Load Libraries

library(ggplot2)
library(tidyverse)
## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2 ──
## ✔ tibble  3.1.6     ✔ dplyr   1.0.8
## ✔ tidyr   1.2.0     ✔ stringr 1.4.0
## ✔ readr   2.1.3     ✔ forcats 0.5.1
## ✔ purrr   0.3.4     
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
library(skimr)
library(psych)
## 
## Attaching package: 'psych'
## 
## The following objects are masked from 'package:ggplot2':
## 
##     %+%, alpha
library(reshape2)
## 
## Attaching package: 'reshape2'
## 
## The following object is masked from 'package:tidyr':
## 
##     smiths
library(stats)

Shortcuts:

Commenting out multiple lines of code: cmd + shift + c

The shorthand for inserting chunks into an R Markdown file is cmd + options + i

Importing Data

We will be using a data set built into the reshape2 package called french fries (see here: https://rdrr.io/cran/reshape2/man/french_fries.html). The variables are:

time in weeks from start of study treatment (type of oil) subject replicate potato-y flavour buttery flavour grassy flavour rancid flavour painty flavour

df1 <- french_fries #load in data - if we were importing data, we would use the csv file

head(df1) #shows the first 6 lines
##    time treatment subject rep potato buttery grassy rancid painty
## 61    1         1       3   1    2.9     0.0    0.0    0.0    5.5
## 25    1         1       3   2   14.0     0.0    0.0    1.1    0.0
## 62    1         1      10   1   11.0     6.4    0.0    0.0    0.0
## 26    1         1      10   2    9.9     5.9    2.9    2.2    0.0
## 63    1         1      15   1    1.2     0.1    0.0    1.1    5.1
## 27    1         1      15   2    8.8     3.0    3.6    1.5    2.3
View(df1) #shows the entire data set

Inspecting our data

There are many ways to inspect our data - see below!

str(df1) #lets you know what each variable type is
## 'data.frame':    696 obs. of  9 variables:
##  $ time     : Factor w/ 10 levels "1","2","3","4",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ treatment: Factor w/ 3 levels "1","2","3": 1 1 1 1 1 1 1 1 1 1 ...
##  $ subject  : Factor w/ 12 levels "3","10","15",..: 1 1 2 2 3 3 4 4 5 5 ...
##  $ rep      : num  1 2 1 2 1 2 1 2 1 2 ...
##  $ potato   : num  2.9 14 11 9.9 1.2 8.8 9 8.2 7 13 ...
##  $ buttery  : num  0 0 6.4 5.9 0.1 3 2.6 4.4 3.2 0 ...
##  $ grassy   : num  0 0 0 2.9 0 3.6 0.4 0.3 0 3.1 ...
##  $ rancid   : num  0 1.1 0 2.2 1.1 1.5 0.1 1.4 4.9 4.3 ...
##  $ painty   : num  5.5 0 0 0 5.1 2.3 0.2 4 3.2 10.3 ...
summary(df1) #get descriptive info for each variable
##       time     treatment    subject         rep          potato      
##  1      : 72   1:232     10     : 60   Min.   :1.0   Min.   : 0.000  
##  2      : 72   2:232     15     : 60   1st Qu.:1.0   1st Qu.: 4.000  
##  3      : 72   3:232     16     : 60   Median :1.5   Median : 7.200  
##  4      : 72             19     : 60   Mean   :1.5   Mean   : 6.953  
##  5      : 72             51     : 60   3rd Qu.:2.0   3rd Qu.: 9.900  
##  6      : 72             52     : 60   Max.   :2.0   Max.   :14.900  
##  (Other):264             (Other):336                 NA's   :1       
##     buttery           grassy            rancid           painty      
##  Min.   : 0.000   Min.   : 0.0000   Min.   : 0.000   Min.   : 0.000  
##  1st Qu.: 0.000   1st Qu.: 0.0000   1st Qu.: 0.400   1st Qu.: 0.000  
##  Median : 0.700   Median : 0.0000   Median : 2.700   Median : 0.600  
##  Mean   : 1.824   Mean   : 0.6642   Mean   : 3.852   Mean   : 2.522  
##  3rd Qu.: 2.925   3rd Qu.: 0.9000   3rd Qu.: 6.500   3rd Qu.: 4.100  
##  Max.   :11.200   Max.   :11.1000   Max.   :14.900   Max.   :13.100  
##  NA's   :4        NA's   :1         NA's   :1        NA's   :2
summary(df1$treatment) #get descriptive info for only the variable "treatment" - using a $ specifies the column that you could like to use
##   1   2   3 
## 232 232 232

You could also use the describe functions in the psych package, which are almost identical to the summary() function

describe(df1) #describe() is a function from the psych package
##            vars   n mean   sd median trimmed  mad min  max range  skew kurtosis
## time*         1 696 5.36 2.82    5.0    5.34 2.97   1 10.0   9.0  0.04    -1.19
## treatment*    2 696 2.00 0.82    2.0    2.00 1.48   1  3.0   2.0  0.00    -1.50
## subject*      3 696 6.47 3.41    6.5    6.46 3.71   1 12.0  11.0  0.01    -1.21
## rep           4 696 1.50 0.50    1.5    1.50 0.74   1  2.0   1.0  0.00    -2.00
## potato        5 695 6.95 3.58    7.2    7.00 4.45   0 14.9  14.9 -0.11    -0.95
## buttery       6 692 1.82 2.41    0.7    1.36 1.04   0 11.2  11.2  1.45     1.32
## grassy        7 695 0.66 1.32    0.0    0.36 0.00   0 11.1  11.1  3.48    17.87
## rancid        8 695 3.85 3.78    2.7    3.39 4.00   0 14.9  14.9  0.80    -0.47
## painty        9 694 2.52 3.39    0.6    1.88 0.89   0 13.1  13.1  1.29     0.57
##              se
## time*      0.11
## treatment* 0.03
## subject*   0.13
## rep        0.02
## potato     0.14
## buttery    0.09
## grassy     0.05
## rancid     0.14
## painty     0.13
describeBy(df1, group = df1$treatment)
## 
##  Descriptive statistics by group 
## group: 1
##            vars   n mean   sd median trimmed  mad min  max range  skew kurtosis
## time*         1 232 5.36 2.83    5.0    5.34 2.97   1 10.0   9.0  0.04    -1.20
## treatment*    2 232 1.00 0.00    1.0    1.00 0.00   1  1.0   0.0   NaN      NaN
## subject*      3 232 6.47 3.42    6.5    6.46 3.71   1 12.0  11.0  0.01    -1.22
## rep           4 232 1.50 0.50    1.5    1.50 0.74   1  2.0   1.0  0.00    -2.01
## potato        5 232 6.89 3.74    7.2    6.91 4.45   0 14.9  14.9 -0.08    -1.05
## buttery       6 231 1.78 2.47    0.6    1.27 0.89   0 11.2  11.2  1.55     1.57
## grassy        7 232 0.65 1.38    0.0    0.34 0.00   0 11.1  11.1  4.26    26.29
## rancid        8 232 4.07 3.87    2.8    3.64 4.15   0 14.3  14.3  0.70    -0.72
## painty        9 232 2.58 3.33    1.2    1.93 1.78   0 13.1  13.1  1.40     1.14
##              se
## time*      0.19
## treatment* 0.00
## subject*   0.22
## rep        0.03
## potato     0.25
## buttery    0.16
## grassy     0.09
## rancid     0.25
## painty     0.22
## ------------------------------------------------------------ 
## group: 2
##            vars   n mean   sd median trimmed  mad min  max range  skew kurtosis
## time*         1 232 5.36 2.83   5.00    5.34 2.97   1 10.0   9.0  0.04    -1.20
## treatment*    2 232 2.00 0.00   2.00    2.00 0.00   2  2.0   0.0   NaN      NaN
## subject*      3 232 6.47 3.42   6.50    6.46 3.71   1 12.0  11.0  0.01    -1.22
## rep           4 232 1.50 0.50   1.50    1.50 0.74   1  2.0   1.0  0.00    -2.01
## potato        5 232 7.00 3.58   7.05    7.03 4.60   0 14.1  14.1 -0.08    -1.04
## buttery       6 230 1.97 2.49   0.80    1.52 1.19   0 10.1  10.1  1.29     0.74
## grassy        7 232 0.66 1.25   0.00    0.36 0.00   0  7.1   7.1  2.59     7.55
## rancid        8 232 3.62 3.80   2.20    3.12 3.26   0 14.9  14.9  0.89    -0.34
## painty        9 231 2.46 3.51   0.10    1.78 0.15   0 12.7  12.7  1.31     0.47
##              se
## time*      0.19
## treatment* 0.00
## subject*   0.22
## rep        0.03
## potato     0.24
## buttery    0.16
## grassy     0.08
## rancid     0.25
## painty     0.23
## ------------------------------------------------------------ 
## group: 3
##            vars   n mean   sd median trimmed  mad min  max range  skew kurtosis
## time*         1 232 5.36 2.83    5.0    5.34 2.97   1 10.0   9.0  0.04    -1.20
## treatment*    2 232 3.00 0.00    3.0    3.00 0.00   3  3.0   0.0   NaN      NaN
## subject*      3 232 6.47 3.42    6.5    6.46 3.71   1 12.0  11.0  0.01    -1.22
## rep           4 232 1.50 0.50    1.5    1.50 0.74   1  2.0   1.0  0.00    -2.01
## potato        5 231 6.97 3.44    7.3    7.04 4.00   0 14.5  14.5 -0.19    -0.80
## buttery       6 231 1.72 2.27    0.7    1.29 1.04   0 10.2  10.2  1.49     1.60
## grassy        7 231 0.68 1.33    0.0    0.38 0.00   0 10.5  10.5  3.29    15.47
## rancid        8 231 3.87 3.69    2.8    3.40 4.15   0 13.3  13.3  0.79    -0.35
## painty        9 231 2.53 3.36    0.5    1.92 0.74   0 11.8  11.8  1.15     0.05
##              se
## time*      0.19
## treatment* 0.00
## subject*   0.22
## rep        0.03
## potato     0.23
## buttery    0.15
## grassy     0.09
## rancid     0.24
## painty     0.22

For character strings, it is useful to get a count of how many cases fall into these categories.

table(df1$treatment) #this is a raw count of each category specified (treatment)
## 
##   1   2   3 
## 232 232 232
ggplot(df1, aes(x = potato)) +
  geom_histogram() +
  stat_bin(bins = 20) #graph a histogram for one column of data
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## Warning: Removed 1 rows containing non-finite values (stat_bin).
## Removed 1 rows containing non-finite values (stat_bin).

Boolean Expressions

less than is > less than or equal to <= greater than is > greater than or equal to is >= equals is == not equal to is !=

Logical Operators

or is ‘OR’ and is ‘&’

More Data Inspecting

skim(df1) #for each variables, it gives you the n missing, min, max, empty n, number of unique values, mean, stdev, quartiles, mini histrogram
Data summary
Name df1
Number of rows 696
Number of columns 9
_______________________
Column type frequency:
factor 3
numeric 6
________________________
Group variables None

Variable type: factor

skim_variable n_missing complete_rate ordered n_unique top_counts
time 0 1 FALSE 10 1: 72, 2: 72, 3: 72, 4: 72
treatment 0 1 FALSE 3 1: 232, 2: 232, 3: 232
subject 0 1 FALSE 12 10: 60, 15: 60, 16: 60, 19: 60

Variable type: numeric

skim_variable n_missing complete_rate mean sd p0 p25 p50 p75 p100 hist
rep 0 1.00 1.50 0.50 1 1.0 1.5 2.00 2.0 ▇▁▁▁▇
potato 1 1.00 6.95 3.58 0 4.0 7.2 9.90 14.9 ▅▇▇▇▂
buttery 4 0.99 1.82 2.41 0 0.0 0.7 2.92 11.2 ▇▂▁▁▁
grassy 1 1.00 0.66 1.32 0 0.0 0.0 0.90 11.1 ▇▁▁▁▁
rancid 1 1.00 3.85 3.78 0 0.4 2.7 6.50 14.9 ▇▃▂▂▁
painty 2 1.00 2.52 3.39 0 0.0 0.6 4.10 13.1 ▇▂▁▁▁ 
#Can use skim_without_charts() function for no mini histograms

Subsetting Data

df1_treat1 <- subset(df1, treatment == 1)

filter() is the Tidyverse way of subsetting the data

df1_treat2 <- dplyr::filter(df1, treatment == 2)

Subsetting by treatment == 1 AND rep ==1

df1_treat1_rep1 <- dplyr::filter(df1, treatment == 1 & rep == 1)

Selecting variables (columns) using the select() function

df1_special <- dplyr::select(df1, subject, treatment, rep, potato, buttery) #Selecting only a few variables
head(df1_special)
##    subject treatment rep potato buttery
## 61       3         1   1    2.9     0.0
## 25       3         1   2   14.0     0.0
## 62      10         1   1   11.0     6.4
## 26      10         1   2    9.9     5.9
## 63      15         1   1    1.2     0.1
## 27      15         1   2    8.8     3.0

Deleting variables using the select() function

df1_delete <- dplyr::select(df1, -subject, -treatment, -rep, -potato, -buttery) #A minus sign deletes a column from a data set
head(df1_delete)
##    time grassy rancid painty
## 61    1    0.0    0.0    5.5
## 25    1    0.0    1.1    0.0
## 62    1    0.0    0.0    0.0
## 26    1    2.9    2.2    0.0
## 63    1    0.0    1.1    5.1
## 27    1    3.6    1.5    2.3

Piping in Tidyverse %>%

Pipes are a way of telling R, “then do this” without having to run separate lines of code.

Shortcut for piping is shift + command + m

df1_final <- df1 %>% 
  dplyr::filter(treatment == 1) %>% 
  dplyr::select(subject, treatment, rep, potato, buttery)
head(df1_final)
##    subject treatment rep potato buttery
## 61       3         1   1    2.9     0.0
## 25       3         1   2   14.0     0.0
## 62      10         1   1   11.0     6.4
## 26      10         1   2    9.9     5.9
## 63      15         1   1    1.2     0.1
## 27      15         1   2    8.8     3.0
pl_hist_buttery <- df1_final %>% 
  ggplot(aes(x = buttery)) + 
  geom_histogram() + 
  theme_minimal()

pl_hist_buttery #show the plot of buttery
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## Warning: Removed 1 rows containing non-finite values (stat_bin).

Exporting Data from R to a CSV file

The arguments are your data frame and the path + name you want the file to save

write.csv(df1_final, "~/Documents/R Scripts General/output/final-data.csv")

Export Plots from R

I’m saving my histogram plot from above as an image file.

ggsave(plot = pl_hist_buttery, file =  "~/Documents/R Scripts General/output/pl_hist_buttery.png",
       width = 4, height = 3.5)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## Warning: Removed 1 rows containing non-finite values (stat_bin).

Re-coding variable values

Let’s re-code treatment into a new variable, where the character values are instead numeric values (e.g., we want the treatment 1 to be peanut oil and so on ).

table(df1$treatment) #Check how many levels of treatment in the data 
## 
##   1   2   3 
## 232 232 232

Base-R method for recoding values from one variable into another

df1$treatment_name <- NA #Makes a new column name
df1$treatment_name[df1$treatment == 1] <- "Peanut Oil" #If treatment = 1, then make the treatment name "Peanut Oil"
df1$treatment_name[df1$treatment == 2] <- "Canola Oil" #If treatment = 2, then make the treatment name "Canola Oil"
df1$treatment_name[df1$treatment == 3] <- "Vegetable Oil" #If treatment = 3, then make the treatment name "Vegetable Oil"

View(df1) #check your work - specifically the new treatment_name column!

Removing NA Values

df2 <- df1_special %>% 
  na.omit() #omit NA values and save it as a new data set

Renaming

Rename columns - rename the individual variables using new_name = old_name In the example below, the original variable “potato” is now named “potato_flavor”

df2 <- df2 %>% rename("potato_flavor" = potato, "buttery_flavor" = buttery)
head(df2)
##    subject treatment rep potato_flavor buttery_flavor
## 61       3         1   1           2.9            0.0
## 25       3         1   2          14.0            0.0
## 62      10         1   1          11.0            6.4
## 26      10         1   2           9.9            5.9
## 63      15         1   1           1.2            0.1
## 27      15         1   2           8.8            3.0

Summarizing

Summarize your data by a grouping variable

df2 %>% 
  group_by(treatment) %>% 
  summarize(mean_potato = mean(potato_flavor)) #get potato flavor means for the 3 treatments
## # A tibble: 3 × 2
##   treatment mean_potato
##   <fct>           <dbl>
## 1 1                6.87
## 2 2                7.01
## 3 3                6.97

A more complicated summarizing scenario…

df2 %>% 
  group_by(treatment, rep) %>% #group by treatment and repetition
  summarize(mean_potato = mean(potato_flavor), mean_buttery = mean(buttery_flavor)) #take the mean of potato and buttery flavors
## `summarise()` has grouped output by 'treatment'. You can override using the
## `.groups` argument.
## # A tibble: 6 × 4
## # Groups:   treatment [3]
##   treatment   rep mean_potato mean_buttery
##   <fct>     <dbl>       <dbl>        <dbl>
## 1 1             1        6.74         1.80
## 2 1             2        7.00         1.76
## 3 2             1        7.18         1.99
## 4 2             2        6.84         1.96
## 5 3             1        6.94         1.81
## 6 3             2        7.00         1.63

Visualize Data

Generate a scatterplot using potato, and buttery variables at repetition == 1

df1_final %>% #take your data and then... 
  filter(rep == 1) %>% #take only first repetitions and then... 
  ggplot(aes(x = potato, y = buttery)) +
  geom_point() #plot potato flavor on the x axis and buttery flavor on the y axis
## Warning: Removed 1 rows containing missing values (geom_point).

I can make the same graph as above but make it look a little nicer with some aesethics…

df1_final %>% #take your data and then... 
  filter(rep == 1) %>% #take only first repetitions and then... 
  ggplot(aes(x = potato, y = buttery)) + #plot potato flavor on the x axis and buttery flavor on the y axis
  geom_point(size = 2, color = "dodgerblue") + # this is a scatterplot (geom_point) and the color is blue
  theme_minimal() + #the theme is minimal (no grey background)
  xlab("Potato Flavor") + #add x axis label
  ylab("Buttery Flavor") + #add y axis label
  ggtitle("Scatterplot of Potato Flavor and Buttery Flavor at the First Taste") #add a title
## Warning: Removed 1 rows containing missing values (geom_point).