Generate a list of 10 random numbers between 1 and 100:
sample(1:100, 10)## [1] 100 85 83 99 66 47 1 51 74 57Generate a list of 10 random numbers between 1 and 100 allowing repetitions:
sample(1:100, 10, replace=T)## [1] 69 65 40 49 79 55 48 94 6 30Generate a list of 10 random numbers between 1 and 10 allowing repetitions:
sample(1:10, 10, replace=T)## [1] 3 1 4 4 2 5 7 2 6 5Randomly order the numbers 1 to 10:
sample(1:10, 10, replace=F)## [1] 2 3 4 8 6 5 9 1 7 10Generate 10 random numbers using the numbers 1 and 2:
sample(1:2, 10, replace=T)## [1] 2 1 2 1 1 1 2 2 2 2Load the data set StudentSurvey from the package Lock5Data:
require(Lock5Data)## Loading required package: Lock5Datadata(StudentSurvey)Select 10 random observations:
sample(StudentSurvey, 10)## Year Gender Smoke Award HigherSAT Exercise TV Height Weight
## 67 FirstYear M No Olympic Verbal 10 20 72 195
## 273 Sophomore F No Nobel Math 3 2 NA NA
## 345 FirstYear M Yes Nobel Math 11 6 68 135
## 215 Sophomore M No Olympic Math 15 15 68 140
## 144 Junior M No Olympic Verbal 18 20 75 175
## 125 Sophomore F No Olympic Verbal 12 2 80 150
## 94 Sophomore M No Nobel Math 14 5 73 206
## 287 Senior M No Olympic Math 15 5 73 200
## 42 Sophomore F No Nobel Verbal 12 4 64 135
## 330 Sophomore M Yes Academy Math 3 5 70 165
## Siblings BirthOrder VerbalSAT MathSAT SAT GPA Pulse Piercings
## 67 2 2 600 600 1200 3.10 60 1
## 273 2 2 610 640 1250 3.50 88 2
## 345 1 1 540 670 1210 3.30 84 0
## 215 1 2 570 580 1150 3.00 64 0
## 144 0 1 560 500 1060 2.93 74 0
## 125 4 1 650 600 1250 3.15 68 2
## 94 1 1 560 660 1220 2.00 48 0
## 287 2 1 570 610 1180 3.00 57 0
## 42 2 3 580 540 1120 2.40 73 7
## 330 3 2 630 680 1310 3.25 77 0
## orig.ids
## 67 67
## 273 273
## 345 345
## 215 215
## 144 144
## 125 125
## 94 94
## 287 287
## 42 42
## 330 330Select 10 random observations and save them into the data set sample10. This command will also generates a column named origin.ids that gives information on where in the original data set the random observations fall.
sample10<-sample(StudentSurvey, 10)Select 10 random observations and save them into the data set sample10. The option origin.ids=F makes it so that no column named origin.ids is generated, so that the sample has no information on where in the original data set the random observations fall.
sample10<-sample(StudentSurvey, 10, orig.ids=F)Select 5 random observations for which the class year is Junior:
sample(subset(StudentSurvey, Year=="Junior"), 10, orig.ids=F)## Year Gender Smoke Award HigherSAT Exercise TV Height Weight
## 220 Junior M No Olympic Math 10 15 71 180
## 31 Junior M No Academy Math 8 8 62 175
## 192 Junior F No Nobel Math 8 6 65 140
## 40 Junior F No Nobel Verbal 5 5 64 130
## 331 Junior M Yes Olympic Math 3 3 67 150
## 321 Junior M Yes Nobel Math 12 5 75 185
## 53 Junior M No Olympic Math 15 0 72 175
## 14 Junior F No Nobel Verbal 6 1 68 135
## 325 Junior F Yes Nobel Math 15 10 64 145
## 46 Junior M No Olympic Verbal 5 14 73 159
## Siblings BirthOrder VerbalSAT MathSAT SAT GPA Pulse Piercings
## 220 1 2 620 640 1260 3.40 54 0
## 31 2 3 610 620 1230 3.50 55 0
## 192 2 3 670 710 1380 3.80 84 2
## 40 0 NA 570 570 1140 2.40 48 4
## 331 2 2 620 640 1260 2.90 66 0
## 321 1 2 670 680 1350 3.00 76 0
## 53 2 1 650 670 1320 3.60 68 1
## 14 2 3 650 650 1300 3.08 66 4
## 325 1 1 480 690 1170 3.50 64 8
## 46 1 1 620 620 1240 3.31 61 0Select 5 samples of size 2 using the command do to repeat the random sample selection:
do(5)*sample(StudentSurvey, 2, orig.ids=F)## Year Gender Smoke Award HigherSAT Exercise TV Height Weight
## 71 Senior F No Olympic Math 2 6 67 145
## 127 Sophomore M No Nobel Verbal 20 6 70 195
## 239 Sophomore M No Academy Math 5 10 74 210
## 170 Sophomore M No Olympic Math 8 10 70 183
## 346 Sophomore F Yes Nobel Verbal 18 5 69 140
## 29 FirstYear M No Olympic Math 12 6 68 190
## 65 FirstYear M No Olympic Math 15 6 70 175
## 344 Sophomore M Yes Olympic Math 5 5 70 165
## 271 FirstYear M No Olympic Math 3 5 74 145
## 339 Sophomore M Yes Nobel 6 2 74 185
## Siblings BirthOrder VerbalSAT MathSAT SAT GPA Pulse Piercings .row
## 71 3 2 540 640 1180 3.50 62 5 1
## 127 0 1 640 640 1280 3.50 76 0 2
## 239 1 2 510 530 1040 2.70 78 0 1
## 170 3 2 470 500 970 2.20 84 4 2
## 346 1 1 680 580 1260 3.20 87 4 1
## 29 1 1 550 560 1110 2.50 74 0 2
## 65 2 3 500 540 1040 NA 63 0 1
## 344 1 1 530 600 1130 3.25 70 0 2
## 271 1 2 560 620 1180 3.00 70 0 1
## 339 1 2 600 600 1200 3.48 80 0 2
## .index
## 71 1
## 127 1
## 239 2
## 170 2
## 346 3
## 29 3
## 65 4
## 344 4
## 271 5
## 339 5Select 100 samples of 30 students each, and save the mean SAT score for each sample into a data set called sample30SAT:
sample30SAT<-do(100)*mean(~SAT, data=sample(StudentSurvey, 30, orig.ids=F))Find the mean of the 100 sample mean SAT scores we just created, and compare it to the real mean SAT score of the entire data set:
mean(~result, data=sample30SAT)## [1] 1203.664mean(~SAT, data=StudentSurvey)## [1] 1203.627The two means are very close to each other indicating that the sample distribution is unbiased.