Questions and Answers

1 Questions about R

1.1 Why doesn’t my `select`/`filter` statement work?

When you are loading packages, sometimes different packages have the same function names in them, and the functions themselves will do very different things. For example, there is a select function in the tidyverse, but also another select function in the package MASS that does something very different. If we load the tidyverse before loading MASS, then the MASS version of select is the one that will be used?!

library(tidyverse)
library(MASS)

diamonds %>% select(carat, cut, color)

To get around this make sure that you load the tidyverse after MASS, to be safe you should always load the tidyverse last.

library(MASS) 
library(tidyverse)

diamonds %>% select(carat, cut, color)

# A tibble: 53,940 × 3
   carat cut       color
   <dbl> <ord>     <ord>
 1  0.23 Ideal     E    
 2  0.21 Premium   E    
 3  0.23 Good      E    
 4  0.29 Premium   I    
 5  0.31 Good      J    
 6  0.24 Very Good J    
 7  0.24 Very Good I    
 8  0.26 Very Good H    
 9  0.22 Fair      E    
10  0.23 Very Good H    
# ℹ 53,930 more rows

you can also specify the package that select comes from (in this case from a package within the tidyverse called dplyr):

diamonds %>% dplyr::select(carat, cut, color)

# A tibble: 53,940 × 3
   carat cut       color
   <dbl> <ord>     <ord>
 1  0.23 Ideal     E    
 2  0.21 Premium   E    
 3  0.23 Good      E    
 4  0.29 Premium   I    
 5  0.31 Good      J    
 6  0.24 Very Good J    
 7  0.24 Very Good I    
 8  0.26 Very Good H    
 9  0.22 Fair      E    
10  0.23 Very Good H    
# ℹ 53,930 more rows

Finally, there is a package that helps your deal with conflicts called conflicted. At the top of your script you can define which version of a function you prefer, this will be the version that will always be used. If you want to specify another version of a function, you can still use other versions of the function using the :: notation:

library(conflicted)
conflicts_prefer(dplyr::select)
conflicts_prefer(dplyr::mutate)
conflicts_prefer(dplyr::summarise)
conflicts_prefer(dplyr::filter)

df %>% select(name, age) # uses select from dplyr
df %>% MASS::select(name) # uses select from MASS

To find out if you have any conflicts, you can run conflict_scout().

1.2 How can I unload packages

If you are finding yourself with a conflict as mentioned above and want to unload packages, then you need to run the following code:

# adapted from: @mmfrgmpds https://stackoverflow.com/questions/7505547/detach-all-packages-while-working-in-r
while(!is.null(sessionInfo()$loadedOnly)){
  lapply(names(sessionInfo()$loadedOnly), require, character.only = TRUE)
  invisible(lapply(paste0('package:', names(sessionInfo()$otherPkgs)), detach, character.only=TRUE, unload=TRUE, force=TRUE))    
}

1.3 Why am I getting the error ‘could not find function “%>%”’

The pipe operator %>% is loaded when you load the tidyverse package - make sure you have installed tidyverse and loaded it

install.packages("tidyverse")  # install
library(tidyverse)             # load

1.4 I am getting the error: ‘“Error: Mapping should be created with `aes()` or `aes_()`.”’ when using ggplot

This may be caused by having a bracket after the geom rather than before it

data<-data.frame(x=5:1,
                 y=10:6)
ggplot(data, aes(x, y) +           # Reproduce error message
         geom_point())

data<-data.frame(x=5:1,
                 y=10:6)
ggplot(data, aes(x, y)) +           # Fixed error by moving bracket
         geom_point()

1.5 My axis labels are too long

If you want to use long axis titles you may find they overrun the space available

data<-data.frame(x=5:1,
                 y=10:6)
ggplot(data, aes(x=x, y=y)) +           
         geom_col()+
  ylab("A very long description for the y-axis label that will overflow and not look very nice")

To insert a line break in the axis label, add \n to the text where you want line breaks.

data<-data.frame(x=5:1,
                 y=10:6)
ggplot(data, aes(x=x, y=y)) +           
         geom_col()+
  ylab("A very long description \n for the y-axis label that will \n overflow and not look very nice")

1.6 How can I find out the full item labels in the PISA data frame?

The code below will give you a list of all the item labels

lapply(PISA_2022, attr, "label")