00:01
Hey sue, we have some questions about hypothesis testing here.
00:04
The number one, the null hypothesis is the no -effect hypothesis, the research hypothesis, the relationship researchers are interested in, or h -1.
00:17
So the correct answer here is a, the no effect.
00:21
In the null hypothesis, you assume there is no relationship between your variables.
00:30
So, for example, if you're running a drug trial, and you want to know if a drug affects blood pressure, your null hypothesis is, it doesn't affect blood pressure.
00:39
Your alternative hypothesis would be, it does.
00:42
And the null hypothesis is h0, not h1.
00:49
Which brings us on to part two.
00:51
The alternative hypothesis, a, something to nullify, states the relationship they're interested in, the starting point for empirical research, or the no effect.
01:02
So we've covered it's not the no effect.
01:04
This h1 is what they're trying to look at and it's b.
01:13
Perhaps you think maybe your drug does affect blood pressure.
01:17
Your alternative hypothesis would be it changes blood pressure.
01:22
The null hypothesis would be it doesn't.
01:24
So a is referring to h0, what you're hoping to disprove.
01:30
And the whole point of this is you start with the null hypothesis and then you disprove it.
01:36
Rather like, question three, in the us legal system, the presumption of innocence is analogous to scientific method, the alternative hypothesis, the null hypothesis or the research hypothesis.
01:53
So in the us, and most places really, you start by assuming the person is innocent and it must be proven that they are not.
02:03
You don't start by assuming they're guilty...