WEBVTT 1 00:00:00.720 --> 00:00:02.250 Hello students, 2 00:00:02.250 --> 00:00:07.250 and welcome to Biostats ER Chapter eight, Example nine. 3 00:00:08.070 --> 00:00:11.220 In this example, we will learn how to calculate sample size 4 00:00:11.220 --> 00:00:14.520 for two-independent samples dichotomous data 5 00:00:14.520 --> 00:00:16.623 to determine confidence interval. 6 00:00:17.580 --> 00:00:19.560 This problem is not from our textbook, 7 00:00:19.560 --> 00:00:23.490 and as always, I'm going to read the problem first. 8 00:00:23.490 --> 00:00:26.340 The issue here is that we have a treatment group 9 00:00:26.340 --> 00:00:30.330 and a placebo group who have reported on pain relief. 10 00:00:30.330 --> 00:00:34.200 In the placebo group, we have 20% pain relief reported, 11 00:00:34.200 --> 00:00:38.520 and in the treatment group we have 10% pain relief reported. 12 00:00:38.520 --> 00:00:42.240 Now we want to determine what sample size do we need 13 00:00:42.240 --> 00:00:45.100 to generate a 90% confidence interval 14 00:00:46.020 --> 00:00:48.370 for the difference between these two groups 15 00:00:49.470 --> 00:00:53.310 that is at most 5% from the true mean. 16 00:00:53.310 --> 00:00:54.870 So here I have summarized some 17 00:00:54.870 --> 00:00:56.913 of the information that is provided. 18 00:00:59.220 --> 00:01:04.220 And then I have provided you with the formula 19 00:01:04.590 --> 00:01:07.140 that we need to solve this problem. 20 00:01:07.140 --> 00:01:10.260 So once we start inserting the values here, 21 00:01:10.260 --> 00:01:13.020 P1 is 0.2. 22 00:01:13.020 --> 00:01:15.690 1 - P1 is 0.8. 23 00:01:15.690 --> 00:01:18.540 P2 is 0.1, 24 00:01:18.540 --> 00:01:21.510 and 1 - P2 is 0.9. 25 00:01:21.510 --> 00:01:23.320 And the Z value here is 26 00:01:26.323 --> 00:01:30.930 1.645 and the margin of error here is 0.05. 27 00:01:32.190 --> 00:01:34.650 So after we insert all these values 28 00:01:34.650 --> 00:01:37.083 and perform the calculation, 29 00:01:38.400 --> 00:01:41.760 our answer here is 270.6, 30 00:01:41.760 --> 00:01:45.390 and we always round up for sample size calculation 31 00:01:45.390 --> 00:01:47.250 and that gives us 271. 32 00:01:47.250 --> 00:01:52.143 So for each group, we need 271 subjects.