Skip to main content

POL 200: Analyzing Politics: Stats and Polls

Which Polls Should You Believe?

As with all information sources, you need to look at poll data critically. Where did it come from? Who was polled? What methods were used? What might make you question its results?

Nate Cohn recently outlined some of the issues for interpreting election polls in a New York Times article, "The Savvy Person's Guide to Reading the Latest Polls." Or see "Harry's Guide to 2016 Election Polls" from 538/On the Media.

Find this and additional Breaking News Consumer Handbooks at On the Media's website.

In the news:

"Pollsters Scramble as Fewer People Take Their Phone Calls" Ryan Knutson in the Wall Street Journal.

Sources for Opinion Polls

Strategies for Finding and Using Stats

There is a lot of data out there. To find and use it, ask yourself: 

WHO would be interested in collecting it? Exmples include government agencies, interest groups, trade associations, think tanks, researchers studying the issue . . . If you don't know, look at sources for statistics in secondary sources such as the Statistical Abstract or in the references at the end of an article.

WHERE would it be published? Does the creator of the data set make it free online? Is it historical, so likely to be published in print? 

WHEN was it published? What might have been going on in the world that would affect either how it was collected or how you might interpret it?

HOW was it gathered? Knowing what populations were included, what method was used for reaching those populations, when it was gathered and so on will help you assess the value and the limitations of the data. 

WHAT will you use it for? Make sure when you include social statistics in your work that you represent the data accurately and fairly. 

Statistics

Political Analysis from FiveThirtyEight

Loading ...

Visualizing Data

There are some nifty tools for visualizing data. Geographic Information Systems combine data with geo-location information to create maps. Other tools can show change over time. As with any use of information, use your best judgement. Make sure you know the limitations of the data itself and make sure that it's being repreesnted in a way that doesn't distort it. 

H/T to Kristin Partlo of Carleton whose awesome guide to statistics and data is awesome. 

An Example of a Data Visualization

This data is from an analysis conducted by Clio Andris et all, "The Rise of Partisanship and Super-Cooperators in the U.S. House of Representatives," PLoS One April 21, 2015 http://dx.doi.org/10.1371/journal.pone.0123507 with data from the Office of the Clerk of the U.S. House of Representatives collected by GovTrack.

Recommended Reading

Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 Unported License