Statistics Y1 Flashcards
What is a census
When you sample every member of the population
What is sampling error?
The difference between the estimate of a parameter from sample data compared to the true value
What is one advantage of a census and one disadvantage
- the sample error is 0
Often impractical or too costly
Give 6 questions you can ask in order to check how valid the sample is?
- is the data you are collecting relevant? (Are you asking the right questions to get the right answers?_
- is the sample set likely to be biased? (This would introduce a systematic error)
- Does the method of collection distort the data? Eg avoid leading questions
- Is the right person collecting the data? (Bias can be introduced if people are likely to give a certain answer to a person in an authority position for example)
- Is the sample large enough? (This depends how accurate you want the sample to be?)
- Is the sampling procedure appropriate for the circumstances?
What is simple random sampling?
- every member of the parent population is equally likely to be selected
What is stratified sampling?
- put population into subgroups which are expected to have different characteristics
- subgroups are called strata
- sample all of the strata
(Usually use simple random sampling within the stratum)
What is proportional stratified sampling?
- the number sampled per strata is proportional to the size of the population of that strata
What is systematic sampling? - what issue do you have to be aware of with this?
- choose the nth item every m items in a list
- be aware of cyclic patterns within the sampling frame
What is quota sampling
- n number of a certain group must be sampled within the sample
What is opportunity sampling and when is it useful
- when circumstances make a sample readily available and it is chosen on those grounds
- useful for an initial pilot study before a wider investigation takes place
How do you find the median of n items
(n+1)/2
The median is more formally called what
A measure of tendency
What do Q1, Q2 and Q3 denote
Q1, the lower quartile
Q2, the median
Q3, upper quartile
What percentile is Q1
25th percentile
What % of data has values less than the 90th percentile
90%