My comments

Population:

  • Akritas says, “In statistics the word population is used to denote the set of all objects or subjects relevent to the particular study that are exposed to the same treatment or methd”.(Pg 2)
  • Hathaway says, “the word population is used to identify the group of objects or subjects upon which inference is desired. Typically this is the group of object or subjects from which a random sampling process is applied”
  • On a test, quiz, or homework just make sure you are a little word in describing your population.

Statistical Population:

  • Akritas says, “The collection of values is called the statistical population”
  • Hathaway says, “The collection of values are the measures on the population units”
  • The key idea is to think about what you will write down in your data sheet for each measured unit.

My code

Here is example code for the case where we know that Toyota Tacoma’s are assembled in San Antionio, Texas and Baja California, Mexico. Let’s suppose that 13% are assembled in Baja California and the other 87% in San Antonio. We are asked to evaluate an issue with the alignment of the headlights as they come off the assembly line. Headquarters is investigating this issue after varied dealers receiving cars from both plants haver reported customer complaints. On any given day 690 cars are assembled in total.

  • How could we do a SRS of 100 Tacomas for a given day?
  • How could we do a SRS of 100 Tacomas for a given week?
  • How could we do a stratified random sample between the two plants?
# For a given day we could randomly select the day of the week.  After the day of the week is settled we could use
tacoma_day = c(rep("San Antonio",600),c(rep("Baja",90)))
random_picks = sample(690,100)
table(tacoma_day[random_picks])
## 
##        Baja San Antonio 
##          11          89
# the random_picks could the specific car order.

# For a given week we would just use the same code with 7*690 

# For a stratified sample there a few ways this could be done. One example is
tacoma_sa = sample(600,100*.87)
tacoma_baja = sample(90,100*.13)