 Hi, this is Dr. Don. I want to take a few minutes and show you how you can use the calculator, the two sample t-test calculator you can download free from my website, works to solve a typical two sample t-test problem. Now this calculator does essentially the same thing that the data analysis tool pack calculator does, but it's got an added feature that automatically calculates both the left tail, right tail p-values and shows you the difference in those as well as the two tail test. So let's use an example. Here's our data and this is data from a manager that has two clinics, medical clinics that she's responsible for and she wanted to know was there a difference in the throughput time, patient throughput time. That's the time from when a patient signs in at the desk until they check out at the exit desk. That's the throughput. And she has two clinics she's responsible for and she was curious to see if there was a significant difference in the average throughput time. Now we're going to run this as the two tail test and in that case we would make the assumption, the null assumption that there is no significant difference and that's where she is. She doesn't really know. But for example say that she had been getting complaints about clinic two being slow, then she might assume that clinic one's time is significantly less than clinic two, faster or the other way around, that clinic two is faster than clinic one. But we're just going to focus on this two tail test. We want to know either way, is there a significant difference. So here's the data. She calculated the average patient throughput on each of 30 days to give a better snapshot of the capability of the clinics. And she came up with an average of 65.2 minutes for clinic one and 75.4 for clinic two. That's the average of the averages obviously. And it looks like clinic two, clinic one is doing better. But let's grab this data. I'm just going to highlight it, control C to copy it, go here to the T test calculator and I'm going to paste that in over the sample data that's there and I'm just going to paste in the values. So now we've got our data in there and the calculator gives us those sample means again, clinic one and clinic two calculates the standard deviations and you can see there is some difference there, the amount of variation in each of those samples of 31 days. We're going to assume that there is no difference. That's the main difference is zero and leave it set at zero. And then we're going to use a 5% significance level. In other words, we want to be 95% confident that we're making the right decision about whether or not there's a significant difference. Now in this calculator, you can select, if you know that the variance is equal, you can select that. But in most cases, you don't know if the variances are equal and you haven't run a statistical test to find that out. So it's better just to assume that they're not equal. And then all the yellow cells here, the calculator calculates. Now I'm not going to go over everything here. Just to point out, it gives you the test statistics which shows that clinic two is 10 minutes, a little over 10 minutes faster, smaller average throughput time than clinic two. And we're going to jump down here to the two-tail test. And the upper and lower critical values, I'm not going to get into the theory here. We're just going to focus on the p-value. And remember the p-value is the probability of getting as extreme data or even more extreme data with our null assumption, in this case, no difference. So we're going to look at that p-value and it comes out to be .0429. So that is less than our 5%. So we would conclude there is a statistically significant difference in the throughput times of these two clinics. And of course, if you did a left-tail or right-tail, you can see we did a left-tail test that we assumed that one was faster. We got another significant result there. But if we assume that one was slower, we've got a non-significant. Hope this helps.