R Manual

Haven

Haven contains functions that lets R import SPSS, SAS, Stata, and other foreign files.

Documentation can be viewed here

# Check if haven is already installed and if it is, load it.
if (!require(haven)){
  # If it's not intalled, then tell R to install it.
  install.packages("haven", dependencies = TRUE)
  # Once it's installed, tell R to load it.
  library(haven)
}

Tidyverse

Tidyverse contains many functions useful to cleaning, tidying, and manipulating data

Documentation can be viewed here

if (!require(tidyverse)){
  install.packages("tidyverse", dependencies = TRUE)
  library(tidyverse)
}

summarytools

Summarytools provide useful functions for summarizing and visualizing data

Documentation can be viewed here

if (!require(summarytools)){
  install.packages("summarytools", dependencies = TRUE)
  require(summarytools)
}

psych

Summarytools provide useful functions for summarizing and visualizing data

Documentation can be viewed here

if (!require(psych)){
  install.packages("psych", dependencies = TRUE)
  require(psych)
}

corrr

quick and easy correlations

Documentation can be viewed here

if (!require(corrr)){
  install.packages("corrr", dependencies = TRUE)
  require(corrr)
}

GGally

quick and easy correlations

Documentation can be viewed here

if (!require(GGally)){
  install.packages("GGally", dependencies = TRUE)
  require(GGally)
}

Import an SPSS file

Use the read_sav() function from the Haven package to import an SPSS file into R.

Documentation can be viewed here

dataset <- read_sav("https://osf.io/98mt6/download")

Clean dataset

View dataset

Use the View() function to look at your dataset like you would in SPSS or Excel

Documentation can be viewed here

#View(dataset)

View questions or ‘labels’

This is useful if you used haven to import an SPSS file created by Qualtrics

Documentation can be viewed here

dataset %>%
  map(~ attr(., "label"))

Remove columns

Use the Select() function to de-select a column.

Put ‘-’ before the column name to remove a column (you can list more than one column).

Just type the column name to keep only that column (you can list more than one column).

Documentation can be viewed here

dataset %>%
  select(-IPAddress) -> dataset

Remove practice runs

Use the slice() funciton to remove rows.

Put the range of rows you’d like to remove with minus signs in front.

Documentation can be viewed here

dataset %>% 
  slice(-1:-3) -> dataset

Recode variables

Use the tidyverse, mutate, and case_when to recode variables Documentation can be viewed here

Recode to a factor

dataset %>%
  mutate(CoinFlipFactor = case_when(CoinFlip==1 ~ "Heads",
                                    CoinFlip==2 ~ "Tails")) -> dataset

Recode to a dummy variable

dataset %>%
  mutate(FlipHeadsDummy = case_when(CoinFlip==1 ~ 1,
                                   CoinFlip==2 ~ 0)) -> dataset

Summarize variables

Use dfSummary function from summarytools package to summarize and visualize data

Documentation can be viewed here

#First select the variables you'd like to summarize
dataset %>%
  select (CoinFlip, FFM_5, Potter3) -> exampleDF

#Then print them with this command
print(dfSummary(exampleDF, graph.magnif = .75), method = 'render')

Use the describe() function of the psych package to summarize data

Documentation can be viewed here

#First select the variables you'd like to summarize
dataset %>%
  select (CoinFlip, FFM_5, Potter3) %>%
  describe()

Create Composite Variable

Psych

Documentation can be viewed here

Extraversion

#create dataframe with only relevant variables to work with
Extraversion <- data.frame (dataset$FFM_1, dataset$FFM_6, dataset$FFM_11, dataset$FFM_16, dataset$FFM_21, dataset$FFM_26, dataset$FFM_31, dataset$FFM_36)
#create list of 'keys'. The  numbers just refer to the order of the question in the data.frame() you just made. The most important thing is to mark the questions that should be reversed scored with a '-'. 
Extraversion.keys <- make.keys(Extraversion, list(Extraversion=c(1,-2,3,4,-5,6,-7,8)))
#score the scale
Extraversion.scales <- scoreItems (Extraversion.keys, Extraversion)
#save the scores
Extraversion.scores <- Extraversion.scales$scores
#save the scores back in 'dataset'
dataset$Extraversion <- Extraversion.scores[,]
#print the cronbach alpha
Extraversion.scales$alpha

Agreeableness

#create dataframe with only relevant variables to work with
Agreeableness <- data.frame (dataset$FFM_2, dataset$FFM_7, dataset$FFM_12, dataset$FFM_17, dataset$FFM_22, dataset$FFM_27, dataset$FFM_32, dataset$FFM_37, dataset$FFM_42)
Agreeableness.keys <- make.keys(Agreeableness, list(Agreeableness=c(-1,2,-3,4,5,-6,7,-8,9)))
Agreeableness.scales <- scoreItems (Agreeableness.keys, Agreeableness)
Agreeableness.scores <- Agreeableness.scales$scores
dataset$Agreeableness <- Agreeableness.scores[,]
Agreeableness.scales$alpha

Conscientiousness

#create dataframe with only relevant variables to work with
Conscientiousness <- data.frame (dataset$FFM_3, dataset$FFM_8, dataset$FFM_13, dataset$FFM_18, dataset$FFM_23, dataset$FFM_28, dataset$FFM_33, dataset$FFM_38, dataset$FFM_43)
my.keys <- make.keys(Conscientiousness, list(Conscientiousness=c(1,-2,3,-4,-5,6,7,8,-9)))
my.scales <- scoreItems (my.keys, Conscientiousness)
my.scores <- my.scales$scores
dataset$Conscientiousness <- my.scores[,]
my.scales$alpha

Neuroticism

#create dataframe with only relevant variables to work with
Neuroticism <- data.frame (dataset$FFM_4, dataset$FFM_9, dataset$FFM_14, dataset$FFM_19, dataset$FFM_24, dataset$FFM_29, dataset$FFM_34, dataset$FFM_39)
my.keys <- make.keys(Neuroticism, list(Neuroticism=c(1,-2,3,4,-5,6,-7,8)))
my.scales <- scoreItems (my.keys, Neuroticism)
my.scores <- my.scales$scores
dataset$Neuroticism <- my.scores[,]
my.scales$alpha

Openness

#create dataframe with only relevant variables to work with
Openness <- data.frame (dataset$FFM_5, dataset$FFM_10, dataset$FFM_15, dataset$FFM_20, dataset$FFM_25, dataset$FFM_30, dataset$FFM_35, dataset$FFM_40, dataset$FFM_41, dataset$FFM_44)
my.keys <- make.keys(Openness, list(Openness=c(1,2,3,4,5,6,-7,8,-9,10)))
my.scales <- scoreItems (my.keys, Openness)
my.scores <- my.scales$scores
dataset$Openness <- my.scores[,]
my.scales$alpha

Tidyverse

Documentation can be viewed here and here

dataset %>%
  rowwise() %>%
  mutate(extraversionTidy = mean(c(FFM_1, 6-FFM_6, FFM_11, FFM_16, 6-FFM_21, FFM_26, 6-FFM_31, FFM_36))) -> dataset

Correlations

Calculate correlations

Explore correlations

Documentation can be viewed here

dataset %>%
  select(Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism) %>%
  correlate() %>%
  shave() %>%
  fashion()

Calculate p-values

Documentation can be viewed here

dataset %>%
  select(Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism) -> Big5df

corr.test(x = Big5df, 
          y = NULL,
          use = "pairwise",
          method = "pearson",
          adjust = "holm", 
          alpha = .05,
          ci = TRUE)

Deciding between Pearson and Spearman correlations

Documentation can be viewed here

dataset %>%
  select(Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism) %>%
  ggpairs()

t-tests

Documentation can be viewed here

independent samples t-tests

dataset %>%
  with(t.test(Agreeableness ~ CoinFlipFactor))

paired samples t-tests

dataset %>%
  with(t.test(Agreeableness, Neuroticism, paired = TRUE))