Census vs. Survey
A parameter is a numerical description of a population characteristic
A statistic is a numerical description of a sample characteristic
Census | Population | Parameter |
---|---|---|
Survey | Sample | Statistic |
Characteristics of a Well-Designed and Well-Conducted Survey
Representative sample
- Must represent the population that you want to draw inference about
Random sample
Does not introduce bias
Bias
What is it?
How might it occur?
Wording of the question
Sample technique
Under-coverage: miss a certain group of people
Non-response: someone in your sample choose not to respond
Random Sampling
Simple Random Sample (SRS)
Random digit table
Calculator
- MATH ➡️ PRB ➡️ randInt(lower, upper, number of numbers)
![П-84 Pius Stver Editj(n ф TEns [NSTRUMENTS ](./media/image77.png)
- Stratified Random Sampling
Cluster Sample
Systematic Random Sample
Comparison
Example
If you wanted to select a sample of school children and you did this by selecting an SRS of classrooms this would be an example of what type of sampling?
- Cluster
If you wanted to select a sample of school children and you did this by randomly selecting 5 children from each grade this would be an example of what type of sampling?
- Stratified
Non-Random Sampling
Convenience Sample
- Selecting those that are easy of convenient
Voluntary Response Sample