data handling notes
- Created by: loupardoe
- Created on: 05-04-16 14:57
Fullscreen
Sampling and bias
- sample data must be representative
- the population is the whole group you want to find out about
- often you can't survey the whole population
- a sample is a small group from the population
- for a sample to be representative it needs to be: a random sample (every member has an equal chance of being in it) and big enough for the size of the population
- to spot bias you need to think about when, where and how the sample is taken and how many members are in it
- if certain groups are excluded, the sample isn't random and that can lead to bias from things like age, gender and different interests
- if the sample is too small, it's also likely to be biased
Sampling methods
random sampling
- assign a number to every member of the population
- create a list of random numbers
- match the random numbers to members of the population
stratified sampling
- find the proportion…
Comments
No comments have yet been made