995 Words Jun 9th, 2015 4 Pages
Business Statistics SCMA 1000 Winter 2015 Section 2
Assignment 1

Due Tuesday February 25, 2015

Sampling exercise
The purpose of this exercise is to convey some basic concepts in regards to sampling while at the same time deriving sampling distributions empirically. Deriving sampling distributions empirically works best when there are a large number of samples. The idea here is that each student in the class will create 20 samples for two populations, using two different sampling procedures, for a total of 80 samples. These samples will be combined into common datasets which will be used in class (and made available to all students in the class.
The four sampling contexts will be: 1. Discrete numerical population, sampling
Sampling context 1: Discrete numerical population, sampling without replacement
The population
The population is comprised of 10 elements, with values from 1 to 10. It a discrete variable in that the variable can only take on integer values. Also it is a uniform distribution in that each value appears once in the population.
Below is a graphical representation of the population:

Sampling procedure
The kinds of samples we are going to generate are called simple random samples but sampling will be done without replacement. Sampling without replacement means that an element is no longer available for selection once it has been selected. In other words, the element is not replaced once selected.
The sample size will be 5.
Cut out 10 cardboard squares measuring 3 cm x 3 cm
Number the squares from 1 through 10.
Place cardboard squares in a container.

Randomly select 5 squares without replacing the squares (You should have 5 squares in front of you.)
Record the numbers on the 5 squares (these are called your observed values) in columns D to H of the spreadsheet.
Do this 20 times.
For these 20 rows (samples), ‘DNWITHOUT’ should appear in Column C.
Sampling context 2: Discrete numerical population, sampling with replacement
The population
Same population as above.
Sampling procedure
The kinds of samples we are going to generate are called simple random samples but…