Sampling.

Sampling Error

Sampling Error

Occurs when the sample doesn't exactly match the population. The error is random and will occur even for samples which are well-chosen to avoid bias.

Measurement Error

Inaccuracies in measurement. Asking people their height (continuous data so won't be exact), wording of questions

Coverage Error

occur when a sample doesn't fully reflect the population. To avoid this, samples should be sufficiently large and unbiased.

Non-response errors

Occur when a large # of people selected for a survey choose not to respond to it

There are two types of sampling. What are they?

Probability Sampling and Non-probability Sampling

Probability Sampling

Random Sampling; Usually requires a list of the entire population; Most difficult and expensive and implement; better validity

Non-Probability Sampling

Uses a method that is NOT random; Usually does not sample from the entire population; Less work and cheaper; Validity is worse because it is not a full representation of the population

Probability Sampling has three options:

Simple Random Sampling, Systemic Sampling, Stratified Sampling Method

Non-probability Sampling has two options:

Convenience Sampling and Quota Sampling

Simple Random Sampling:

Obtain a complete list of the population and randomly select participants from it. Each member of the population has an equal chance of being selected.

Systemic Sampling:

Obtain a list of the population and pick the Nth person. (Picking every 6th person, for example.)

Stratified Sampling Method:

Dividing the population into similar subpopulations (called strata). Take data from the stratas.

Convenience Sampling:

The researcher surveys people who are easy to contact and willing to participate

Quota Sampling:

Participants are non-randomly selected from population subgroups that the researcher determines

Null Hypothesis

This statement is assumed to be true unless we have enough evidence to reject it.

Alternative Hypothesis

The hypothesis that states there is a difference between two or more sets of data.