1
00:00:00,107 --> 00:00:03,926
♪ [音乐] ♪

2
00:00:20,880 --> 00:00:22,077
- [Thomas Stratmann] 大家好！

3
00:00:22,077 --> 00:00:24,268
在接下来的一系列视频中

4
00:00:24,268 --> 00:00:26,858
我们将向你介绍一个
炫酷的新工具

5
00:00:26,858 --> 00:00:30,414
来帮助你理解数据

6
00:00:30,414 --> 00:00:31,981
那就是“线性回归”

7
00:00:32,885 --> 00:00:34,668
假设你有这么一种理论

8
00:00:34,668 --> 00:00:37,249
你发现外貌出众的人

9
00:00:37,249 --> 00:00:39,067
好像总能得到特殊的优待

10
00:00:39,642 --> 00:00:40,878
你在想

11
00:00:40,878 --> 00:00:43,798
“还有什么地方
也能看到这种现象呢？”

12
00:00:44,132 --> 00:00:45,637
对教师来说
这种现象也存在吗？

13
00:00:45,637 --> 00:00:48,259
有没有可能
外貌出众的老师

14
00:00:48,259 --> 00:00:50,010
也会得到特别优待呢？

15
00:00:50,350 --> 00:00:53,899
学生们会不会
对这些老师更好

16
00:00:53,899 --> 00:00:57,209
给他们打更高的学生评价分？

17
00:00:57,866 --> 00:01:00,467
如果确实如此
外貌对评价分的影响

18
00:01:00,467 --> 00:01:03,573
是很大还是很小呢？

19
00:01:04,159 --> 00:01:07,519
假设有位教师刚刚开始
到一所大学上班

20
00:01:07,519 --> 00:01:08,759
- [男人] 老兄，早啊

21
00:01:08,759 --> 00:01:11,810
- 仅从他的外貌
我们能对他的学生评价分

22
00:01:11,810 --> 00:01:13,371
做出怎样的预测？

23
00:01:13,940 --> 00:01:17,216
由于评价分能影响加薪

24
00:01:17,671 --> 00:01:21,709
如果这个理论属实
老师们可能会

25
00:01:21,709 --> 00:01:24,519
采取一些令人惊讶的手段
来提高他们的得分

26
00:01:24,519 --> 00:01:25,731
- [Lloyd Christmas] 耶！

27
00:01:25,731 --> 00:01:27,461
- 如果你想弄清楚

28
00:01:27,461 --> 00:01:30,801
更出众的外貌是否真的会
带来更高的评价分

29
00:01:31,441 --> 00:01:34,450
你会怎样检验这个假说呢？

30
00:01:34,956 --> 00:01:36,552
你可以收集数据

31
00:01:36,761 --> 00:01:40,025
首先，让学生从1到10

32
00:01:40,025 --> 00:01:42,076
给老师的外貌打分

33
00:01:42,076 --> 00:01:44,807
由此你可以得出
这位老师的颜值平均分

34
00:01:45,229 --> 00:01:48,552
然后你可以从25名学生处

35
00:01:48,552 --> 00:01:50,421
收集这位老师的教学评价分

36
00:01:50,421 --> 00:01:53,273
我们通过散点图

37
00:01:53,273 --> 00:01:54,738
来同时查看这两个变量

38
00:01:54,981 --> 00:01:57,419
我们用横轴表示外貌

39
00:01:57,852 --> 00:02:00,589
纵轴表示教学评价分

40
00:02:01,223 --> 00:02:04,903
例如，这一点代表着 Peate 教授

41
00:02:04,903 --> 00:02:06,423
- [Bib Fortuna] De wana wanga.

42
00:02:06,423 --> 00:02:08,811
- 他得到了3分的外貌分

43
00:02:08,811 --> 00:02:11,866
8.425的教学评价分

44
00:02:12,084 --> 00:02:14,958
这边的是 Helmchen 教授

45
00:02:14,958 --> 00:02:16,797
- [Ben Stiller, "Zoolander"]
帅到不像话！

46
00:02:16,797 --> 00:02:18,721
- 他的外貌得分非常高

47
00:02:18,721 --> 00:02:20,872
但评价分没那么高

48
00:02:21,101 --> 00:02:22,283
你能看出规律吗？

49
00:02:22,283 --> 00:02:25,533
当我们沿x轴从左向右移动

50
00:02:25,533 --> 00:02:27,963
从难看向好看移动

51
00:02:27,963 --> 00:02:31,186
评价分呈现出上升趋势

52
00:02:31,870 --> 00:02:35,174
对了，我们在这个系列视频中
使用的数据

53
00:02:35,174 --> 00:02:38,923
不是编造出来的
而是来自于

54
00:02:38,923 --> 00:02:40,897
在德克萨斯大学做过的
真实研究

55
00:02:41,337 --> 00:02:46,023
另外你可能不知道
“pulchritude”只不过是

56
00:02:46,023 --> 00:02:47,880
“颜值”的另一种
比较高端、学术的说法

57
00:02:48,405 --> 00:02:51,474
有些时候

58
00:02:51,474 --> 00:02:55,594
用散点图很难判断出
两个变量之间的确切关系

59
00:02:55,594 --> 00:02:59,104
尤其是随着我们
从左向右移动

60
00:02:59,104 --> 00:03:01,318
数值的波动很大的时候

61
00:03:02,000 --> 00:03:04,908
处理这种波动的一种方法是

62
00:03:04,908 --> 00:03:08,144
画一条直线
穿过这团数据

63
00:03:08,144 --> 00:03:10,775
让这条直线

64
00:03:10,775 --> 00:03:12,613
尽可能贴切地概括这些数据

65
00:03:13,295 --> 00:03:17,181


66
00:03:17,669 --> 00:03:20,888


67
00:03:20,888 --> 00:03:24,278


68
00:03:24,278 --> 00:03:26,456


69
00:03:27,087 --> 00:03:29,536


70
00:03:30,067 --> 00:03:32,596


71
00:03:32,596 --> 00:03:35,358


72
00:03:36,107 --> 00:03:39,827


73
00:03:40,794 --> 00:03:43,807


74
00:03:43,807 --> 00:03:45,587


75
00:03:46,070 --> 00:03:50,237


76
00:03:50,237 --> 00:03:53,026


77
00:03:53,544 --> 00:03:55,907


78
00:03:55,907 --> 00:03:59,469


79
00:03:59,768 --> 00:04:03,939


80
00:04:04,377 --> 00:04:07,420


81
00:04:07,857 --> 00:04:10,764


82
00:04:11,158 --> 00:04:13,903


83
00:04:14,389 --> 00:04:17,304


84
00:04:17,770 --> 00:04:21,262


85
00:04:21,579 --> 00:04:25,569


86
00:04:25,569 --> 00:04:28,429


87
00:04:28,429 --> 00:04:30,229


88
00:04:30,229 --> 00:04:31,297


89
00:04:31,297 --> 00:04:34,109


90
00:04:34,683 --> 00:04:36,749


91
00:04:37,019 --> 00:04:38,749


92
00:04:39,233 --> 00:04:41,665


93
00:04:41,665 --> 00:04:43,515


94
00:04:44,844 --> 00:04:47,890


95
00:04:47,890 --> 00:04:49,770


96
00:04:49,770 --> 00:04:52,039


97
00:04:52,838 --> 00:04:55,439


98
00:04:55,439 --> 00:04:58,340


99
00:04:58,833 --> 00:05:00,430


100
00:05:00,430 --> 00:05:03,639


101
00:05:03,639 --> 00:05:06,900


102
00:05:07,344 --> 00:05:09,965


103
00:05:09,965 --> 00:05:12,456


104
00:05:12,456 --> 00:05:15,645


105
00:05:16,052 --> 00:05:18,956


106
00:05:19,228 --> 00:05:22,972


107
00:05:23,660 --> 00:05:27,668


108
00:05:28,080 --> 00:05:32,095


109
00:05:32,095 --> 00:05:34,335


110
00:05:34,335 --> 00:05:37,383


111
00:05:37,861 --> 00:05:40,388


112
00:05:40,388 --> 00:05:43,537


113
00:05:43,537 --> 00:05:45,848


114
00:05:46,346 --> 00:05:49,807


115
00:05:50,289 --> 00:05:53,620


116
00:05:53,620 --> 00:05:54,900


117
00:05:54,900 --> 00:05:58,166


118
00:05:58,922 --> 00:06:02,069


119
00:06:02,069 --> 00:06:05,781


120
00:06:06,626 --> 00:06:09,575


121
00:06:09,846 --> 00:06:14,577


122
00:06:14,577 --> 00:06:18,994


123
00:06:19,408 --> 00:06:21,758


124
00:06:21,758 --> 00:06:24,477


125
00:06:24,477 --> 00:06:26,219


126
00:06:26,219 --> 00:06:28,368


127
00:06:28,368 --> 00:06:30,737


128
00:06:31,762 --> 00:06:35,509


129
00:06:35,509 --> 00:06:39,070


130
00:06:41,169 --> 00:06:42,445


131
00:06:42,445 --> 00:06:45,247


132
00:06:45,568 --> 00:06:47,139


133
00:06:47,139 --> 00:06:48,700


134
00:06:48,700 --> 00:06:50,404


135
00:06:50,865 --> 00:06:53,976


136
00:06:54,313 --> 00:06:55,364


137
00:06:55,598 --> 00:06:58,325


138
00:06:58,325 --> 00:07:01,642


139
00:07:01,892 --> 00:07:04,406