1
00:00:04,650 --> 00:00:05,470
Hello, in this video,
2
00:00:05,470 --> 00:00:14,610
I want to talk with you just a little bit about statistics that we can measure and how we want to think about differences between them.
3
00:00:14,610 --> 00:00:21,220
In the previous video or an earlier video, I talked about how bar charts emphasized relative differences.
4
00:00:21,220 --> 00:00:24,450
So in this one, I'm going to talk just a little bit about what that means.
5
00:00:24,450 --> 00:00:30,040
So are learning outcomes for this video to review a little bit of the statistics or the metrics that we've been talking about,
6
00:00:30,040 --> 00:00:37,990
to talk for you to be able to compute absolute and relative differences and to interpret a relative difference between two quantities.
7
00:00:37,990 --> 00:00:43,870
So as we've talked about, we can compute various statistics over our data means median mode success rates.
8
00:00:43,870 --> 00:00:46,720
We can compute percentiles, counts many.
9
00:00:46,720 --> 00:00:57,530
Basically, any statistic you can think of for that, you can compute over a of a set of numbers we can use as some kind of a statistic or a metric.
10
00:00:57,530 --> 00:01:04,210
And this often serves as the metric for analysis or evaluation. Try to evaluate a program or tried to evaluate a technology.
11
00:01:04,210 --> 00:01:08,920
We have some metric that is measuring its effectiveness.
12
00:01:08,920 --> 00:01:15,310
And we want to see whether it's improved or changed somehow.
13
00:01:15,310 --> 00:01:19,090
So when we compare to values, though, with a few different ways to do it.
14
00:01:19,090 --> 00:01:24,040
So let's take a couple the population estimate in 2018 of Boise in Salt Lake City.
15
00:01:24,040 --> 00:01:27,230
And there's two ways that we few different ways that we can compare it.
16
00:01:27,230 --> 00:01:33,460
Two of them are the absolute difference. Boise has twenty eight thousand more people than Salt Lake City.
17
00:01:33,460 --> 00:01:42,180
The other is the relative difference. Boise has fourteen point five percent or fourteen percent more people than Salt Lake City.
18
00:01:42,180 --> 00:01:49,020
So the absolute value is the difference between two values is actually the absolute value of the difference between two values.
19
00:01:49,020 --> 00:01:54,750
We can also talk about a science difference or a real difference where we don't have the absolute value if we need the direction on the difference.
20
00:01:54,750 --> 00:02:01,440
That becomes useful. But what we're talking about here is the actual difference in the underlying units.
21
00:02:01,440 --> 00:02:11,940
So in our example case, number of people. And but the another way we can do it is to talk about the relative difference.
22
00:02:11,940 --> 00:02:18,960
So this is the it's the difference normalized by.
23
00:02:18,960 --> 00:02:22,440
The reference quantity, and we have to be clear on which ones,
24
00:02:22,440 --> 00:02:29,700
the reference quantity and the reference quantity is the one we're starting from whom we're computing the relative difference.
25
00:02:29,700 --> 00:02:39,990
So, for example, 50 is 25 percent, more than 40 because you take 25 percent of fortius 10 add add that and you get 50.
26
00:02:39,990 --> 00:02:46,410
But 40 is 20 percent is only 20 percent less than 50 because you take 50.
27
00:02:46,410 --> 00:02:50,670
10 is 20 percent a 50, whereas it's 25 percent of 40.
28
00:02:50,670 --> 00:02:56,750
You subtract it. And you get and you get 40.
29
00:02:56,750 --> 00:03:01,010
So this different this order difference is really, really important.
30
00:03:01,010 --> 00:03:09,480
So for the 50, what we've got is we have 50 minus 40, over 40.
31
00:03:09,480 --> 00:03:20,010
That's this one. And we have for the minus 50, over 50, negative 10, over 50.
32
00:03:20,010 --> 00:03:25,080
That's 20 percent. Ten over 40 is 25 percent.
33
00:03:25,080 --> 00:03:33,330
So we need to be really, really careful about the order, another way we can compares with the ratio, we just divide one quantity by the other.
34
00:03:33,330 --> 00:03:40,470
So this is when we say slike this year's sales of 20 million are twice as much as last year's 10 million.
35
00:03:40,470 --> 00:03:47,010
This is an apt. It's an absolute change of one of 10 million and it's a relative change of one hundred percent.
36
00:03:47,010 --> 00:03:51,030
Twenty million is twice as much as one is 10 million.
37
00:03:51,030 --> 00:03:54,090
And it is one hundred percent higher than 10 million.
38
00:03:54,090 --> 00:03:59,940
Now, one thing to think about, if I say this year's returns are two times larger than last year's.
39
00:03:59,940 --> 00:04:03,750
What does that mean? Does it mean it's two times?
40
00:04:03,750 --> 00:04:08,100
Does it mean it's 200 percent more, which would be three times?
41
00:04:08,100 --> 00:04:13,350
Is it clear? I would submit that this way of framing it is ambiguous.
42
00:04:13,350 --> 00:04:20,400
And so we should avoid it. The appropriate comparison really depends on context and problem.
43
00:04:20,400 --> 00:04:25,320
There's not a hard and fast rule when you need one or another.
44
00:04:25,320 --> 00:04:31,740
Relative comparisons are quite common because they they can be compared across a variety of contexts.
45
00:04:31,740 --> 00:04:36,570
But we still also need to pay attention to the underlying absolute difference in what
46
00:04:36,570 --> 00:04:41,190
the act what this what the change being made in this relative change actually is.
47
00:04:41,190 --> 00:04:43,850
One example of a high profile relative change.
48
00:04:43,850 --> 00:04:53,940
If the Netflix prize, which was a run by Netflix a number of years ago, they paid a million dollars to the team that was able to beat It's The Beat,
49
00:04:53,940 --> 00:04:58,740
their internal movie recommender on the metric that they chose by 10 percent.
50
00:04:58,740 --> 00:05:06,120
They wanted a 10 percent improvement. And this metric it was. Lower is better, so they wanted you to.
51
00:05:06,120 --> 00:05:10,380
They wanted to decrease in the metric by 10 percent.
52
00:05:10,380 --> 00:05:17,340
We can also talk about a difference between differences, because if we compute a difference, that difference itself is just another value.
53
00:05:17,340 --> 00:05:22,290
So we could say ten sales grew 10 percent more this year than last year.
54
00:05:22,290 --> 00:05:32,200
So if we define growth as the one year sales minus the other year sales, then we can look at the growth of this year.
55
00:05:32,200 --> 00:05:37,000
And the growth of last year and we can compute the difference and difference.
56
00:05:37,000 --> 00:05:44,080
And so we can have a 10 percent increase in growth. Difference in difference has come up a lot in various contexts.
57
00:05:44,080 --> 00:05:50,020
And so it's important to be able to reason about those as well. And again, be clear both in writing and understanding.
58
00:05:50,020 --> 00:05:58,120
So we don't want we don't want to. Visual comparison bar charts emphasized relative difference because the height of the bar is right there.
59
00:05:58,120 --> 00:06:04,120
And the eye very naturally compares the difference between bars to the height of the bar itself.
60
00:06:04,120 --> 00:06:10,840
Point plots emphasize absolute difference because you don't have the reference point of the size of the bar.
61
00:06:10,840 --> 00:06:17,170
They're both of them. Make it pretty clear to see the also compare the differences.
62
00:06:17,170 --> 00:06:23,590
See how different the differences are. Those are of evident both in bar charts and point plots.
63
00:06:23,590 --> 00:06:28,690
So to wrap up, there are three primary ways to compute statistics, absolute relative and ratio.
64
00:06:28,690 --> 00:06:32,470
You need to be very clear and unambiguous when you're writing the results of a
65
00:06:32,470 --> 00:06:36,370
comparison and also when you're trying to understand what others have written.
66
00:06:36,370 --> 00:06:39,340
Seek to accurately understand it. And if you're providing feeB,
67
00:06:39,340 --> 00:06:50,900
if you're in a context where you're providing feedback and it's not clear that clarity is something you want to ask for revision.