Why don’t you just ask everyone?
You’d think using all the data would always be best, but in real life that can be too slow, too expensive, or just impossible. If you want to know what a whole year group thinks, you usually ask a smaller set of people and use that to judge the bigger group.
After you’ve chosen that smaller set, you use it to make a conclusion about the whole group. That only works well if your chosen people actually reflect the wider group instead of coming from one easy-to-reach corner.
A population is the full set you want information about.
A sample is the smaller part of that full set that you actually collect data from.
A sampling frame is the complete list you use to pick your sample from.
Think of your sample like a trailer for a film. If the trailer only shows one type of scene, you get the wrong idea about the whole movie. A good sample should be representative, meaning it keeps the main features of the population in similar proportions.
A larger sample often helps because repeated samples won’t vary as wildly, but size is not magic. If your method is biased, collecting loads of data just gives you a bigger pile of misleading results.
9 more sections in this topic
Pick this up in your Library: it holds the whole topic, notes, cheatsheet and questions.