00:01
All right, so this problem site gives us three different examples of setting up hypothesis tests and finding and evaluating those tests at different levels of significance.
00:11
And so the first example is discussing the salaries of customer service representatives, and our null hypothesis is that the average customer service representative makes 23 ,000.
00:23
But to test this, a sample of 25 was taken, with a sample of 24 ,000, and a sample standard deviation of 5 ,000.
00:33
And so we're trying to test at the 5 % significance level if there's a statistically significant difference in the sample mean and the population mean, which would lead us to reject our null hypothesis that the population mean is 23 ,000.
00:49
And so because a sample size is 25, we're going to not rely on a z distribution and instead is a t distribution.
00:57
Anytime your population is 30 or fewer, you can assume it's a large enough sample to have a perfectly normal distribution.
01:05
And so instead we're going to be concerned with a t distribution with degrees of freedom equal to n minus 1.
01:11
So in this case, we have 24 degrees of freedom.
01:15
So that is what this column right here is showing.
01:19
And so we're interested in this row with 24 degrees of freedom.
01:23
And then because we're trying to see if there's a statistically significant difference, it doesn't matter in the direction.
01:29
We're using a two -sided distribution, and our significance level is 5%.
01:37
And so you can see here this chart is only showing a one -sided distribution.
01:41
And so instead of looking at the 0 .05, we're instead going to look at half of that 0 .025, so that on either side, this sums up to a total of 0 .05.
01:53
And so if the t statistic is greater than what has a probability, of 0 .025 in either direction.
02:03
That means that totally broadly speaking combined we have enough evidence to reject at the 5 % significance level.
02:11
So here this means that our t statistic has to be greater than or equal to 2 .06390.
02:22
So let's open a new page.
02:25
So a rejection range occurs when t is greater than or equal to 2 .06 930.
02:33
And t is calculated as our sample mean minus our population mean or null hypothesis divided by our standard sample deviation over the square root of our population size.
02:50
And so now we're just going to be plugging everything in.
02:53
So we have 24 ,000 minus 23 ,000 divided by 5 ,000 over the square root of 25.
03:04
So this is going to simplify it to 1 ,000 divided by 5 ,000 over 5.
03:14
So this 5 multiplies up.
03:16
We have 5 ,000 over 5 ,000, which equals 1.
03:25
So right now our test statistic is equal to 1, which is less than the required test statistic.
03:36
And this should be absolute value, sorry.
03:38
But it's less than the required test statistic to reject the null hypothesis.
03:43
So in this case, we're not rejecting the null hypothesis.
03:46
And so now let's put our rejection range in terms of our variable of interest.
03:52
So to do this, we take t, which is equal to 2 .0693.
04:00
It's pretty greater than it equal to.
04:02
And this is going to be equal to x bar minus 23 ,000, divided by, 5 ,000 over the squared event.
04:13
So this is basically saying, what does a sample mean that we need with a standard deviation of 5 ,000? sorry, we can put this in right away.
04:23
That would lead us to reject the null hypothesis.
04:26
And this will give us a rejection range.
04:29
And again, this is the absolute value...