Would you like to inspect the original subtitles? These are the user uploaded subtitles that are being translated:
1
00:00:00,004 --> 00:00:01,006
- [Instructor] As much as we like to think
2
00:00:01,006 --> 00:00:04,000
that statistics is an exact science,
3
00:00:04,000 --> 00:00:06,009
in fact, there's a lot of room for uncertainty.
4
00:00:06,009 --> 00:00:07,008
In this movie,
5
00:00:07,008 --> 00:00:11,003
I want to go over some of the limits of hypothesis testing
6
00:00:11,003 --> 00:00:15,006
to give you some cautions about your data and your analysis.
7
00:00:15,006 --> 00:00:16,008
The first is that,
8
00:00:16,008 --> 00:00:19,007
even if your analysis meets the 95% standard
9
00:00:19,007 --> 00:00:21,002
that most analysts use,
10
00:00:21,002 --> 00:00:24,000
you will still be wrong one out of 20 times.
11
00:00:24,000 --> 00:00:26,004
And if you've ever performed any type of analysis
12
00:00:26,004 --> 00:00:27,006
or even flipped a coin
13
00:00:27,006 --> 00:00:30,001
and gotten four or five tails in a row,
14
00:00:30,001 --> 00:00:33,000
then you know how likely it is that events can happen
15
00:00:33,000 --> 00:00:38,001
one out of 16, one out of 20, or even one out of 100 times.
16
00:00:38,001 --> 00:00:39,008
Also, you might run into errors
17
00:00:39,008 --> 00:00:41,007
in data collection or analysis
18
00:00:41,007 --> 00:00:44,003
that can skew your results.
19
00:00:44,003 --> 00:00:45,006
Some errors in data collection
20
00:00:45,006 --> 00:00:48,000
are not randomized during your samples
21
00:00:48,000 --> 00:00:50,002
or using old data.
22
00:00:50,002 --> 00:00:52,008
So make sure that your data is collected properly
23
00:00:52,008 --> 00:00:55,005
and that it isn't too old
24
00:00:55,005 --> 00:00:57,004
and that circumstances haven't changed
25
00:00:57,004 --> 00:01:00,001
to make the analysis invalid.
26
00:01:00,001 --> 00:01:03,004
And also, remember that correlation is not causation.
27
00:01:03,004 --> 00:01:05,008
Just because two things happen together
28
00:01:05,008 --> 00:01:08,009
doesn't mean that one caused the other.
29
00:01:08,009 --> 00:01:13,008
That said, correlation does tell you where to start looking.
30
00:01:13,008 --> 00:01:15,006
If two things happen together
31
00:01:15,006 --> 00:01:18,001
and it's plausible that there is a way
32
00:01:18,001 --> 00:01:20,001
that one could have caused the other,
33
00:01:20,001 --> 00:01:22,009
then that's a great place to start looking.
34
00:01:22,009 --> 00:01:25,000
As always, this is where your intuition
35
00:01:25,000 --> 00:01:28,000
and knowledge of the business will help out a lot.
2706
Can't find what you're looking for?
Get subtitles in any language from opensubtitles.com, and translate them here.