评论
打印
So, at The Washington Post, we are really lucky because we have subject-matter experts throughout the newsroom.
在《华盛顿邮报》，我们真的很幸运，因为我们的新闻编辑部有各个主题的专家。
We meet with editors and reporters of -- who cover what we will be asking about.
我们会见了各个编辑和记者，他们报道了我们将询问的问题。
So if it's a national poll about politics, we meet with the politics team.
如果是一项关于政治的全国性民意调查，我们会与政治团队会面。
If it is a poll of Virginia voters, we meet with folks who cover Virginia, and we ask them what they want to know.
如果是一项针对弗吉尼亚州选民的民意调查，我们会与报道弗吉尼亚州的人们会面，询问他们想知道的事情。
And then we write the questionnaire.
然后写出一份调查问卷。
We ensure that the questions are balanced and easy to understand, and then we go out into the field.
我们确保问题是中立和谐并易于理解的，然后再进行实地考察。
So going out into the field means actually getting people to take our poll.
实地考察意味着让人们参加我们的民意调查。
We're not selling anything, just doing an opinion poll on some interesting subjects in the news.
我们不出售任何东西，只是就新闻中一些有趣的话题做了一次民意调查。
Do you feel that way strongly or somewhat?
你是强烈地感觉到这种感觉，还是有些感觉？
Do you usually think of yourself as a Democrat, a Republican, an independent, or what?
你通常认为自己是民主党人、共和党人、无党派人士，还是其他什么？
We work with these organizations called "field houses," and they're the people who actually ask the questions, whether that's by telephone or by text message or some other way.
我们与这些被称为“实地之家”的组织合作，他们是真正提出问题的人，无论是通过电话、短信还是其他方式。
They're the people who are getting the responses to our poll questions.
他们是收集民意调查的答案的人。
How are you, sir? Would you support or oppose a nationwide ban...?
你好吗，先生？你会支持还是反对全国性的禁令？
An opinion poll nationwide? Those are all questions I have for you today.
一项全国性的民意调查？这就是我今天想要问你的所有问题。
Thank you so much for your participation and have a great rest of your evening.
非常感谢您的参与，祝您有一个愉快的夜晚。
And the way we figure out who to reach out to to take a poll is based on this really fundamental aspect of polling called "sampling."
而我们决定联系谁进行民意调查，是根据民意调查的这个基本面决定的，称为“抽样”。
And sampling is figuring out that sample of people who we'll reach out to, and those people respond to the poll questions.
抽样就是找出我们要联系的人的样本，这些人会对民意调查的问题做出回应。
And the sample that we reach out to should be representative of the total universe that we're surveying.
我们接触到的样本应该能够代表我们正在调查的整个群体。
So if it's, say, Virginia voters, the sample that we're reaching out to, which is not every single voter in Virginia -- it's just the people that we randomly chose to respond to the survey -- needs to reflect everyone there.
因此，如果是弗吉尼亚州的选民，我们接触到的样本并不是弗吉尼亚州的每一个选民--只是我们随机选择的人--由他们来反映那里的每个人。
So we want to ensure that the population is reflective.
因此，我们希望这些人群具有映射性。
An analogy I like to use is making soup.
我喜欢用做汤来打比方。
If you're making soup for your friends who are coming over, you're not going to taste the entire bowl before they get there, because then you got nothing to serve them.
如果你正在为要到访的朋友煮汤，你不会在他们到达之前品尝整碗汤，那样你就没有东西可以端给他们了。
And you would be very, very full. And you'd be a crappy hostess.
你会吃得非常非常饱。你会是个差劲的女主人。
But if you take a ladle and stir it up so that you can mix it up and make sure that, you know, the salt is all over and that the, you know, the beans are all over and not just in one part... Uh-huh.
但如果你拿起一把勺子把它们搅拌均匀，确保放好了盐，放好了豆子，而不是只在其中一部分放了调料……嗯，嗯。
Like, the carrots. Uh-huh. Totally.
比如，胡萝卜。嗯，嗯。肯定的。
And then take a tiny taste and be like, "Yeah, that's the ticket," and that is sampling.
然后尝一小口，感觉，“没错，就是这个感觉”，这就是抽样。
And you make sure that, in the spoon, that you, like, got a little bean and you got a little piece of carrot, and you make sure that it's got the, you know -- that it's... Everybody's in there.
你要确保，在勺子里，有一颗豆子和一片胡萝卜，这些一定要有--它包括了所有食材。
That everybody's. Everyone's in the spoon. It's a nice mix.
什么都要有。每一种都在勺子里。这才是很好的组合。
It's a little -- a little hot bath of public opinion.
这是--一个小小的舆论热水浴。
And if it needs more salt, you go back, and you get more salt in there.
如果它需要更多盐，你就把它放回去，加更多盐。
Okay. So that helps me understand what a sample is.
好的。这能帮我理解什么是样本。
But then, getting back to this idea of this poll that's out in the field, once you get the responses back from people, then what happens?
但是，回到实地调查这个说法上来，你从人们那里得到回复后，会发生什么？
Then we get a data set that comes back to us, and it just looks like a list of numbers and -- and nonsense to most people.
我们会拿到一个返给我们的数据集，它看起来就像数字列表--对大多数人来说，这就是一堆废话。
And we are able to analyze that data by looking at the different responses, but also breaking them down by crosstabs.
我们可以通过不同的回答来分析这些数据，当然也可以通过交叉表对它们进行分解。
So, at the end of all those issue questions, we also ask people about themselves.
在所有问题的最后，我们还会询问人们关于他们自己的情况。
So we ask them their gender and their zip code and their racial or ethnic identity and their age.
我们会问他们的性别、邮政编码、种族或民族身份以及他们的年龄。
And we're able to group people together by those different demographic groups and see if there are trends or interesting things that pop up with those different groups.
我们可以根据不同的人口统计群体把人们分组，看看这些不同的群体是否存在某种趋势或有趣的现象。
And then we write the analysis up, we make graphics, we usually team ourselves with different reporters in the newsroom to write a story, and then we release it.
然后我们会写分析报告，制作图表，再与新闻编辑室的其他记者合作写出一篇报道，然后发表报道。
