Subsetting Data

• Selecting specific variable(s)

data$method     # extracts 'method' variable values

##  [1] 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2

is.vector(data$method)      # Note that a single variable in a dataset is a vector 

## [1] TRUE

data[, "method"]

##  [1] 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2

data[, c("method", "score")]

##    method score
## 1       1    35
## 2       1    23
## 3       1    14
## 4       1    17
## 5       1    23
## 6       1    35
## 7       1    27
## 8       1    33
## 9       1    32
## 10      1    31
## 11      2    34
## 12      2    27
## 13      2    51
## 14      2    36
## 15      2    39
## 16      2    45
## 17      2    31
## 18      2    40
## 19      2    25
## 20      2    32

• Selecting specific observation(s)

data[, 1:2]     # returns the first two columns

##    method score
## 1       1    35
## 2       1    23
## 3       1    14
## 4       1    17
## 5       1    23
## 6       1    35
## 7       1    27
## 8       1    33
## 9       1    32
## 10      1    31
## 11      2    34
## 12      2    27
## 13      2    51
## 14      2    36
## 15      2    39
## 16      2    45
## 17      2    31
## 18      2    40
## 19      2    25
## 20      2    32

data[c(2, 4), ]       # returns the second and fourth rows

##   method score
## 2      1    23
## 4      1    17

data[19, 2]     # returns the value in 19th row and 2nd column

## [1] 25

• Selecting specific subgroup(s)/subset(s)

data[data$method==1, ]     # returns rows that satisfies 'method == 1'

##    method score
## 1       1    35
## 2       1    23
## 3       1    14
## 4       1    17
## 5       1    23
## 6       1    35
## 7       1    27
## 8       1    33
## 9       1    32
## 10      1    31

subset(data, method==1)    # returns a subset of data that satisfies 'method == 1' 

##    method score
## 1       1    35
## 2       1    23
## 3       1    14
## 4       1    17
## 5       1    23
## 6       1    35
## 7       1    27
## 8       1    33
## 9       1    32
## 10      1    31

data[data$score > 30, ]    # returns observations(rows) for 'scores larger than 30'

##    method score
## 1       1    35
## 6       1    35
## 8       1    33
## 9       1    32
## 10      1    31
## 11      2    34
## 13      2    51
## 14      2    36
## 15      2    39
## 16      2    45
## 17      2    31
## 18      2    40
## 20      2    32