerical variables (ground low-intensity, ground high-intensity, standing low-intensity, standing high-intensity). The latter variables are representing the duration of each time period in seconds.

My data are: dataset

The image below represents an example of what I kind want to make: Time-motion analysis description relative to total fight time, considering modalities and positions of actions Coswig, V. S., Gentil, P., Bueno, J. C., Follmer, B., Marques, V. A., & Del Vecchio, F. B. (2018). Physical fitness predicts technical-tactical and time-motion profile in simulated Judo and Brazilian Jiu-Jitsu matches. PeerJ, 6, e4851.

I have read a lot of guides and watched many YT tutorials, but most of them are using 2 categorical and 1 numerical variable, thus, it does not work in my case.

Any help or guidance would be highly appreciated.

Thank you in advance.

Upvotes: 1

Views: 527

Answers (1)

Ray
Ray

Reputation: 2288

You will find a lot of friends here, if you provide a reproducible example and show what you have done and where things go wrong.

data

ds <- tribble(
    ~GROUP, ~GLI, ~GHI,~SLI, ~SHI,~GT,~ST,~EFFORT, ~PAUSE, ~HI, ~LI
    ,"REG", 158, 48, 26, 4, 205, 30, 235, 10, 51, 184
    ,"INT", 217, 62, 20, 1, 279, 21, 300, 11, 63, 237
)

{ggplot} works best with long data. Here tidyr is your friend and pivot_longer()

ds <- ds %>% 
 pivot_longer(
         cols=c(GLI:SHI)          # wich cols to take
       , names_to = "intensity"   # where to put the names aka intensitites
       , values_to = "duration"   # where to put the values you want to plot
    ) %>% 
#-------------------- calculate the shares of durations per group
    group_by(GROUP) %>% 
    mutate(share = duration / sum(duration)
)

This gives you a tibble like this:

# A tibble: 8 x 10
# Groups:   GROUP [2]
  GROUP    GT    ST EFFORT PAUSE    HI    LI intensity duration   share
  <chr> <dbl> <dbl>  <dbl> <dbl> <dbl> <dbl> <chr>        <dbl>   <dbl>
1 REG     205    30    235    10    51   184 GLI            158 0.669  
2 REG     205    30    235    10    51   184 GHI             48 0.203  
3 REG     205    30    235    10    51   184 SLI             26 0.110  
4 REG     205    30    235    10    51   184 SHI              4 0.0169 
5 INT     279    21    300    11    63   237 GLI            217 0.723  
6 INT     279    21    300    11    63   237 GHI             62 0.207  
7 INT     279    21    300    11    63   237 SLI             20 0.0667 
8 INT     279    21    300    11    63   237 SHI              1 0.00333

With the last columns providing you your categories and % durations, the grouping is done with the GROUP variable. And then you can print it with ggplot.

ds %>%
    ggplot() + 
    geom_col(aes(x = GROUP, y = share, fill = intensity), position = position_stack())  + 
    scale_y_continuous(labels=scales::percent)

enter image description here

You can then "beautify" the plot, chosing desired theme, colours, legends, etc. Hope this gets you started!

Upvotes: 1

Related Questions