HE 201
Lecture 13 Oct 11, 2013
Sampling
The process of choosing members of a population to be included in a sample
Research uses data from a sample to make inferences about a population
Selecting a Research Sample:
Ex. How do undergraduate student spend their free time?
Is it going to be a specific university?
People representation of different programs
Which year are they in their undergrad
Need an operational definition of free time
Gender, age, department
How do you select sample
What is your desired sample and who to you want to generalize your results to?
Population
The defined group of individuals from which a sample is drawn
Sample should closely represent population
Generalizability
Goal: generalize results from the limited setting in which they were originally
obtained to a larger population
When results are generalizable, they can be applied to different populations in
different settings
Selection
How do we ensure that the sample selected is representative of the pop in
which we’re interested?
If you select sample effectively, your results will be generalizable
You are trying to create a “mini population”.
Two major types of Sampling
Probability Sampling
The likelihood of any one member of pop being selected is known
I.e. 8000 wlu students, sample 200, probability is 0.025
Nonprobability Sampling
Likelihood of selecting any one member of the pop is not known
I.e., telephone book sampling not good, because not everyone is not in the
phonebook (people will be left out of sampling) Probability Sampling Strategies
Simple Random Sampling:
Each member of the pop has an equal and independent chance of being
selected as part of the sample
Equal- no bias that one person will be chosen over another
Independent- choice of one person will not affect the choice of another
Steps in Simple Random Sampling
Define the pop from which you want to select the sample
List all members of pop
Assign numbers to each member of pop
Use a criterion to select the same you want (e.g. every even number)
Types of Simple Random Sampling
Use table of random numbers
Use a computer to generate random numbers
Systematic Random Sampling
Divide the size of pop by size of desired sample
E.g. pop is 50, sample is 10, then select every 5 name
Choose one name from list at random and select every 5 name from there
(Systematic aspect of it takes away the equality of being chosen. However, it
is still random)
Stratified Sampling
Assure that the profile of the same matches the profile of pop
Similar to simple random sampling but pop is now divided in to groups on
the basis of the variables of interest
E.g. WLU has 65% females and 35% males, want sample of 100, therefore
select 65 females and 35 males randomly from the two

More
Less