Samples And Populations Flashcards
What is population in a statistical context?
it is all possible scores in a variable
Why would we choose to look at a sample instead of the population?
Sometimes its not possible or would be too time consuming to look at the whole population so it is easier to look at a sample and then generalise. We can find out a lot more about a sample.
What are the problems with using samples?
One sample might provide different results to another sample from the same population
What is descriptive statistics and inferential statistics?
Descriptive: Explains that one set of data, e.g. mean, standard deviation, mode
Inferential: Allow you to infer and reach a logical conclusion about a wider situation, e.g. t-test, ANOVA, correlation
What is inference?
Drawing logical conclusions from known data.
What are the issues with inferential statistics?
It is open to error.
We can’t be (as) sure (as possible) that our sample represents our population?
What is random sampling?
Selecting a sample randomly but with organisation e.g. drawing numbers from a hat, random number tables, random number generator (calculator, excel)
When drawing numbers from a hat, why would you put the pulled number back in the hat?
Because of odds, if one is removed then the odds for picking another increase, affecting the outcome. For true randomness, the odds have to remain the same.
What are random number tables?
These are tables filled randomly with numbers by a computer.
Pick starting point at random, select X amount of scores then skip X amount of scores and repeat until sample group is filled.
How do we do random number assignment in excel?
Use formula Randbetween and then put it into how many cell boxes you need for your sample. Sometimes you will get the same number twice.
When random sampling should you have a big or small sample and why?
A big sample because it helps us come closer to approximating the community mean.
With sampling, we should work out the standard error. What will the standard error show us?
A smaller standard error suggests our sample is more representative of the population than a large standard error
What is a sample?
A subset of the population