Is Nate Silver right?

In this Politico article, “Is Nate Silver right?”, Steven Shepard analyzes the famed statistician’s election polling methods on Shepard indicates how polls are intended to “bring order to the chaos” and provide a unbiased position on the election status. As someone who is very up to date in the election and checks multiple polls (including 538’s), I thought this article was an interesting read. Especially with 538 gaining so much ground in the recent election, Shepard’s analysis of Silver’s method is timely and relevant.

While I was glad that Shepard wrote this article, I do have critiques on how useful it actually was. The title itself is a question: “Is  Nate Silver right?”. However, do you feel that this question was actually answered in the article? Shepard also used a rhetorical question in the second paragraph. Do you think this usage helped strengthen or weaken the lede and nut graph?

The number of polling organizations that exist can be confusing and Shepard named a bunch throughout the article. Who do you think the audience was? Do you think he clearly described the status of polls in this election for those who are not up to date?

In terms of sources, Shepard only quoted two: a Huffington Post polling editor and Nate Silver himself. Do you think quoting Silver strengthens or weakens the argument? There also were not any polling experts consulted aside from those who were affiliated with current polls. Does this make Shepard’s argument less reliable? Would the article have been stronger if Shepard used expert opinions instead of simply comparing between existing polls?

By showing the history of 538’s polls within the past month, I think it helped clarified how much their stance on Clinton has changed. But did the article leave you wanting more? Do you think Shepard focused on the core question or did he go off track?



About Erin Dolan

I am a junior at the University of Michigan with an International Studies major and Sustainability minor.

7 Responses to “Is Nate Silver right?”

  1. This article argued a very interesting point about the poll outcome discrepancies between various election polls. However, I don’t think it was very comprehensible to people who are not up-to-date on the election polls. Shepard throws around a lot of numbers and they don’t seem to be very effective in supporting his main argument regarding the discrepancies in the election outcomes in various polls. These statistics are overwhelming and at times distracting from the overall message of the article.

    Additionally, it might have been beneficial to quote an opinion of a statistician in this article to get a professional opinion on which way of calculating poll outcome would be best. This would have given this article more credible and perhaps guided readers that are not closely following the election in a direction that helps them understand all these numbers and how they should be interrupted.

  2. I thought the rhetorical question in the second paragraph did a nice job of setting up the article in terms of the content of the article. Also, the article does respond to the question because it provides explanations for the discrepancies in predictions- the various speeds and types of models. Since the election hasn’t occurred yet, the article offers probable outcomes but can’t definitively answer the question the title proposes, yet did give examples of multiple polling results that were in consensus with Silver’s opinions.

    The article left me wanting to more about RedClearPolitics’ methods of predictions, since the article stated that this, along with Huffington Post, is the top polling aggregation site. The opinions of an expert or employee from this polling website would have added a better comparison since the differences would have been between two equally-popular websites. Overall, the article was simple to follow and provided a concise summary of current polling predictions. Therefore, the intended audience could have been anyone interested in the election, without having to seek other sources.

  3. I think this article presents a lot of interesting information, but does not necessarily answer the question. It mentions how arbitrary the polls can be, making it hard to determine which method of polling is most effective/accurate. I agree that there could have been more quotes from professionals – maybe an outside perspective on which forms of polling have been more successful in the past.
    I do think that quoting Nate Silver is beneficial to the article. It provides specific insight to the very question being asked, getting one side of the story. But I agree that the experts may have helped. The tossing of numbers and the discussing of the numbers shifting seemed less helpful than perhaps a more longitudinal study of polls and their success in the past.

  4. I think that the title of this article is a really important piece to the overall impact of the article. If you do not recognize the name Nate Silver, there is a small chance you will click on it. That already narrows down the audience of this article dramatically. Not only to educated people who care about the election, but to people who are interested in polling analysis enough to read this article.

    But, there is nothing wrong with a piece that is directed at a small portion of people. I think the title is intriguing as a question. I think its an example of good journalism that this article does not explicitly answer the question of whether he is right or not. Its giving informed information about both sides and allowing the reader to make a well informed decision.

    I do not feel like the lack of sources in this article was a weakness considering it was an analysis done by the reader. The author uses a large amount of sources for comparison for analysis and I think that is the important part to making this article credible.

  5. I agree with the general consensus that the article did not answer the question posed in the title: Is Nate Silver right? However, I did not think that this was necessary to write a good story. I thought that the title was used to get the reader to start thinking about the validity of polling and whether or not they are completely accurate of the political climate. This is a topic that the article did a good job answering, through their analysis of a variety of polling processes, which added good depth to the article.

    However, I do think the article failed to address the original premise of why poll, or even how polls are conducted. As someone who is not very familiar with the polling process or their use, it would have been beneficial to add a bit of background to the this article. It seems like the the target audience of this piece was more informed about this process due to where it was published. Yet, I still think that making the article more accessible would increase readership and inform more people about an interesting topic relevant to political elections.

  6. Now that the election is over, I thought I would share this piece about how and why Nate Silver ultimately got it wrong:

  7. I am so proud of this discussion, which smartly foreshadowed the biggest coverage issue facing journalists after the election. Many are analyzing how the failure to see a Trump win will change journalism.

    Here’s one story on that topic from a journalism training center called the Poynter Institute:

    This is Slate’s take on how journalists should learn from mistakes made covering the election and adjust coverage of the Trump presidency:

    Finally, here’s a good look by the local online journalism nonprofit, Bridge, about how Trump’s election may impact Michigan:

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s