 If you find this video helpful, please consider subscribing to my YouTube channel, TheStatsFiles. Just look for the big red button. Hi, this is Dr. Don. I have a problem out of Chapter 8 when we're dealing with two sample hypothesis tests. In this particular problem, though, we're given summary data here, the sample means, the sample standard deviations, and the sample sizes for two samples. And we're asked to construct a 95% confidence interval for the mean difference, mean one minus mean two, with the sample statistics from here. Okay, we're told to assume the populations are approximately normal, which almost always is true in intro statistics. And we're also told that we should assume equal variances, which is important. Now they give us these big, long, complex formulas, and you can crunch through those and get the right answers, but it's a whole lot easier to let StatCrunch do that for you. You can go up to question help, and then click on StatCrunch, open StatCrunch, or you can navigate to statcrunch.com and open it that way. Okay, I'm over here in StatCrunch, and we're going to go, as we usually do, to Stat. This is T stats. Now why are we going to T stats? Let's look back at our problem. We're not given the population standard deviation sigmas. We're given S, the sample standard deviation. That's one clue. But if you look in the formula they give you, they show you T critical. That's another clue you use the T. But the main one you should catch is that you don't know the population standard deviation sigmas. So we go back over here to StatCrunch, and we go to T stats. We've got two samples, and we've got summary data. So we bring up our dialog box, and we're going to enter this data. One, eight, two, four, 18. And we go down to the next part, and this is the critical part. It says, note the default was changed to off, which means it's not going to pool variances. You pool variances if you have equal variances. So we do, so we need to check that. Then we go down. We want the confidence interval. It's all we have to do. We want the 95, so it's already got 0.95 in there, and I'm just going to click compute. And we bring up our answers here. We've got, again, our main difference, which was 149. We've got our standard error calculated for us, the degrees of freedom calculated. But there they are, the lower limit, 4.85, upper limit, 293.15. If we look over here, they say round to the nearest integer, and in my StatLab it's critical. You give them exactly what they ask for. So if we round 4.8 to the nearest integer, that would be 5, and 293.15 rounded would be 293, which is the values they have here. Hope this helps. And if it does help, please consider subscribing to my YouTube channel, the stats files. Just click the big red subscribe button.