Stats sampling technique

Population

The entire set of items in the group being studied

Census

Measuring every member of a population

• Accurate

• Some testing destroys the item eg. Battery

Sampling frame

List of all sampling units

Simple Random Sampling

Equal chance of being selected, give every item in the sampling frame a number then use a random number generator to select an item

• Bias free

• Needs a sampling frame

Systematic sampling

Take every kth unit, pick random number between 1 and k for start point

Eg. if k is 10, pick a random number between 1 and 10 eg 3, then use 3rd, 13th..

• quick to use

• Needs a sampling frame

Stratified sampling

Proportionally represents strata(groups) in the sample to reflect the population, use either random sampling or systematic sampling to fill the groups

• Reflects the population

• Need clear strata for the population → needs a sampling frame

Opportunity sampling

Sample based on who/what is available

• easy + cheap

• Not random → potential bias

• Unlikely to be representative of the population

Quota sampling

Starts with quotas to be filled, but not necessarily representative of the population. Groups are filled using opportunity sampling

eg. if a uni has about 60% female and 40% male you set quotas of 60 women and 40 men and then use opportunity sampling to fill those quotas

• No sampling frame needed

• Not random → potential bias

• May not be representative of the population

When can you use Binomial distribution

• Only 2 outcomes

• Probability is fixed

• Fixed number of trials

• Probability has to be independant