MonaChalabi_2017S[莫娜 查拉比][甄别不可信统计数据的3种方法].pdf

《MonaChalabi_2017S[莫娜 查拉比][甄别不可信统计数据的3种方法].pdf》由会员分享，可在线阅读，更多相关《MonaChalabi_2017S[莫娜 查拉比][甄别不可信统计数据的3种方法].pdf（6页珍藏版）》请在淘文阁 - 分享文档赚钱的网站上搜索。

1、www.XiYuS锡育软件Now,Im going to be talking about statistics today.今天我想讲一下统计学。00:12If that makes you immediately feel a little bit wary,thats OK,that doesnt make you some kind of crazy conspiracytheorist,it makes you skeptical.如果你马上有所警觉,没关系,我不会让你变成那种疯狂的阴谋论者,只会让你学会质疑。00:15And when it comes to numbers,esp

2、ecially now,you shouldbe skeptical.尤其是现在,说到数字、数据,你应该时刻保持怀疑态度。00:22But you should also be able to tell which numbers arereliable and which ones arent.但你需要能够辨别哪些数字是可信的,哪些存在问题。00:26So today I want to try to give you some tools to be able to dothat.因此为了让你们具有这种甄别能力,我会试着提供一些方法。00:30But before I do,I just

3、want to clarify which numbers Imtalking about here.但在我开始之前,我想说明我所说的统计 数据是什么类型的。00:34Im not talking about claims like,9 out of 10 womenrecommend this anti-aging cream.我讨论的并不是这种广告:“10个女人中有9个 会推荐这款抗衰老乳霜。”00:38I think a lot of us always roll our eyes at numbers like that.我想我们中很多人对 这样的数字不以为然。00:42Whats d

4、ifferent now is people are questioning statistics like,The US unemployment rate is five percent.但不同的是,现在人们对这样的 数字也开始抱有怀疑:“美国的失业率是 5%。”00:45statistics:n.统计;统计学;统计统计资料 wary:adj.谨慎的;机警的;惟恐的;考虑周到的 conspiracy:n.阴谋;共谋;阴谋集团theorist:n.理论家 skeptical:adj.怀疑的；怀疑论的，不可知论的 anti-aging:n.抗衰老What makes this claim di

5、fferent is it doesnt come from aprivate company,it comes from the government.这个数字不是出于一个私人公司,而是来自政府。00:50About 4 out of 10 Americans distrust the economic data thatgets reported by government.事实上,如今10个美国人中差不多有4个 不相信政府公布的 那些经济数据。00:55Among supporters of President Trump its even higher;itsabout 7 out o

6、f 10.在特朗普总统的支持者中,情况更糟,10人中有7人不相信那些数据。01:00I dont need to tell anyone here that there are a lot ofdividing lines in our society right now,and a lot of them startto make sense,once you understand peoples relationshipswith these government numbers.无须我告诉大家,当今我们的社会中 划分了许多的分界线,一旦你明白“人”和那些“政府数据”之间的关系,这些分界线

7、中的 大多数就不言自明了。01:04On the one hand,there are those who say these statistics arecrucial,that we need them to make sense of society as awhole in order to move beyond emotional anecdotes andmeasure progress in a subjective way.一方面,有些人说,这些统计数据很重要,我们需要它们来理解整个社会,进而脱离情感喜好、偏见等,以一种客观的方式衡量发展程度。01:14distrust:vt

8、.不信任/n.不信任/怀疑 supporters:n.支持者,支持单位;支持单元 Trump:n.王牌;法宝;喇叭/vt.胜过;打出王牌赢/vi.出王牌;吹喇叭 make sense:有意义;讲得通;言之有理 On the one hand:一方面 crucial:adj.重要的；决定性的；定局的；决断的 make sense of:搞清的意思 as a whole:总的来说 emotional:adj.情绪的；易激动的；感动人的anecdotes:n.奇闻轶事(anecdote的复数)subjective:adj.主观的；个人的；自觉的And then there are the other

9、s,who say that these statisticsare elitist,maybe even rigged;they dont make sense andthey dont really reflect whats happening in peopleseveryday lives.正相反,另一方面,有一些人说 这些数据隶属于精英统治,它们甚至有可能是被操纵的;它们没什么意义,也不能真正反映 人们的日常生活中发生了什么。01:25It kind of feels like that second group is winning theargument right now.现

10、在看来,似乎是第二种人 在这两个阵营的针锋相对中获得了胜利。01:35Were living in a world of alternative facts,where peopledont find statistics this kind of common ground,this startingpoint for debate.我们所生活的世界里,胡说八道已成常态,人们对统计数据不能达成共识,也不会把数据作为争论时的出发点。01:38This is a problem.这是一个问题。01:45There are actually moves in the US right now to

11、 get rid ofsome government statistics altogether.如今在美国,人们有一个举措是 完全丢弃政府的某项统计数据。01:46TED演讲者：Mona Chalabi|莫娜 查拉比演讲标题：3 ways to spot a bad statistic|甄别不可信统计数据的3种方法内容概要：How do parents protect their children and help them feel secure again when theirhomes are ripped apart by war?In this warm-hearted talk,

12、psychologist Aala El-Khani sharesher work supporting-and learning from-refugee families affected by the civil war in Syria.She asks:How can we help these loving parents give their kids the warm,secure parenting theymost need?有时候很难知道哪些统计数字是可信的，但我们也不能完全忽略这些数据正相反，我们要学会看到这些统计数字背后的东西。在这场愉轻松、有趣的演讲中，数据记者（d

13、ata journalist）莫娜 查拉比分享了几点实用的建议，它将帮助你质疑、解释并且真正理解这些统计数字的含义。Right now theres a bill in congress about measuring racialinequality.现在,国会有一项关于 衡量种族不平等程度的议案。01:51The draft law says that government money should not beused to collect data on racial segregation.草案规定,政府资金不能被用在 搜集关于种族隔离的数据上。01:55This is a tota

14、l disaster.这完全是一场灾难。01:59If we dont have this data,how can we observediscrimination,let alone fix it?如果我们没有相关数据,我们怎样观察种族歧视现象,更不要说解决它了？02:01In other words:How can a government create fair policies ifthey cant measure current levels of unfairness?换句话说:如果政府都不能衡量 如今的种族不平等程度,那他们又如何制定合理的政策呢？02:06elitist:n

15、.优秀人才;杰出人物/adj.优秀人才的;杰出人才的 segregation:n.隔离,分离;种族隔离 discrimination:n.歧视；区别，辨别；识别力 let alone:更不必说;听任;不打扰 unfairness:n.不公平,不公正;不正当This isnt just about discrimination,its everything-thinkabout it.不只是在种族歧视方面,还包括其他方面 想一想吧,02:12How can we legislate on health care if we dont have gooddata on health or pove

16、rty?如果我们没有关于医疗健康 以及贫穷问题的可靠数据,要如何为医疗保健问题立法？02:16How can we have public debate about immigration if wecant at least agree on how many people are entering andleaving the country?如果我们甚至对进入或离开 这个国家的人数都无法达成共识,我们又如何 我们又如何02:20Statistics come from the state;thats where they got theirname.“统计”（statistics）来自于

17、“国家事务”（state）,是它的词源所在。02:26The point was to better measure the population in order tobetter serve it.它的重点在于更准确地衡量人口 来更好地造福于人民。02:29So we need these government numbers,but we also have tomove beyond either blindly accepting or blindly rejectingthem.所以我们需要这些政府数据,但我们也不能盲目接受 或者盲目排斥它们。02:33We need to lea

18、rn the skills to be able to spot bad statistics.我们需要学会甄别不可信的数据。02:38I started to learn some of these when I was working in astatistical department thats part of the United Nations.当我在统计部门工作时 它是联合国的一部分,我开始学会了一些甄别的技巧。02:41Our job was to find out how many Iraqis had been forcedfrom their homes as a res

19、ult of the war,and what theyneeded.我们的工作是弄清楚 有多少伊拉克人是因为战争而被迫离家,以及他们需要什么。02:47legislate:vt.用立法规定;通过立法/vi.立法;制定法律 immigration:n.外来移民;移居 blindly:adv.盲目地;轻率地;摸索地rejecting:拒绝/拒收 statistical:adj.统计的;统计学的 Iraqis:adj.伊拉克的;伊拉克人的/n.伊拉克人 as a result:结果It was really important work,but it was also incrediblydiff

20、icult.这的确是一项非常重要的工作,但其难度也让人难以置信。02:53Every single day,we were making decisions that affected theaccuracy of our numbers-decisions like which parts of thecountry we should go to,who we should speak to,whichquestions we should ask.每一天,我们都在做出 会影响我们统计数据精确度的决定 决定我们应该去 这个国家的哪一块区域调查,和哪种人交谈,我们应该询问他们什么问题。02:

21、56And I started to feel really disillusioned with our work,because we thought we were doing a really good job,butthe one group of people who could really tell us were theIraqis,and they rarely got the chance to find our analysis,letalone question it.我开始对我们的工作感到失望,因为我们以为我们的 工作非常棒、很有意义,但是真正能告诉我们实情的是那些

22、伊拉克人,他们却很难有机会看到我们的 分析结果,更别提质疑它了。03:06So I started to feel really determined that the one way tomake numbers more accurate is to have as many people aspossible be able to question them.所以我开始坚信,能让数据更加精确可信的一种方法,就是让尽可能多的人能够去质疑它。03:18So I became a data journalist.所以我成为了一名“数据记者”。03:25My job is finding the

23、se data sets and sharing them with thepublic.我的工作是找到这些数据集,并且向公众发布。03:26Anyone can do this,you dont have to be a geek or a nerd.任何人都可以做到这件事;你无须是一个极客或者老手。03:30incredibly:adv.难以置信地；非常地 disillusioned:adj.醒悟的;幻想破灭的;不抱幻想的/v.使幻想破灭(disillusion的过去分词);唤醒 good job:好运;幸运的事情;干的不错 geek:n.做低级滑稽表演的人,反常的人;畸形人;野人/n.(

24、俚)极客,即智力超群,善于钻研但不懂与人交往的怪才 nerd:n.呆子;讨厌的人You can ignore those words;theyre used by people trying tosay theyre smart while pretending theyre humble.你可以忽视那些术语;那些想要表现得很聪明 又假装谦虚的人才喜欢这些专业词汇。03:34Absolutely anyone can do this.毫无疑问任何人都能做到这些事。03:39I want to give you guys three questions that will help you be

25、able to spot some bad statistics.我想向你们提出3个问题,会帮助你甄别那些不可靠的数据。03:40So,question number one is:Can you see uncertainty?那么第一个问题是:你能看到不确定性吗？03:45One of things thats really changed peoples relationshipwith numbers,and even their trust in the media,has been有一件事真正改变了 人们和数据的关系,甚至改变了人们对媒体的信任,那就是政治上的民意调查的滥用。03:

26、49with numbers,and even their trust in the media,has beenthe use of political polls.滥用。03:49I personally have a lot of issues with political polls because Ithink the role of journalists is actually to report the facts andnot attempt to predict them,我个人对政治民意调查颇有微词,因为我认为,记者的任务就是 原原本本地报道事实,而不应该尝试去预测什么,

27、03:56especially when those predictions can actually damagedemocracy by signaling to people:dont bother to vote forthat guy,he doesnt have a chance.尤其是当这些“预测”实际上会有损民主的时候,它向人们示意:不用费心给那个家伙投票了,他是没机会入选的。04:04Lets set that aside for now and talk about the accuracy ofthis endeavor.我们把这事先搁到一边,来谈谈这种做法的精确度如何。

28、04:11uncertainty:n.不确定,不可靠 trust in:vt.信任;依靠;存放 polls:n.民意调查;投票;投票模块(poll的复数形式)/v.对进行民意测验,使投票(poll的单三形式)journalists:n.新闻记者(journalist的复数)predictions:n.预测,预言(prediction复数形式)signaling:n.发信号;打信号 for now:目前,暂时 endeavor:n.努力;尽力(等于endeavour)/vi.努力;尽力(等于endeavour)/vt.努力;尽力(等于endeavour)Based on national ele

29、ctions in the UK,Italy,Israel and ofcourse,the most recent US presidential election,using pollsto predict electoral outcomes is about as accurate as usingthe moon to predict hospital admissions.根据英国、意大利、以色列的国家选举情况,当然,还有近期的 美国总统选举情况,它们都表明:使用民意 调查来预测选举结果 无异于夜观天象来预测是否 住院都是不可靠的。04:15www.XiYuS锡育软件No,seri

30、ously,I used actual data from an academic study todraw this.说真的,我用一份学术研究中的 真实数据画出了这幅图。04:28There are a lot of reasons why polling has become soinaccurate.民意调查变得如此不准确 是有很多原因的。04:32Our societies have become really diverse,which makes itdifficult for pollsters to get a really nice representativesample

31、of the population for their polls.我们的社会已经变得相当多元化,这使得民意调查者很难在 人群中为他们的调查得到一份04:36People are really reluctant to answer their phones topollsters,and also,shockingly enough,people might lie.人们不是很情愿地 回答民意调查电话,另外,令人震惊的是,人们可能会说谎。04:43But you wouldnt necessarily know that to look at the media.但在媒体的报道中,你不一定会

32、 看到并了解这些过程。04:49Israel:n.以色列(亚洲国家);犹太人,以色列人 presidential:adj.总统的;首长的;统辖的 electoral:adj.选举的;选举人的admissions:n.许可;承认;入会费;入场券(admission的复数形式)polling:n.投票;轮询/v.获得票(poll的ing形式)inaccurate:adj.错误的 diverse:adj.不同的；多种多样的；变化多的 pollsters:民意测验专家/民意调查人(pollster的名词复数)reluctant:adj.不情愿的；勉强的；顽抗的 shockingly:adv.怕人地,非

33、常地;不正当地For one thing,the probability of a Hillary Clinton win wascommunicated with decimal places.例如,希拉里 克林顿 赢得选举的可能性 在媒体的报道中精确到了小数点。04:52We dont use decimal places to describe the temperature.我们描述气温都不会这么精确。04:57How on earth can predicting the behavior of 230 millionvoters in this country be that pr

34、ecise?所以,对这个国家里的 2亿3千万选民的行为的预测,怎么可能会如此精确？05:00And then there were those sleek charts.另外,还有一些美观的图表。05:06See,a lot of data visualizations will overstate certainty,andit works-these charts can numb our brains to criticism.知道吗,很多数据可视化的方法会 夸大数据的准确性,而且很有效 这些图表会麻痹我们的大脑,让我们无法对其准确性生疑。05:08When you hear a sta

35、tistic,you might feel skeptical.当你听到一项统计数据,你或许会怀疑。05:15As soon as its buried in a chart,it feels like some kind ofobjective science,and its not.但当它整合进了图表中,它看起来仿佛就成了客观的科学,但实际上正好相反。05:17So I was trying to find ways to better communicate this topeople,to show people the uncertainty in our numbers.所以我试着

36、找出一些方法 来告诉人们这些数据背后的事实,为人们展示这些数据的不确定性。05:22For one thing:首先;一则 probability:n.可能性；机率；数或然率 Hillary:n.希拉里(美国国务卿)decimal:adj.小数的;十进位的/n.小数 predicting:v.预测(predict的ing形式);预报;预言 sleek:adj.圆滑的;井然有序的/vt.使光滑;掩盖/vi.打扮整洁;滑动 visualizations:n.计可视化(visualization的复数形式);直观化 overstate:vt.夸张;夸大的叙述 numb:vt.使麻木;使发愣;使失去感

37、觉/adj.麻木的;发愣的What I did was I started taking real data sets,and turningthem into hand-drawn visualizations,so that people can seehow imprecise the data is;so people can see that a humandid this,a human found the data and visualized it.我所做的,就是先采集数据,然后把它们转换为手绘的示意图,所以人们可以看到 这些数据的不精确性;可以看到这份图表是人画出来的,有人搜

38、集数据并把它可视化了。05:28For example,instead of finding out the probability of gettingthe flu in any given month,you can see the roughdistribution of flu season.例如,我们不去计算出每个月 可能患流感的概率,但我们可以得到流感 在每个季节的大致分布。05:40This is-这就是05:47(Laughter)a bad shot to show in February.（笑声）现在就是二月,展示 这张图不是很合适。05:48But its also m

39、ore responsible data visualization,because if但这是更加负责任的数据可视图表,因为,你如果在这张图中 展示了精确的概率,那么它就有可能鼓励But its also more responsible data visualization,because ifyou were to show the exact probabilities,maybe that wouldencourage people to get their flu jabs at the wrong time.这张图中 展示了精确的概率,那么它就有可能鼓励人们 在不合适的时间段05

40、:51The point of these shaky lines is so that people rememberthese imprecisions,but also so they dont necessarily walkaway with a specific number,but they can rememberimportant facts.这些摇摇晃晃的线条 是想向人们说明 这些数据的不精确性,它也不会让人们简单地 得到一个鸡肋的具体数字,而是让他们记住重要的事实。06:00hand-drawn:抽花刺绣品 imprecise:adj.不精确的;不严密的;不确切的 vis

41、ualized:adj.直观的;直视的/v.使形象化;想像(visualize的过去分词)show in:领入;领人 probabilities:可能性/统计概率(probability的复数)jabs:n.注射(jab的复数形式);猛戳/v.刺;戳;猛击(jab的三单形式)shaky:adj.摇晃的;不可靠的;不坚定的 walk away with:顺手带走;轻易赢得;轻易获胜;偷走Facts like injustice and inequality leave a huge mark on ourlives.不公正、不平等这样的事实 给我们的生活带来了巨大的影响。06:10Facts l

42、ike Black Americans and Native Americans haveshorter life expectancies than those of other races,and thatisnt changing anytime soon.“美国黑人以及原住民的预期寿命 比其他种族的人短一些”,这样的事实,短时间内也难以改变。06:14Facts like prisoners in the US can be kept in solitaryconfinement cells that are smaller than the size of anaverage par

43、king space.还包括“美国监狱中被单独监禁的 囚犯的活动空间 比起一般的停车位面积还要小”这样的事实。06:22The point of these visualizations is also to remind people ofsome really important statistical concepts,concepts likeaverages.这些数据可视化的 意义也在于提醒人们 一些极其重要的统计概念,例如“平均”的概念。06:30So lets say you hear a claim like,The average swimmingpool in the US

44、 contains 6.23 fecal accidents.比方说你听到了这样一句话:“在美国,一般大小的游泳池 平均含有6.23次大便。”06:37That doesnt mean every single swimming pool in thecountry contains exactly 6.23 turds.这并不意味着这个 国家里每一座游泳池 刚好有6.23次大便。06:43So in order to show that,I went back to the original data,which comes from the CDC,who surveyed 47 swim

45、mingfacilities.所以为了展示这一点,我找到疾病预防控制 中心的原始数据,他们调查了47座游泳池。06:48injustice:n.不公正;不讲道义 expectancies:n.期望,期待 solitary:adj.孤独的；独居的/n.独居者；隐士 confinement:n.限制；监禁；分娩 concepts:n.概念,观念;思想(concept复数形式)averages:n.平均值;中等水平(average的复数形式)/v.均分(average的三单形式);算平均数;对做出总的评价 fecal:adj.排泄物的;残渣的;糟粕的 turds:n.粪,粪便;可鄙的人surveye

46、d:v.调查,俯视(survey的过去式和过去分词形式);考察,调研 facilities:n.设施;工具,设备And I just spent one evening redistributing poop.我花了一个晚上“重新分配这些大便”。06:55So you can kind of see how misleading averages can be.所以你可以看到,“平均”是多么误导人。06:57(Laughter)OK,so the second question that you guys shouldbe asking yourselves to spot bad numbe

47、rs is:（笑声）好,第二个问题,你们在甄别数据时 应当反问自己的是:07:00Can I see myself in the data?我能在这些数据中 看到自己的身影吗？07:07This question is also about averages in a way,because part ofthe reason why people are so frustrated with these nationalstatistics,is they dont really tell the story of whos winningand whos losing from natio

48、nal policy.这个问题在某种程度上 也是关于“平均”的,因为,人们对政府 公布的数据感到失望的 一部分原因就是,他们无法分清,在国家政策下 谁损失、谁获益。07:09Its easy to understand why people are frustrated with globalaverages when they dont match up with their personalexperiences.很容易理解为什么当人们发现 自己的个人经验与全球平均数据 不相符时,他们会感到非常失望。07:20I wanted to show people the way data rel

49、ates to theireveryday lives.我想为人们展示:数据是如何 与他们的日常生活相关联的。07:26I started this advice column called Dear Mona,我开设了一个咨询专栏,叫做亲爱的Mona,07:30where people would write to me with questions and concernsand Id try to answer them with data.人们会写信询问 他们所关心的事情,这些问题包罗万象,07:32redistributing:n.重新分配/v.再分配;重新分布(redistribu

50、te的ing形式)poop:n.船尾;傻子;内幕消息/vt.使精疲力尽;使船尾受击/vi.疲乏;排便 misleading:adj.令人误解的;引入歧途的/v.给带错路;把引入歧途(mislead的ing形式)in a way:在某种程度上;有点儿;十分激动 frustrated:adj.失意的,挫败的;泄气的/v.挫败;阻挠(frustrate的过去式和过去分词形式)match up:使相配,使协调People asked me anything.例如,“和老婆分床睡是正常的吗？”07:36questions like,Is it normal to sleep in a separate