Studying for my MSc and I’m doing a research methology module. I’ve tried reading this multiple times and I just can’t get my head around it.
Please. Help me Obi-Wan Kenobi.
In: Mathematics
So say you have a bunch of points.
The confidence interval is the range you are most likely to find the points the majority of the time. There might be a couple points outside that range but it’s not not likely. Like if you have a strict daily schedule. You can predict where you will be with high confidence but occasionally you might have to change your schedule a little (outside the interval).
Null hypothesis is whatever condition you want really for your test. This what you are assuming is true about your points. For example, the mean is 0 or there is no difference between 2 groups.
Now the alternate hypothesis is that the null is not true. For example, the mean is not 0 or there is a difference between groups.
When you do a statistical test you will either gain evidence against the null or not. If the evidence is strong enough, you can reject the null and accept the alternate hypothesis. This is sometimes determined by p-values.
The p-values you get from some tests is the probability of the outcome matching your points assuming the null is correct. Essentially what are the chances of getting the same thing assuming the null.
So if p is really high (near 1.0) you learn the probability of the outcomes matching your points is high so the null is NOT rejected (i.e. it’s likely true). The chances are pretty good you will get the same as the null.
If p is lower than the probability threshold you set (e.g. 0.05) the probability is low that the outcome matches your data, so the null IS rejected (the alternative hypothesis is true). The chances are really low you’d get the null results.
Latest Answers