Previous: Binary and Multiple Stars: Crash Course Astronomy #34
Next: Metabolism & Nutrition, Part 2: Crash Course Anatomy & Physiology #37



View count:686,920
Last sync:2024-04-04 05:30


Citation formatting is not guaranteed to be accurate.
MLA Full: "Public Opinion: Crash Course Government and Politics #33." YouTube, uploaded by CrashCourse, 3 October 2015,
MLA Inline: (CrashCourse, 2015)
APA Full: CrashCourse. (2015, October 3). Public Opinion: Crash Course Government and Politics #33 [Video]. YouTube.
APA Inline: (CrashCourse, 2015)
Chicago Full: CrashCourse, "Public Opinion: Crash Course Government and Politics #33.", October 3, 2015, YouTube, 09:50,
So today, Craig is finally going to start talking about politics. Now up until this point we've specifically been looking at government - that is answering the questions of who, what, and how in relation to policies. But politics is different in that it looks at why certain policies are made. We're going to start today by looking at public opinion - specifically how the public does (and does not) influence our elected officials.

Produced in collaboration with PBS Digital Studios:

Support is provided by Voqal:

All attributed images are licensed under Creative Commons by Attribution 4.0
Want to find Crash Course elsewhere on the internet?
Facebook -
Twitter -
Tumblr -
Support Crash Course on Patreon:

CC Kids:
Hello, I'm Craig and this is Crash Course Government and Politics and today we're going to begin our discussion of politics rather than government.  Aren't they the same thing, Stan?  Aren't they?  They're not the same. Oh.

I know some of you are saying that we've been talking about politics all along, and in a sense, that's true.  But for the rest of this series, we'll be looking more closely at policies and the factors that influence how they're made, rather than the institutions and structures that make them.

One way to think about this is that government describes the what, the who, and the how of policies, and politics describes the why.  Don't ask me about the where or the when journalism students.  Actually, don't ask me anything, because you..I won't hear you.  This is a You-Tube video.

Another way that I like to think about politics is that following it is like following sports.  With any political event, whether an election or a congressional vote or a Supreme Court decision, you can spend time analyzing and predicting what might happen.  And then, after the fact, you can analyze why your prediction was correct or way off base.  Just like what happens after a big game or race or whatever you choose to follow. 

This is getting very conceptual and today we're going to focus on one particular aspect of politics that looms large in America--Godzilla.  No.  Public opinion.


Public opinion can refer to a lot of things, but one useful definition is that it refers to "How a nation's population collectively views vital policy issues and evaluates political leaders."  

Public opinion matters in America, especially because it's a democracy, which classicists out there will know comes from the Greek word demokratia, which means "ruled by the people."  That's not a drug for balding men?  Oh, that's something else.  And as anyone who's been forced to learn the Gettysburg Address knows, like Abraham Lincoln, America's a government of the people, by the people and for the people.  

So, what the people think, especially about how the government should govern, matters, but it also raises some important questions.  Namely, how do the people express what they want?  How does or should the government respond to the people?  And, the one we'll start with, what if the people don't know what they want or are just plain ignorant?

The framers of the Constitution were somewhat skeptical of the ability of the average American to understand and influence public policies, so they gave Americans direct influence over only one  part of the government, the House of Representatives.  This view that the ignorant masses were not to be fully trusted with the hard work of governing won out over the antj-Federalist view that more popular participation was better, but is it justified?

Many people, including a lot of political scientists, say it's justified.  Public issues are complicated and many people, most of the time, are either uninterested or confused by them.  This isn't necessarily a bad thing, especially for those who see disengagement from politics are an example of rational ignorance.  Given the high cost of being informed, it makes good sense to stay less informed.  

There have been a number of books that show just how uninformed Americans can be.  The most notable was The American Voter in 1960, which showed us how little most Americans knew, or cared, about politics and suggested that people's opinions are so changeable and random that the authors concluded that "most people don't have real opinions at all."  Wow.  I have no opinion about that. 

Oh, and if you're thinking that's fine, but in 1960, Americans had so much less information available to them.  They didn't even have color back then, and everyone wore hats.  Everyone wore hats then.  Today, we have the internet, and 24-hour TV news, but here's a statistic--in 1960, 47% of people were unable to name the member of the house who represented them.  In 2010, it was 59%.

On the other hand, there are political scientist that would argue that looking at individual voters and their responses to questionnaires is the wrong way to go.  For writers like Benjamin Paige and Robert Shapiro, authors of The Rational Public, the key is to look at the collective opinion.  

If you take large numbers of Americans, and aggregate their opinions, you find that they are much more coherent and stable and reflect reasonable judgements about politics and government.  Next time you disagree with me and call me crazy, Stan, just aggregate my opinion.  You will find it doesn't vary so much.

Closely related to this idea of large groups of people basically getting things right about politics, is Condorcet's Jury Theorem, which demonstrated that while only one juror had slightly a better chance of determining a defendant's guilt or innocence than a coin flip, a larger group of jurors would produce a majority that would be more likely than not to get the case right.  

James Surowiecki summed it up well in his book, The Wisdom of Crowds, arguing that "even if one voter does not have clear political views, a larger group, taken together, adds up to a rational public."

So, assuming that like Lincoln, we actually want public opinion to influence government, we need to take into account a few things.  First, we should have a reasonably good idea that the people know what they want.  Second, the people should be able to communicate what they want to government officials.  And third, the government should pay attention to the public's desires and respond accordingly.  

All three of these conditions can provide interesting problems of their own.  Even if you agree with the rational public idea and agree that the population as a whole does have coherent political views, the chances are good that what the public wants consists mostly of generalities and are difficult to turn into actual policies.

For example, after the 2008 financial crisis, there was a general anger with Wall Street banks.  But different polls on the issue showed no consensus about what to do about things like executive compensation, or regulating complex financial transactions.  It's difficult to say that the resulting Dodd-Frank Bill represented expression of the popular will.    

The public communicates what it wants in a number of ways.  Most obviously, voting.  But let's just say that people have other ways than election results of letting their voices be heard.  Or their punches.  But don't do that.  That's a fake eagle.  Don't worry about it.

Even though politicians often claim that winning an election gives them a mandate to govern, a quick look at the unpopularity of ObamaCare suggests that an election win doesn't often translate into solid support of a candidate's policies.  Sometimes this lack of support is due to the fact that politicians don't exactly respond to public opinion.

National campaigns spend around $1 billion a year on polling, but it doesn't mean that politicians do exactly what the polling suggests, and they often deny that polls influence their decisions.  Even as poll-conscience a politician President Bill Clinton didn't always do what the American people said they wanted.  For example, in 1994 the public was solidly against a plan to bail out Mexico with a multi-billion dollar loan.  But Clinton pushed through an Executive Order making the loan anyway, because his advisors said this was good economic policy.  

More often, politicians use public opinion polling to shape their responses to issues.  Rather than defining the issue for the politicians, the polls are used to help them craft a message that will be more acceptable to the public.  And public opinion polling certainly has a role in setting the policy agenda by informing politicians of the issues that seem to matter to Americans in the first place.  

So, in addition to voting and election results, polling is also American can let politicians know what they want.  For instance, whether or not they approve of the President's performance, or specific policies, like whether the government should allow an oil pipeline to be built.  

Politicians, and especially journalists, rely on these polls, but before you go jumping on that band wagon, there are a few things you should know about public opinion polling. And don't just go jumpin' on strange band wagons.  Let's go to the Thought Bubble. 

The first thing you remember when you hear or read some polling data is that there are lots of ways that polls can be wrong.  But there are some questions you should ask before accepting the data.  There are a lot of things that can skew the results of polls; some of which are obvious, and others which are more obscure.  

The biggest questions to ask about a poll is how many respondents were there, and how were they chosen?  It's impossible to get responses to any questions from all 320 million Americans, so pollsters rely on statistical sampling.  In order to get a reliable sample, the magic number for pollsters is between 1000 and 1500.  The smaller the number, the less reliable the results are likely to be.  A poll that's based on a sample that's to small, may suffer from a sampling error.  You can sometimes deduce a size of a polls sample from its margin of error.   

A poll with a small sample will have a large margin of error.  In general, for national public opinion polls, the margin of error will be plus or minus three points.  This means that if the poll says that 53% of people support Policy X, it's better to say that between 50 and 56% of respondents supported it.  But that's just a little math, for fun.  

Polling organizations like Harris, Pew, and Gallup, also strive to make sure that their respondents are a representative sample, free from selection bias.  Selection bias occurs when the people polled are not a representative sample of the population.  Say if they're disproportionately white, or rich, or brownies.  

The classic example of a selection bias error was the 1936 Literary Digest poll that predicted Alf Landon would defeat FDR.  It turns out that Literary Digest's readership, were disproportionately wealthy and Republican.  Another, more recent, source of selection bias is that polls which rely on random digit dialing of landline phones tend to under-count younger people, many of whom only have cell phones.  

Selection bias is a particular problem with online polls.  Anyone who takes an online poll has, by definition, logged into a website and is, therefore, not randomly selected.  Although, news organizations like to report their own polling--CNN, I'm looking at you--you should take these poll numbers with a boulder of salt.  Thanks Thought Bubble.

In addition to demographic factors like age, ethnicity, race, and income level, all of which can influence polls, when the questions are asked matters a lot.  Sometimes these two factors interact.  A poll taken on a Friday evening is likely to include a lot fewer young people responding to it.  Especially me, because every Friday night I like to go out and get my swerve on, which implies that I don't go out.  I haven't gone out since 2003.  

More significant in terms of election polling is how close the poll was to the actual election.  The closer the poll, the more accurate.  Polls taken immediately after the election, called exit polls, can be very unreliable.  And polls taken a few days after the election have limited predictive value.  In fact, just get over it.  The election's over.  Just stop polling.

One of the most important ways that polls can be skewed is through the questions themselves.  Ambiguous or poorly worded questions can result in a failure to identify the true distribution of opinion in a target population.  Quick poll--Do you not not not not unlike Crash Course?  Or me as a host?  Let me know in the comments.

The way questions are framed can change the results of polls.  For instance, respondents are much more favorable to policies that "promote free trade" than those which "destroy American jobs".  

So I want to leave you with the question we started with, in an American democracy, how much should public opinion matter in terms of the way the country is actually governed?  Has your answered changed now that you have more of a sense how informed or uninformed Americans are about politics?  Did you even have an answer before?  Are you even listening?  

And if you think that politicians are right to respond to the public's desires, are you convinced that our public leaders have a good sense of what Americans really want?  I'd be interested to know if your own opinions on these questions change over time.  But polling's expensive, so just let us know in the comments.  Thanks for watching.  See you next time.  

Crash Course Government and Politics is produced in association with PBS Digital Studios.  Support for Crash Course US Government comes from Voqal.  Voqal supports non-profits that us technology and media to advance social equity.  Learn more about their objectives and missions at  Crash Course is made with the help of all of these nice pollsters.  Thanks for watching.