Data collection
Every sampling method is a rule for choosing those ten. Some rules give an answer near the truth. One of them can be wrong by everything, and asking more people will not save it.
scroll ↓
Population and sample
The population is all 100 students: 60 in Year 12, 40 in Year 13. Asking every one of them is a census — exact, slow, expensive.
A sample asks a few and estimates. The whole chapter is about how the choosing is done.
Simple random
Number the students 1 to 100 and draw ten at random. Every student has the same chance, and every possible sample of ten is equally likely.
It needs a sampling frame — a list of the whole population. This sample came out at 70% Year 12 when the truth is 60%: random samples wobble.
Two more rules
Systematic: every 10th from a random start, spreading the sample across the whole list.
Stratified: sample each group in proportion to its size:
Quick, and wrong
Opportunity sampling takes whoever is to hand. No frame, no numbering, no effort — and here, ten students who all sit together and are all in Year 13.
The estimate is 0%. Not every opportunity sample is this bad, but nothing in the method stops it.
Size will not fix it
Take thirty instead of ten, spreading right along the back of the room, and the estimate is still 0%.
A bigger sample buys precision, never fairness. A large biased sample is just a more confident wrong answer — which is why the method matters more than the number.
Chapter 1
Census against sample, sampling frames, all five methods with their advantages and disadvantages, quantitative and qualitative data, the large data set, and twelve graded exam questions with full worked solutions are waiting.