Data in R packages

data()               # the datasets package
data(package="vcd")  # in the vcd package

typically, load data from a package using data()

data(UCBAdmissions)
str(UCBAdmissions)
##  table [1:2, 1:2, 1:6] 512 313 89 19 353 207 17 8 120 205 ...
##  - attr(*, "dimnames")=List of 3
##   ..$ Admit : chr [1:2] "Admitted" "Rejected"
##   ..$ Gender: chr [1:2] "Male" "Female"
##   ..$ Dept  : chr [1:6] "A" "B" "C" "D" ...
sum(UCBAdmissions)
## [1] 4526
margin.table(UCBAdmissions, 1)
## Admit
## Admitted Rejected 
##     1755     2771
margin.table(UCBAdmissions, 2:3)
##         Dept
## Gender     A   B   C   D   E   F
##   Male   825 560 325 417 191 373
##   Female 108  25 593 375 393 341

data frames

library(vcd)         # load the vcd package & make its datasets available
## Warning: package 'vcd' was built under R version 3.1.2
## Loading required package: grid
data(Arthritis)
str(Arthritis)
## 'data.frame':    84 obs. of  5 variables:
##  $ ID       : int  57 46 77 17 36 23 75 39 33 55 ...
##  $ Treatment: Factor w/ 2 levels "Placebo","Treated": 2 2 2 2 2 2 2 2 2 2 ...
##  $ Sex      : Factor w/ 2 levels "Female","Male": 2 2 2 2 2 2 2 2 2 2 ...
##  $ Age      : int  27 29 30 32 46 58 59 59 63 63 ...
##  $ Improved : Ord.factor w/ 3 levels "None"<"Some"<..: 2 1 1 3 3 3 1 3 1 1 ...
head(Arthritis)      # see the first few lines
##   ID Treatment  Sex Age Improved
## 1 57   Treated Male  27     Some
## 2 46   Treated Male  29     None
## 3 77   Treated Male  30     None
## 4 17   Treated Male  32   Marked
## 5 36   Treated Male  46   Marked
## 6 23   Treated Male  58   Marked

making tables from data frame variables

table(Arthritis$Improved)
## 
##   None   Some Marked 
##     42     14     28
table(Arthritis$Treatment, Arthritis$Sex)
##          
##           Female Male
##   Placebo     32   11
##   Treated     27   14
with(Arthritis, table(Treatment, Sex))
##          Sex
## Treatment Female Male
##   Placebo     32   11
##   Treated     27   14

xtabs() is often easier

art.table <- xtabs(~ Sex + Treatment + Improved, data=Arthritis)
ftable(art.table)    # display as flattened table
##                  Improved None Some Marked
## Sex    Treatment                          
## Female Placebo              19    7      6
##        Treated               6    5     16
## Male   Placebo              10    0      1
##        Treated               7    2      5
summary(art.table)   # chi-square test for mutual independence
## Call: xtabs(formula = ~Sex + Treatment + Improved, data = Arthritis)
## Number of cases in table: 84 
## Number of factors: 3 
## Test for independence of all factors:
##  Chisq = 19.6, df = 7, p-value = 0.006501
##  Chi-squared approximation may be incorrect
plot(art.table, shade=TRUE)

Reading data from external files

read a data table from a local file (NB: ‘/’ not ’' for all systems)

# arthritis <- read.csv("N:/psy6136/data/arthritis.csv")
# arthritis <- read.csv(file.choose())  
# or, read the same data from a web URL ...
arthritis <- read.csv("http://euclid.psych.yorku.ca/www/psy6136/data/Arthritis.csv")

levels(arthritis$Improved)
## [1] "Marked" "None"   "Some"

make an ordered factor

arthritis$Improved <- ordered(arthritis$Improved,
                              levels=c("None", "Some", "Marked"))

make a variable discrete: cut and arithmetic

table(10*floor(arthritis$Age/10)) 
## 
## 20 30 40 50 60 70 
##  4 11 11 29 26  3
table(cut(arthritis$Age, breaks=6))
## 
## (22.9,31.5]   (31.5,40]   (40,48.5]   (48.5,57]   (57,65.5] (65.5,74.1] 
##           8           7          10          19          26          14

assign new variable in data frame

arthritis$AgeGroup <- factor(10*floor(arthritis$Age/10))

simple plots

plot(arthritis$Age)      # index plot

plot(arthritis$AgeGroup) # barplot for a factor

plot(Improved ~ AgeGroup, data=arthritis) # spineplot for two factors

plot(arthritis[,2:5])    # scatterplot matrix; not too useful for factors