1 00:00:00,380 --> 00:00:04,790 Alright. So, why are statistical significance tests useful? So, 2 00:00:04,790 --> 00:00:07,210 they do provide a formalized framework for comparing and 3 00:00:07,210 --> 00:00:10,760 evaluating data. And they do enable us to evaluate 4 00:00:10,760 --> 00:00:13,470 whether perceived effects in our data set reflect differences across 5 00:00:13,470 --> 00:00:16,149 the whole population. They do not make a bad 6 00:00:16,149 --> 00:00:19,190 result look good. Significant sets are useful because they provide 7 00:00:19,190 --> 00:00:22,640 a formalized framework for comparing and evaluating data. Different 8 00:00:22,640 --> 00:00:26,500 tests have different assumptions and rules that they incorporate, and 9 00:00:26,500 --> 00:00:28,780 using a particular test ensures that everyone is on the 10 00:00:28,780 --> 00:00:30,990 same page in so far as what we're assuming about 11 00:00:30,990 --> 00:00:34,820 our data. Significance tests also enable us to evaluate whether 12 00:00:34,820 --> 00:00:37,550 perceived effects in our data set reflect differences across the whole 13 00:00:37,550 --> 00:00:40,796 population. As was the case with our company, where ten 14 00:00:40,796 --> 00:00:45,180 out of ten people polled preferred the color blue. Sometimes an 15 00:00:45,180 --> 00:00:47,770 affect that we seen in a small sample does not 16 00:00:47,770 --> 00:00:51,500 reflect what might be true across the entire population. A statistical 17 00:00:51,500 --> 00:00:54,420 significance test let's us formally determine whether or not this might 18 00:00:54,420 --> 00:00:57,940 be the case. Unfortunately, a bad result is not going to 19 00:00:57,940 --> 00:01:00,390 look any better or worse as a result of using a 20 00:01:00,390 --> 00:01:02,950 statistical significance test. If our data's 21 00:01:02,950 --> 00:01:05,010 bad, or there's really no difference 22 00:01:05,010 --> 00:01:07,930 between our two samples. We're not going to be able to 23 00:01:07,930 --> 00:01:11,470 undo that with a test. It is possible though that different tests 24 00:01:11,470 --> 00:01:14,890 might give us different results. The really important thing and we'll 25 00:01:14,890 --> 00:01:16,580 go into this a bit more is that you need to use 26 00:01:16,580 --> 00:01:19,290 the right test in the right situations. Why don't we talk a 27 00:01:19,290 --> 00:01:22,430 little bit about how we might actually run a statistical significance test.