ABOUT THE SPEAKER
Emily Oster - Assumption-busting economist
Emily Oster, a University of Chicago economist, uses the dismal science to rethink conventional wisdom, from her Harvard doctoral thesis that took on famed economist Amartya Sen to her recent work debunking assumptions on HIV prevalence in Africa.

Why you should listen

Emily Oster, an Assistant Professor of Economics at the University of Chicago, has a history of rethinking conventional wisdom.

Her Harvard doctoral thesis took on famed economist Amartya Sen and his claim that 100 million women were statistically missing from the developing world. He blamed misogynist medical care and outright sex-selective abortion for the gap, but Oster pointed to data indicating that in countries where Hepetitis B infections were higher, more boys were born. Through her unorthodox analysis of medical data, she accounted for 50% of the missing girls. Three years later, she would publish another paper amending her findings, stating that, after further study, the relationship between Hepetitis B and missing women was not apparent. This concession, along with her audacity to challenge economic assumptions and her dozens of other influential papers, has earned her the respect of the global academic community. 

She's also investigated the role of bad weather in the rise in witchcraft trials in Medieval Europe and what drives people to play the Powerball lottery. Her latest target: busting assumptions on HIV in Africa.

And she's an advice columnist too >>

 

More profile about the speaker
Emily Oster | Speaker | TED.com
TED2007

Emily Oster: Flip your thinking on AIDS in Africa

艾米莉·奥斯特颠覆我们对非洲艾滋病情况的认知

Filmed:
921,618 views

艾米莉·奥斯特从经济学的角度重新分析了非洲艾滋病情况的数据,并得出了一个惊人的结论:所有我们知道的关于艾滋病在陆地上的传播情况全是错的。
- Assumption-busting economist
Emily Oster, a University of Chicago economist, uses the dismal science to rethink conventional wisdom, from her Harvard doctoral thesis that took on famed economist Amartya Sen to her recent work debunking assumptions on HIV prevalence in Africa. Full bio

Double-click the English transcript below to play the video.

00:26
So I want to talk to you today今天 about AIDS艾滋病 in sub-Saharan撒哈拉以南 Africa非洲.
0
1000
3000
我今天要谈论的是在撒哈拉沙漠以南的非洲那里艾滋病的情况
00:29
And this is a pretty漂亮 well-educated受过良好教育 audience听众,
1
4000
2000
我知道在场的观众都受过很良好的教育
00:31
so I imagine想像 you all know something about AIDS艾滋病.
2
6000
3000
所以我可以猜想到你们都了解一些关于艾滋病的事情
00:34
You probably大概 know that roughly大致 25 million百万 people in Africa非洲
3
9000
2000
你可能知道在非洲大约有两千五百万人
00:36
are infected感染 with the virus病毒, that AIDS艾滋病 is a disease疾病 of poverty贫穷,
4
11000
4000
感染了这种病毒,艾滋病是一种因贫穷而带来的疾病
00:40
and that if we can bring带来 Africa非洲 out of poverty贫穷, we would decrease减少 AIDS艾滋病 as well.
5
15000
4000
如果我们能让非洲摆脱贫困,我们就能同样减轻艾滋病疫情
00:44
If you know something more, you probably大概 know that Uganda乌干达, to date日期,
6
19000
3000
如果你了解足够多的话,你可能知道迄今为止乌干达
00:47
is the only country国家 in sub-Saharan撒哈拉以南 Africa非洲
7
22000
2000
是唯一一个在撒哈拉以南非洲的国家中
00:49
that has had success成功 in combating打击 the epidemic疫情.
8
24000
3000
成功防治艾滋病
00:52
Using运用 a campaign运动 that encouraged鼓励 people to abstain避免, be faithful可信, and use condoms避孕套 --
9
27000
4000
启用了鼓励人们禁欲,忠贞,并使用安全套的运动
00:56
the ABCABC campaign运动 -- they decreased下降 their prevalence流行 in the 1990s
10
31000
4000
被称为ABC准则。在九十年代它们大大减少了艾滋病的传播
01:00
from about 15 percent百分 to 6 percent百分 over just a few少数 years年份.
11
35000
4000
仅仅几年内就从百分之15减到了百分之6
01:04
If you follow跟随 policy政策, you probably大概 know that a few少数 years年份 ago
12
39000
3000
如果你关心时政,你可能知道几年前
01:07
the president主席 pledged承诺 15 billion十亿 dollars美元 to fight斗争 the epidemic疫情 over five years年份,
13
42000
4000
总统承诺动用150亿美元在五年间抗击疫情
01:11
and a lot of that money is going to go to programs程式 that try to replicate复制 Uganda乌干达
14
46000
3000
一大部分钱都会被用来启动一些项目试图仿造乌干达
01:14
and use behavior行为 change更改 to encourage鼓励 people and decrease减少 the epidemic疫情.
15
49000
6000
用行为变革来引导人们并减少疫情。
01:20
So today今天 I'm going to talk about some things
16
55000
2000
然而今天我想讲一些
01:22
that you might威力 not know about the epidemic疫情,
17
57000
2000
你们可能不知道的事情
01:24
and I'm actually其实 also going to challenge挑战
18
59000
2000
事实上,我同时会颠覆
01:26
some of these things that you think that you do know.
19
61000
2000
一些你们自以为知道的事情
01:28
To do that I'm going to talk about my research研究
20
63000
3000
要做到这些我就要讲一下
01:31
as an economist经济学家 on the epidemic疫情.
21
66000
2000
我作为一个经济学家所做的研究。
01:33
And I'm not really going to talk much about the economy经济.
22
68000
2000
我不是真的要谈经济
01:35
I'm not going to tell you about exports出口 and prices价格.
23
70000
3000
我不会谈论什么出口和价格。
01:38
But I'm going to use tools工具 and ideas思路 that are familiar to economists经济学家
24
73000
4000
但我会用一些经济学家们熟悉的工具和思想
01:42
to think about a problem问题 that's more traditionally传统
25
77000
2000
去思考一个习惯上被认为是
01:44
part部分 of public上市 health健康 and epidemiology流行病学.
26
79000
2000
在公共卫生和流行病学领域的问题
01:46
And I think in that sense, this fits适合 really nicely很好 with this lateral thinking思维 idea理念.
27
81000
4000
在这个意义上,这真的很符合横向思维的方式
01:50
Here I'm really using运用 the tools工具 of one academic学术的 discipline学科
28
85000
3000
我会用某个学科的一些工具
01:53
to think about problems问题 of another另一个.
29
88000
2000
来思考另一个学科一些问题
01:55
So we think, first and foremost最重要的是, AIDS艾滋病 is a policy政策 issue问题.
30
90000
3000
我想首先来说,艾滋病是一个政策的问题
01:58
And probably大概 for most people in this room房间, that's how you think about it.
31
93000
3000
可能对于在这个房里的大多人来说,你们就是这么想的。
02:01
But this talk is going to be about understanding理解 facts事实 about the epidemic疫情.
32
96000
4000
但这次的演讲是有关于理解疫情传播的实事
02:05
It's going to be about thinking思维 about how it evolves演变, and how people respond响应 to it.
33
100000
3000
有关于思考它如何逐渐形成,人们又是如何应对的
02:08
I think it may可能 seem似乎 like I'm ignoring无视 the policy政策 stuff东东,
34
103000
3000
我可能会忽略政策一类的东西
02:11
which哪一个 is really the most important重要,
35
106000
2000
实事它们是最重要的
02:13
but I'm hoping希望 that at the end结束 of this talk you will conclude得出结论
36
108000
2000
我希望在演讲完毕你们能知道
02:15
that we actually其实 cannot不能 develop发展 effective有效 policy政策
37
110000
2000
我们无法实施有效的政策
02:17
unless除非 we really understand理解 how the epidemic疫情 works作品.
38
112000
3000
除非我们明白的疫情是如何发生的
02:20
And the first thing that I want to talk about,
39
115000
2000
我想讲的第一点
02:22
the first thing I think we need to understand理解 is:
40
117000
2000
我认为我们必须理解的第一点是
02:24
how do people respond响应 to the epidemic疫情?
41
119000
2000
人们如何应对这种流行病的
02:26
So AIDS艾滋病 is a sexually transmitted发送 infection感染, and it kills杀死 you.
42
121000
4000
艾滋病是经由性传播感染的,它足以致命。
02:30
So this means手段 that in a place地点 with a lot of AIDS艾滋病,
43
125000
2000
也就是说在一个艾滋病泛滥的地方
02:32
there's a really significant重大 cost成本 of sex性别.
44
127000
2000
性交带来的代价也十分大
02:34
If you're an uninfected未感染 man living活的 in Botswana博茨瓦纳, where the HIVHIV rate is 30 percent百分,
45
129000
4000
在博茨瓦那艾滋病感病率是百分之30,如果你是此处的未感染者
02:38
if you have one more partner伙伴 this year -- a long-term长期 partner伙伴, girlfriend女朋友, mistress情妇 --
46
133000
4000
如果你今年将多一个长期的伴侣,或许是女友,或许是情妇
02:42
your chance机会 of dying垂死 in 10 years年份 increases增加 by three percentage百分比 points.
47
137000
4000
在十年内可能死去的概率会提高三个百分点
02:46
That is a huge巨大 effect影响.
48
141000
2000
这种影响是巨大的
02:48
And so I think that we really feel like then people should have less sex性别.
49
143000
3000
所以我们会意识到,人们真的需要减少性行为。
02:51
And in fact事实 among其中 gay同性恋者 men男人 in the US
50
146000
2000
事实上在美国的男同性恋中
02:53
we did see that kind of change更改 in the 1980s.
51
148000
2000
在八十年代我们确实看到了这种改变
02:55
So if we look in this particularly尤其 high-risk高风险 sample样品, they're being存在 asked,
52
150000
4000
如果我们调查这个独特的高危人群,当他们被问到
02:59
"Did you have more than one unprotected无保护 sexual有性 partner伙伴 in the last two months个月?"
53
154000
3000
在过去的两个月里你是否有多于一个未采取保护措施的性伴侣时
03:02
Over a period from '84 to '88, that share分享 drops滴剂 from about 85 percent百分 to 55 percent百分.
54
157000
6000
数据表明从84到88年,比例从百分之85下降到百分之55
03:08
It's a huge巨大 change更改 in a very short period of time.
55
163000
2000
在如此短的时间里,这是一个巨大的变化。
03:10
We didn't see anything like that in Africa非洲.
56
165000
2000
在非洲我们从未看到这样的改变。
03:12
So we don't have quite相当 as good data数据, but you can see here
57
167000
3000
我们没有这么好的数据,但你会发现
03:15
the share分享 of single men男人 having pre-marital婚前 sex性别,
58
170000
2000
单身男人婚前性行为
03:17
or married已婚 men男人 having extra-marital婚外 sex性别,
59
172000
2000
或者已婚男人的婚外性行为的数据比例
03:19
and how that changes变化 from the early '90s to late晚了 '90s,
60
174000
3000
如何从九十年代初到九十年代末,
03:22
and late晚了 '90s to early 2000s. The epidemic疫情 is getting得到 worse更差.
61
177000
3000
从九十年代末到二十世纪初是如何变化的。疫情的传播变得更加严重。
03:25
People are learning学习 more things about it.
62
180000
2000
人们对疾病的了解更多了
03:27
We see almost几乎 no change更改 in sexual有性 behavior行为.
63
182000
2000
但在性行为上却几乎没发生变化。
03:29
These are just tiny decreases降低 -- two percentage百分比 points -- not significant重大.
64
184000
4000
仅仅有两个百分点的微小减弱
03:33
This seems似乎 puzzling令人费解. But I'm going to argue争论 that you shouldn't不能 be surprised诧异 by this,
65
188000
4000
这似乎很令人困惑,但我要说你不应该对此感到吃惊。
03:37
and that to understand理解 this you need to think about health健康
66
192000
3000
想要理解这个,你需要用经济学家的思维
03:40
the way than an economist经济学家 does -- as an investment投资.
67
195000
3000
来考虑有关健康的事 用一种投资的思想
03:43
So if you're a software软件 engineer工程师 and you're trying to think about
68
198000
3000
如果你是一个软件工程师 当你在思考
03:46
whether是否 to add some new functionality功能 to your program程序,
69
201000
3000
在你的设计里是否要加一些新的功能时
03:49
it's important重要 to think about how much it costs成本.
70
204000
2000
要花费多少成本是必须要考虑的。
03:51
It's also important重要 to think about what the benefit效益 is.
71
206000
2000
能带来多少利益也是必须要考虑的。
03:53
And one part部分 of that benefit效益 is how much longer
72
208000
2000
其中有一点就是你认为
03:55
you think this program程序 is going to be active活性.
73
210000
2000
这个设计会在多长时间内有效。
03:57
If version 10 is coming未来 out next下一个 week,
74
212000
2000
如果第十版会在下周发行,
03:59
there's no point in adding加入 more functionality功能 into version nine.
75
214000
3000
就没有必要在第九版中再加功能了。
04:02
But your health健康 decisions决定 are the same相同.
76
217000
2000
而你的健康选择也是如此。
04:04
Every一切 time you have a carrot胡萝卜 instead代替 of a cookie曲奇饼,
77
219000
2000
每一次你放弃饼干而去吃胡萝卜时
04:06
every一切 time you go to the gym健身房 instead代替 of going to the movies电影,
78
221000
3000
每一次你去健身中心而不是去电影院时
04:09
that's a costly昂贵 investment投资 in your health健康.
79
224000
2000
这就是对你健康的极大投资。
04:11
But how much you want to invest投资 is going to depend依靠
80
226000
2000
但你要投资多少就取决于
04:13
on how much longer you expect期望 to live生活 in the future未来,
81
228000
2000
你预期能活多久
04:15
even if you don't make those investments投资.
82
230000
2000
就算你不做这些投资
04:17
AIDS艾滋病 is the same相同 kind of thing. It's costly昂贵 to avoid避免 AIDS艾滋病.
83
232000
3000
艾滋病也依旧存在 避免艾滋是花费巨大的
04:20
People really like to have sex性别.
84
235000
3000
人们真的很喜欢做爱。
04:23
But, you know, it has a benefit效益 in terms条款 of future未来 longevity长寿.
85
238000
6000
就未来长寿而言,它确实有好处。
04:29
But life expectancy期待 in Africa非洲, even without AIDS艾滋病, is really, really low:
86
244000
4000
但在非洲的平均寿命,即使没有艾滋病,依旧非常非常低:
04:33
40 or 50 years年份 in a lot of places地方.
87
248000
3000
在大部分地方是40到50岁。
04:36
I think it's possible可能, if we think about that intuition直觉, and think about that fact事实,
88
251000
4000
可能的话,我们用直觉想想,结合实事想想,
04:40
that maybe that explains说明 some of this low behavior行为 change更改.
89
255000
3000
这或许能解释一部分这种低行为变革的原因。
04:43
But we really need to test测试 that.
90
258000
2000
但我们急需检测一下。
04:45
And a great way to test测试 that is to look across横过 areas in Africa非洲 and see:
91
260000
3000
一种很好的检测方法就是调查非洲的各个地区:
04:48
do people with more life expectancy期待 change更改 their sexual有性 behavior行为 more?
92
263000
4000
在平均寿命高的地方人们改变性行为的习惯是不是多一些?
04:52
And the way that I'm going to do that is,
93
267000
2000
我想做的就是
04:54
I'm going to look across横过 areas with different不同 levels水平 of malaria疟疾.
94
269000
3000
去调查有着不同疟疾疫情的地区。
04:57
So malaria疟疾 is a disease疾病 that kills杀死 you.
95
272000
3000
疟疾是一种致命的疾病。
05:00
It's a disease疾病 that kills杀死 a lot of adults成年人 in Africa非洲, in addition加成 to a lot of children孩子.
96
275000
3000
在非洲它使无数的成人和儿童失去生命。
05:03
And so people who live生活 in areas with a lot of malaria疟疾
97
278000
3000
那住在疟疾病率高的地方的人们
05:06
are going to have lower降低 life expectancy期待 than people who live生活 in areas with limited有限 malaria疟疾.
98
281000
4000
将比疟疾爆发不严重地方的平均寿命要低一些
05:10
So one way to test测试 to see whether是否 we can explain说明
99
285000
2000
所以检验的一种方法就是我们是否
05:12
some of this behavior行为 change更改 by differences分歧 in life expectancy期待
100
287000
3000
能将平均寿命的差异与行为变革相联系
05:15
is to look and see is there more behavior行为 change更改
101
290000
3000
并看看在疟疾疫情较轻的的地区
05:18
in areas where there's less malaria疟疾.
102
293000
2000
是不是行为变革就越多。
05:20
So that's what this figure数字 shows节目 you.
103
295000
2000
大家看看这些数据
05:22
This shows节目 you -- in areas with low malaria疟疾, medium malaria疟疾, high malaria疟疾 --
104
297000
4000
你会发现在疟疾疫情不同的地区
05:26
what happens发生 to the number of sexual有性 partners伙伴 as you increase增加 HIVHIV prevalence流行.
105
301000
4000
当艾滋病感染率变高时性伴侣的数量如何变化
05:30
If you look at the blue蓝色 line线,
106
305000
2000
看这条蓝线
05:32
the areas with low levels水平 of malaria疟疾, you can see in those areas,
107
307000
3000
在这些疟疾疫情较轻的地区
05:35
actually其实, the number of sexual有性 partners伙伴 is decreasing减少 a lot
108
310000
3000
性伴侣的数量在巨减
05:38
as HIVHIV prevalence流行 goes up.
109
313000
2000
当艾滋病感染率变高的时候
05:40
Areas地区 with medium levels水平 of malaria疟疾 it decreases降低 some --
110
315000
2000
在中度疫情的地方也减少了一些
05:42
it doesn't decrease减少 as much. And areas with high levels水平 of malaria疟疾 --
111
317000
3000
但没那么多。在重度疫情的地区
05:45
actually其实, it's increasing增加 a little bit, although虽然 that's not significant重大.
112
320000
5000
也减少了一点点,但非常不显著
05:50
This is not just through通过 malaria疟疾.
113
325000
2000
不仅仅是通过疟疾。
05:52
Young年轻 women妇女 who live生活 in areas with high maternal母系 mortality死亡
114
327000
3000
在这片地区年轻妇女有着高孕产妇死亡率
05:55
change更改 their behavior行为 less in response响应 to HIVHIV
115
330000
3000
为应对艾滋病而改变的行为也比
05:58
than young年轻 women妇女 who live生活 in areas with low maternal母系 mortality死亡.
116
333000
3000
那些同地区低孕产妇死亡率的女性少些。
06:01
There's another另一个 risk风险, and they respond响应 less to this existing现有 risk风险.
117
336000
4000
当有另一个威胁时,他们这种威胁的顾虑少些。
06:06
So by itself本身, I think this tells告诉 a lot about how people behave表现.
118
341000
3000
就此而言,它告诉我们许多人们如何行为的事实。
06:09
It tells告诉 us something about why we see limited有限 behavior行为 change更改 in Africa非洲.
119
344000
3000
它说明了在非洲行为变革十分有限的原因
06:12
But it also tells告诉 us something about policy政策.
120
347000
2000
同时也说明了一些关于政策的事情
06:14
Even if you only cared照顾 about AIDS艾滋病 in Africa非洲,
121
349000
3000
就算你只关心在非洲的艾滋病情况
06:17
it might威力 still be a good idea理念 to invest投资 in malaria疟疾,
122
352000
3000
投资治理疟疾也是一种不错的想法
06:20
in combating打击 poor较差的 indoor室内 air空气 quality质量,
123
355000
2000
可以改进室内欠佳的空气质量
06:22
in improving提高 maternal母系 mortality死亡 rates利率.
124
357000
2000
降低孕产妇死亡率
06:24
Because if you improve提高 those things,
125
359000
2000
因为你一旦改进这些
06:26
then people are going to have an incentive激励 to avoid避免 AIDS艾滋病 on their own拥有.
126
361000
4000
人们就会自觉地防治艾滋病
06:30
But it also tells告诉 us something about one of these facts事实 that we talked about before.
127
365000
4000
但它同样也说明了一个我们之前讨论过的事实
06:34
Education教育 campaigns活动, like the one that the president主席 is focusing调焦 on in his funding资金,
128
369000
4000
教育运动 就如同总统用他的资金所致力的那样
06:38
may可能 not be enough足够, at least最小 not alone单独.
129
373000
2000
是不够的 至少单独是不行的
06:40
If people have no incentive激励 to avoid避免 AIDS艾滋病 on their own拥有,
130
375000
2000
如果人们没有防治艾滋病的自觉
06:42
even if they know everything about the disease疾病,
131
377000
2000
就算人们知道了疾病的知识
06:44
they still may可能 not change更改 their behavior行为.
132
379000
2000
他们也不会改变自己的行为
06:46
So the other thing that I think we learn学习 here is that AIDS艾滋病 is not going to fix固定 itself本身.
133
381000
3000
我们必须了解的另一点是艾滋病不会自我修正
06:49
People aren't changing改变 their behavior行为 enough足够
134
384000
2000
人们的行为改变不足以
06:51
to decrease减少 the growth发展 in the epidemic疫情.
135
386000
3000
减弱传播的增长
06:54
So we're going to need to think about policy政策
136
389000
2000
我们要重新考虑政策
06:56
and what kind of policies政策 might威力 be effective有效.
137
391000
2000
哪一种政策会更有效
06:58
And a great way to learn学习 about policy政策 is to look at what worked工作 in the past过去.
138
393000
3000
看看过去哪些政策有效能极大帮住我们
07:01
The reason原因 that we know that the ABCABC campaign运动
139
396000
2000
我们知道ABC准则在乌干达有效
07:03
was effective有效 in Uganda乌干达 is we have good data数据 on prevalence流行 over time.
140
398000
3000
是因为我们有过去的数据来佐证
07:06
In Uganda乌干达 we see the prevalence流行 went down.
141
401000
2000
我们知道在乌干达传播极大的减弱
07:08
We know they had this campaign运动. That's how we learn学习 about what works作品.
142
403000
3000
而他们有实施了这个运动 所以我们得出它有效的结论
07:11
It's not the only place地点 we had any interventions干预措施.
143
406000
2000
这不是我们有介入的唯一地区
07:13
Other places地方 have tried试着 things, so why don't we look at those places地方
144
408000
4000
别的地方也实施了政策 为什么我们不同样看看这些地方
07:17
and see what happened发生 to their prevalence流行?
145
412000
3000
看看他们的传播率是否变化了
07:20
Unfortunately不幸, there's almost几乎 no good data数据
146
415000
2000
不幸的是,在非洲截止2003年关于艾滋病传播的
07:22
on HIVHIV prevalence流行 in the general一般 population人口 in Africa非洲 until直到 about 2003.
147
417000
5000
在整体人口上的数据再也没有了
07:27
So if I asked you, "Why don't you go and find me
148
422000
2000
如果我问你 为什么不告诉我
07:29
the prevalence流行 in Burkina布基纳法索 Faso布基纳法索 in 1991?"
149
424000
3000
在布基纳法索1991年传播率是多少
07:32
You get on Google谷歌, you Google谷歌, and you find,
150
427000
3000
你上了谷歌 却发现
07:35
actually其实 the only people tested测试 in Burkina布基纳法索 Faso布基纳法索 in 1991
151
430000
3000
在布基纳法索1991年受测试的人
07:38
are STDSTD patients耐心 and pregnant women妇女,
152
433000
2000
只有性病患者和孕妇
07:40
which哪一个 is not a terribly可怕 representative代表 group of people.
153
435000
2000
这并不是极具代表性的一个人群
07:42
Then if you poked a little more, you looked看着 a little more at what was going on,
154
437000
3000
如果你再深入一点 你就会发现更多
07:45
you'd find that actually其实 that was a pretty漂亮 good year,
155
440000
3000
你会发现当年情况很好
07:48
because in some years年份 the only people tested测试 are IVIV drug药物 users用户.
156
443000
3000
因为在那几年被测试者只有静脉吸毒者
07:51
But even worse更差 -- some years年份 it's only IVIV drug药物 users用户,
157
446000
2000
实际上情况更糟 几年只有静脉吸毒者
07:53
some years年份 it's only pregnant women妇女.
158
448000
2000
几年只有孕妇
07:55
We have no way to figure数字 out what happened发生 over time.
159
450000
2000
我们无法得知那几年到底发生了什么
07:57
We have no consistent一贯 testing测试.
160
452000
2000
我们没有前后一致的测试
07:59
Now in the last few少数 years年份, we actually其实 have doneDONE some good testing测试.
161
454000
5000
在最后几年 我们确实做了一些比较好的测试
08:04
In Kenya肯尼亚, in Zambia赞比亚, and a bunch of countries国家,
162
459000
3000
在肯尼亚 在赞比亚 和一些国家
08:07
there's been testing测试 in random随机 samples样本 of the population人口.
163
462000
3000
这些测试都是在人口中随机进行的
08:10
But this leaves树叶 us with a big gap间隙 in our knowledge知识.
164
465000
3000
但它却在我们的认知里留下一个巨大的鸿沟
08:13
So I can tell you what the prevalence流行 was in Kenya肯尼亚 in 2003,
165
468000
3000
我能告诉你在肯尼亚2003年的感染情况
08:16
but I can't tell you anything about 1993 or 1983.
166
471000
3000
但我不能告诉你关于1993年到1983年的任何情况
08:19
So this is a problem问题 for policy政策. It was a problem问题 for my research研究.
167
474000
4000
现在的政策有问题 我过去的研究也有问题
08:23
And I started开始 thinking思维 about how else其他 might威力 we figure数字 out
168
478000
4000
我开始思考我们能不能得到别的什么
08:27
what the prevalence流行 of HIVHIV was in Africa非洲 in the past过去.
169
482000
2000
关于过去非洲艾滋病的感染情况
08:29
And I think that the answer回答 is, we can look at mortality死亡 data数据,
170
484000
4000
结论是我们分析死亡率
08:33
and we can use mortality死亡 data数据 to figure数字 out what the prevalence流行 was in the past过去.
171
488000
4000
用死亡率来估算艾滋病的感染率
08:37
To do this, we're going to have to rely依靠 on the fact事实
172
492000
2000
要完成这个 我们要建立在
08:39
that AIDS艾滋病 is a very specific具体 kind of disease疾病.
173
494000
2000
艾滋病是一种非常特殊的疾病的事实上
08:41
It kills杀死 people in the prime主要 of their lives生活.
174
496000
2000
它在人们年轻的时候置人于死地
08:43
Not a lot of other diseases疾病 have that profile轮廓. And you can see here --
175
498000
3000
并没有太多其他的疾病也像这样 你可以看到
08:46
this is a graph图形 of death死亡 rates利率 by age年龄 in Botswana博茨瓦纳 and Egypt埃及.
176
501000
4000
这是一个以年龄划分关于在博茨瓦那和埃及的人口死亡率的图表
08:50
Botswana博茨瓦纳 is a place地点 with a lot of AIDS艾滋病,
177
505000
2000
博茨瓦纳是一个艾滋病疫情严重的地方
08:52
Egypt埃及 is a place地点 without a lot of AIDS艾滋病.
178
507000
2000
埃及是一个没有太多艾滋病患者的地方
08:54
And you see they have pretty漂亮 similar类似 death死亡 rates利率 among其中 young年轻 kids孩子 and old people.
179
509000
3000
它们有着在儿童和老年人群中相似的死亡率
08:57
That suggests提示 it's pretty漂亮 similar类似 levels水平 of development发展.
180
512000
3000
这说明他们有着相似的发展水平
09:00
But in this middle中间 region地区, between之间 20 and 45,
181
515000
3000
但在中间层 在20到45岁间
09:03
the death死亡 rates利率 in Botswana博茨瓦纳 are much, much, much higher更高 than in Egypt埃及.
182
518000
4000
在博茨瓦纳的死亡率要远高于在埃及的
09:07
But since以来 there are very few少数 other diseases疾病 that kill people,
183
522000
4000
又由于少有其他的致命性疾病
09:11
we can really attribute属性 that mortality死亡 to HIVHIV.
184
526000
3000
这种死亡率只能是由艾滋病导致的
09:14
But because people who died死亡 this year of AIDS艾滋病 got it a few少数 years年份 ago,
185
529000
4000
但又由于当年死于艾滋病的是在几年前就得病的
09:18
we can use this data数据 on mortality死亡 to figure数字 out what HIVHIV prevalence流行 was in the past过去.
186
533000
5000
我们就可以用这些数据来了解过去艾滋病的感染情况
09:23
So it turns out, if you use this technique技术,
187
538000
2000
结果是如果你用这种方法
09:25
actually其实 your estimates估计 of prevalence流行 are very close
188
540000
2000
你所估算的感病率将会与
09:27
to what we get from testing测试 random随机 samples样本 in the population人口,
189
542000
3000
我们随机抽样所得的结果很相近
09:30
but they're very, very different不同 than what UNAIDS联合国艾滋病规划署 tells告诉 us the prevalences患病率 are.
190
545000
5000
却与联合国艾滋病联合工作组提供的数据大相径庭
09:35
So this is a graph图形 of prevalence流行 estimated预计 by UNAIDS联合国艾滋病规划署,
191
550000
3000
这是一个由联合国艾滋病联合工作组统计的感病数据
09:38
and prevalence流行 based基于 on the mortality死亡 data数据
192
553000
2000
另外一个是用死亡率估测出来的
09:40
for the years年份 in the late晚了 1990s in nine countries国家 in Africa非洲.
193
555000
4000
在九十年代末非洲十九个国家中的感染率数据
09:44
You can see, almost几乎 without exception例外,
194
559000
2000
几乎毫无例外的
09:46
the UNAIDS联合国艾滋病规划署 estimates估计 are much higher更高 than the mortality-based死亡率为基础 estimates估计.
195
561000
4000
联合国艾滋病联合工作组的测算数据远高于这一组
09:50
UNAIDS联合国艾滋病规划署 tell us that the HIVHIV rate in Zambia赞比亚 is 20 percent百分,
196
565000
4000
联合国艾滋病联合工作组说在赞比亚艾滋病感染率是百分之20
09:54
and mortality死亡 estimates估计 suggest建议 it's only about 5 percent百分.
197
569000
4000
但用死亡率估测的确只有百分之5
09:58
And these are not trivial不重要的 differences分歧 in mortality死亡 rates利率.
198
573000
3000
这在死亡率中绝不是微小的差距
10:01
So this is another另一个 way to see this.
199
576000
2000
从另一个角度来看
10:03
You can see that for the prevalence流行 to be as high as UNAIDS联合国艾滋病规划署 says,
200
578000
2000
如果感染率真有联合国艾滋病联合工作组说的那么高的话
10:05
we have to really see 60 deaths死亡 per 10,000
201
580000
2000
我们应该看到在这个年龄层中
10:07
rather than 20 deaths死亡 per 10,000 in this age年龄 group.
202
582000
4000
半分之0.6的死亡率 而不是百分之0.2
10:11
I'm going to talk a little bit in a minute分钟
203
586000
2000
我想花一点时间来谈一下
10:13
about how we can use this kind of information信息 to learn学习 something
204
588000
3000
我们怎么用这种信息来了解事物
10:16
that's going to help us think about the world世界.
205
591000
2000
这将会有助于我们了解世界
10:18
But this also tells告诉 us that one of these facts事实
206
593000
2000
但它同样也说明了
10:20
that I mentioned提到 in the beginning开始 may可能 not be quite相当 right.
207
595000
3000
我在开始提到的一些事实并不正确
10:23
If you think that 25 million百万 people are infected感染,
208
598000
2000
如果你认为2500万人感染了
10:25
if you think that the UNAIDS联合国艾滋病规划署 numbers数字 are much too high,
209
600000
3000
如果你认为联合国艾滋病联合工作组的数据太高了
10:28
maybe that's more like 10 or 15 million百万.
210
603000
2000
或许应该是1000到1500万人
10:30
It doesn't mean that AIDS艾滋病 isn't a problem问题. It's a gigantic巨大 problem问题.
211
605000
4000
这却不意味着艾滋病不是一个问题 这绝对是个大问题
10:34
But it does suggest建议 that that number might威力 be a little big.
212
609000
4000
只是之前的数据有些过大
10:38
What I really want to do, is I want to use this new data数据
213
613000
2000
我真正想做的是用这些新数据
10:40
to try to figure数字 out what makes品牌 the HIVHIV epidemic疫情 grow增长 faster更快 or slower比较慢.
214
615000
4000
去找到是什么影响了艾滋病传播的增长快慢
10:44
And I said in the beginning开始, I wasn't going to tell you about exports出口.
215
619000
3000
我在开始说 我不会谈什么进口之类的事
10:47
When I started开始 working加工 on these projects项目,
216
622000
2000
我在开始研究这些项目时
10:49
I was not thinking思维 at all about economics经济学,
217
624000
2000
并不是全想的跟经济学有关的东西
10:51
but eventually终于 it kind of sucks you back in.
218
626000
3000
但事实上它会将你引导回去
10:54
So I am going to talk about exports出口 and prices价格.
219
629000
3000
我要讲一些跟出口和价格有关的东西
10:57
And I want to talk about the relationship关系 between之间 economic经济 activity活动,
220
632000
3000
我想谈一下经济活动
11:00
in particular特定 export出口 volume, and HIVHIV infections感染.
221
635000
4000
特别是出口量和艾滋病传染之间的关系
11:04
So obviously明显, as an economist经济学家, I'm deeply familiar
222
639000
4000
很明显我作为一个经济学家 我更加熟悉
11:08
with the fact事实 that development发展, that openness透明度 to trade贸易,
223
643000
2000
对外贸易的发展和开放
11:10
is really good for developing发展 countries国家.
224
645000
2000
将会给发展中国家带来极大的好处
11:12
It's good for improving提高 people's人们 lives生活.
225
647000
3000
这有助于改善民生
11:15
But openness透明度 and inter-connectedness相互联系, it comes with a cost成本
226
650000
2000
但开放和全球联系 是会有成本的
11:17
when we think about disease疾病. I don't think this should be a surprise.
227
652000
3000
一旦和疾病联系起来 我觉得大家不应该吃惊
11:20
On Wednesday星期三, I learned学到了 from Laurie劳瑞 Garrett加勒特
228
655000
2000
在周三 劳里·加勒特跟我说
11:22
that I'm definitely无疑 going to get the bird flu流感,
229
657000
2000
我肯定会得禽流感的
11:24
and I wouldn't不会 be at all worried担心 about that
230
659000
3000
但我丝毫不用担心
11:27
if we never had any contact联系 with Asia亚洲.
231
662000
3000
只要我和亚洲走的不近
11:30
And HIVHIV is actually其实 particularly尤其 closely密切 linked关联 to transit过境.
232
665000
4000
艾滋病与过境的联系很紧密
11:34
The epidemic疫情 was introduced介绍 to the US
233
669000
2000
这个传染病
11:36
by actually其实 one male steward管家 on an airline航空公司 flight飞行,
234
671000
4000
是由一个在航班上的男管理员带来美国的
11:40
who got the disease疾病 in Africa非洲 and brought it back.
235
675000
2000
他在非洲染了病 然后就带了回来
11:42
And that was the genesis创世纪 of the entire整个 epidemic疫情 in the US.
236
677000
3000
这是艾滋病在美国传播的起源
11:45
In Africa非洲, epidemiologists流行病学家 have noted注意 for a long time
237
680000
4000
在非洲 流行病学家早注意到
11:49
that truck卡车 drivers司机 and migrants移民 are more likely容易 to be infected感染 than other people.
238
684000
4000
卡车司机和移民比其他人群更易感染
11:53
Areas地区 with a lot of economic经济 activity活动 --
239
688000
2000
在那些经济活动较多
11:55
with a lot of roads道路, with a lot of urbanization城市化 --
240
690000
3000
公路较多 城市化更快的地方
11:58
those areas have higher更高 prevalence流行 than others其他.
241
693000
2000
比其他的地方流行强度更大
12:00
But that actually其实 doesn't mean at all
242
695000
2000
事实上这并不意味着什么
12:02
that if we gave people more exports出口, more trade贸易, that that would increase增加 prevalence流行.
243
697000
4000
如果出口扩大,贸易加强,它就会增加传播度
12:06
By using运用 this new data数据, using运用 this information信息 about prevalence流行 over time,
244
701000
4000
在使用了新数据 使用了跨时段的传播度信息后
12:10
we can actually其实 test测试 that. And so it seems似乎 to be --
245
705000
4000
我就能真正地检测它 那么这意味着
12:14
fortunately幸好, I think -- it seems似乎 to be the case案件
246
709000
2000
很幸运的是就我看来
12:16
that these things are positively积极 related有关.
247
711000
2000
这些事件是正相关的
12:18
More exports出口 means手段 more AIDS艾滋病. And that effect影响 is really big.
248
713000
4000
出口越多意味着艾滋病患者越多 而且这种影响是巨大的
12:22
So the data数据 that I have suggests提示 that if you double export出口 volume,
249
717000
4000
我研究的数据表明如果出口量扩大两倍
12:26
it will lead to a quadrupling翻两番 of new HIVHIV infections感染.
250
721000
5000
会导致艾滋病感染病例数扩大四倍
12:31
So this has important重要 implications启示 both for forecasting预测 and for policy政策.
251
726000
3000
这对于预测和政策都有重要的影响
12:34
From a forecasting预测 perspective透视, if we know where trade贸易 is likely容易 to change更改,
252
729000
4000
从预测的角度 如果我们知道何处贸易将发生变化
12:38
for example, because of the African非洲人 Growth发展 and Opportunities机会 Act法案
253
733000
3000
比方说因为非洲增长与机遇法案
12:41
or other policies政策 that encourage鼓励 trade贸易,
254
736000
2000
或其他鼓励贸易的政策出台
12:43
we can actually其实 think about which哪一个 areas are likely容易 to be heavily严重 infected感染 with HIVHIV.
255
738000
5000
我们可以想见哪些地方将受到艾滋病的侵袭
12:48
And we can go and we can try to have pre-emptive先发制人 preventive预防 measures措施 there.
256
743000
6000
我们就可去那里 采取先发制人的预防措施
12:54
Likewise同样, as we're developing发展 policies政策 to try to encourage鼓励 exports出口,
257
749000
3000
同样地 如果我们要出台鼓励出口的政策
12:57
if we know there's this externality外部性 --
258
752000
2000
我们知道有这样一种外部性
12:59
this extra额外 thing that's going to happen发生 as we increase增加 exports出口 --
259
754000
2000
因为我们增加出口所带来的其他事情的变化
13:01
we can think about what the right kinds of policies政策 are.
260
756000
3000
我们就可以更好的制定政策
13:04
But it also tells告诉 us something about one of these things that we think that we know.
261
759000
3000
它同样也说明了一些我们觉得我们了解的事情
13:07
Even though虽然 it is the case案件 that poverty贫穷 is linked关联 to AIDS艾滋病,
262
762000
3000
尽管艾滋病是和贫困紧密相连
13:10
in the sense that Africa非洲 is poor较差的 and they have a lot of AIDS艾滋病,
263
765000
3000
而且非洲很穷 他们那也有很多艾滋病患者
13:13
it's not necessarily一定 the case案件 that improving提高 poverty贫穷 -- at least最小 in the short run,
264
768000
4000
但改变贫穷的现状也不是必须的 至少在短期不是
13:17
that improving提高 exports出口 and improving提高 development发展 --
265
772000
2000
加大出口和扩大发展
13:19
it's not necessarily一定 the case案件 that that's going to lead
266
774000
2000
也并不一定能够
13:21
to a decline下降 in HIVHIV prevalence流行.
267
776000
2000
减轻艾滋病的传播
13:24
So throughout始终 this talk I've mentioned提到 a few少数 times
268
779000
2000
至始至终我提到过几次
13:26
the special特别 case案件 of Uganda乌干达, and the fact事实 that
269
781000
2000
乌干达的特殊案例
13:28
it's the only country国家 in sub-Saharan撒哈拉以南 Africa非洲 with successful成功 prevention预防.
270
783000
4000
它是唯一一个在撒哈拉以南的非洲国家里的成功预防案例
13:32
It's been widely广泛 heralded预示.
271
787000
2000
它广为人知
13:34
It's been replicated复制 in Kenya肯尼亚, and Tanzania坦桑尼亚, and South Africa非洲 and many许多 other places地方.
272
789000
6000
在肯尼亚、坦桑尼亚、南非以及很多地方被复制使用
13:40
But now I want to actually其实 also question that.
273
795000
4000
我现在真正想置疑的是
13:44
Because it is true真正 that there was a decline下降 in prevalence流行
274
799000
3000
因为在九十年代在乌干达
13:47
in Uganda乌干达 in the 1990s. It's true真正 that they had an education教育 campaign运动.
275
802000
4000
传播率确实下降了 他们也确实采取的是教育运动
13:51
But there was actually其实 something else其他 that happened发生 in Uganda乌干达 in this period.
276
806000
6000
但在那个时期在乌干达还发生了一些别的事
13:57
There was a big decline下降 in coffee咖啡 prices价格.
277
812000
2000
在咖啡价格上有很大的下降
13:59
Coffee咖啡 is Uganda's乌干达 major重大的 export出口.
278
814000
2000
咖啡是乌干达的主要出口产品
14:01
Their exports出口 went down a lot in the early 1990s -- and actually其实 that decline下降 lines线 up
279
816000
5000
在九十年代早期他们出口量大幅下降 而这种降幅的变化
14:06
really, really closely密切 with this decline下降 in new HIVHIV infections感染.
280
821000
4000
这和新感染艾滋病数的降幅极度接近
14:10
So you can see that both of these series系列 --
281
825000
3000
你们可看到这些
14:13
the black黑色 line线 is export出口 value, the red line线 is new HIVHIV infections感染 --
282
828000
3000
黑线是出口值 红线是新感染艾滋病数
14:16
you can see they're both increasing增加.
283
831000
2000
你发现它们都增长了
14:18
Starting开始 about 1987 they're both going down a lot.
284
833000
2000
从1987年起 它们又都开始大幅下降
14:20
And then actually其实 they track跟踪 each other
285
835000
2000
它们的轨迹在年代末的增加变化上
14:22
a little bit on the increase增加 later后来 in the decade.
286
837000
2000
又有一些重叠
14:24
So if you combine结合 the intuition直觉 in this figure数字
287
839000
2000
将你的直觉和
14:26
with some of the data数据 that I talked about before,
288
841000
3000
我之前讲的一些数据联系起来
14:29
it suggests提示 that somewhere某处 between之间 25 percent百分 and 50 percent百分
289
844000
4000
则发现大约有百分之25到50的
14:33
of the decline下降 in prevalence流行 in Uganda乌干达
290
848000
2000
在乌干达的传播减少比率
14:35
actually其实 would have happened发生 even without any education教育 campaign运动.
291
850000
4000
是在没有任何教育运动的情况下也会发生的
14:39
But that's enormously巨大 important重要 for policy政策.
292
854000
2000
这对于制定政策极度重要
14:41
We're spending开支 so much money to try to replicate复制 this campaign运动.
293
856000
2000
我们花了大量的钱试图复制这项运动
14:43
And if it was only 50 percent百分 as effective有效 as we think that it was,
294
858000
3000
但如果仅仅只有我们预想的效果的一半的话
14:46
then there are all sorts排序 of other things
295
861000
2000
我们何不把我们的钱投在
14:48
maybe we should be spending开支 our money on instead代替.
296
863000
2000
一些其他的事上呢
14:50
Trying to change更改 transmission传输 rates利率 by treating治疗 other sexually transmitted发送 diseases疾病.
297
865000
4000
通过应对其他性传播疾病来试图改变传播速度
14:54
Trying to change更改 them by engaging in male circumcision割礼.
298
869000
2000
通过包皮环切术来改变他们
14:56
There are tons of other things that we should think about doing.
299
871000
2000
有数以万计的事我们还可以做
14:58
And maybe this tells告诉 us that we should be thinking思维 more about those things.
300
873000
4000
我们或许应该多考虑一下这些方面的事
15:02
I hope希望 that in the last 16 minutes分钟 I've told you something that you didn't know about AIDS艾滋病,
301
877000
5000
希望在刚刚的16分钟我告诉了大家一些关于艾滋病不为人知的信息
15:07
and I hope希望 that I've gotten得到 you questioning疑问 a little bit
302
882000
2000
希望我引起了大家对一些自己知道的事
15:09
some of the things that you did know.
303
884000
2000
的一些疑问
15:11
And I hope希望 that I've convinced相信 you maybe
304
886000
2000
我希望我使你们相信或许
15:13
that it's important重要 to understand理解 things about the epidemic疫情
305
888000
2000
为了制定政策
15:15
in order订购 to think about policy政策.
306
890000
2000
去了解传染病的事情是很重要的
15:18
But more than anything, you know, I'm an academic学术的.
307
893000
2000
但是你要知道 我是一个学者
15:20
And when I leave离开 here, I'm going to go back
308
895000
2000
当我离开这 我会回到
15:22
and sit in my tiny office办公室, and my computer电脑, and my data数据.
309
897000
3000
我的小办公室里 对着我的电脑 我的数据
15:25
And the thing that's most exciting扣人心弦 about that
310
900000
2000
和那些每一次我开始研究时
15:27
is every一切 time I think about research研究, there are more questions问题.
311
902000
3000
就会出现的让我无比兴奋的新问题
15:30
There are more things that I think that I want to do.
312
905000
2000
我想要做的还要多得多
15:32
And what's really, really great about being存在 here
313
907000
2000
能来这里真的非常非常棒
15:34
is I'm sure that the questions问题 that you guys have
314
909000
2000
我确信你们思考的问题
15:36
are very, very different不同 than the questions问题 that I think up myself.
315
911000
3000
将会跟我自己想的截然不同
15:39
And I can't wait to hear about what they are.
316
914000
2000
我迫不及待地想知道它们
15:41
So thank you very much.
317
916000
2000
非常感谢大家。
Translated by Jie Zhao
Reviewed by Zheng Li

▲Back to top

ABOUT THE SPEAKER
Emily Oster - Assumption-busting economist
Emily Oster, a University of Chicago economist, uses the dismal science to rethink conventional wisdom, from her Harvard doctoral thesis that took on famed economist Amartya Sen to her recent work debunking assumptions on HIV prevalence in Africa.

Why you should listen

Emily Oster, an Assistant Professor of Economics at the University of Chicago, has a history of rethinking conventional wisdom.

Her Harvard doctoral thesis took on famed economist Amartya Sen and his claim that 100 million women were statistically missing from the developing world. He blamed misogynist medical care and outright sex-selective abortion for the gap, but Oster pointed to data indicating that in countries where Hepetitis B infections were higher, more boys were born. Through her unorthodox analysis of medical data, she accounted for 50% of the missing girls. Three years later, she would publish another paper amending her findings, stating that, after further study, the relationship between Hepetitis B and missing women was not apparent. This concession, along with her audacity to challenge economic assumptions and her dozens of other influential papers, has earned her the respect of the global academic community. 

She's also investigated the role of bad weather in the rise in witchcraft trials in Medieval Europe and what drives people to play the Powerball lottery. Her latest target: busting assumptions on HIV in Africa.

And she's an advice columnist too >>

 

More profile about the speaker
Emily Oster | Speaker | TED.com