public opinion is the holy grail of democratic politics it's impossible to ignore public opinion but at the same time it's hard to put your hands on it what's going on in the minds of the people that you could see behind me are they thinking about the problems facing the nation are they thinking about the conflicts in the middle east or the growing income gap well probably not more likely they're thinking about where they're headed what they'll do once they get there what they might have for lunch if they're planning ahead what they're going to do this weekend as the journalist walter lipman put it we are concerned in public affairs but immersed in our private ones that's a problem for democratic theory theoretically government in a democracy rests on the will of the people but that assumes that the will of the people is an informed and engaged will that people are interested in politics and so that's one of the puzzles of democratic politics where public opinion fits into the equation how attentive to public opinion should government be how attentive to public opinion is government and what happens when government ignores public opinion [Music] [Applause] the rise of democracy led political theorists to ask what was meant by the phrase the will of the people did that mean officials had to pay heed to people's everyday opinions edmund burke an 18th century british theorist said no he argued that citizens were ill-equipped to make policy judgments and that representatives should decide what was best for them said burke your representative owes you his judgment and he betrays you if he sacrifices it to your opinion jeremy bentham the founder of utilitarian theory thought differently bentham argued that citizens deserve to be heard indeed that public opinion was the best protection against miss rule that view that people's everyday opinions matter gradually prevailed but a practical question remained how was public opinion to be determined was it to be judged by what was said from the pulpit and in the community hall by what newspaper reporters were hearing on the streets by studying election results the introduction of scientific polling in the 1930s more or less settled the question of how public opinion would be determined polls gained early and instant credibility when the first ever gallup presidential poll accurately predicted a landslide victory for franklin d roosevelt over elf landon in 1936. many newspapers had predicted otherwise they had reported voters couldn't wait to get rid of roosevelt as it turned out newspaper publishers were living in an echo chamber they and their wealthy friends hated roosevelt's new deal and had been listening to each other polls are now the dominant method of gauging and researching public opinion in this session we'll examine that research focusing on the nature of opinion polling the attributes of public opinion and the influence of public opinion on officials and public policy let's start with a few observations on polling in a poll a small number of properly selected individuals or as they're called the sample are used to estimate the opinions of an entire population such as the residents of a city or country now how is that possible how can interviews with say a thousand respondents possibly provide a reliable estimate of what millions of americans are thinking the answer is found in the laws of probability consider the hypothetical example of a huge jar filled with a million marbles half of them blue and half of them red if a blindfolded person reaches into the jar the probability of selecting a marble of any given color is 50 50. and if 1000 marbles are chosen in this random way it is likely that roughly half of them will be red and roughly half will be blue opinion sampling works on the same principle if respondents are selected at random they can be used to estimate what the population as a whole is thinking now which do you think is more important in determining the accuracy of a poll do you think it's the size of the sample or do you think it's the size of the population whether the pole is of a city or a state or the nation as a whole it's sample size think again of the jar full of marbles half red half blue as long as the marbles were thoroughly mixed it wouldn't matter whether the jar held 1 million 10 million 100 million marbles in each case the odds of drawing a red marble or a blue one would be 50 50. the size of the sample would matter however if you drew only 10 marbles from the jar you'd stand a good chance of getting six or more marbles of one color rather than five of each if you tried to generalize from a sample of 10 you'd stand a good chance of being wrong on the other hand if you drew a thousand marbles from the jar the odds greatly improved that the sample distribution will come closer to the actual number for example the mathematical odds of drawing as many as 600 marbles of the same color are more than a hundred thousand to one close to the odds of being struck dead by a meteorite falling from the sky the accuracy of a poll is expressed in terms of sampling error this error occurs because a sample is being used to estimate the characteristics of the full population the error refers to the difference between the estimate obtained from the sample and what the actual figure would be if the entire population had been contacted as you would expect the larger the sample the smaller the sampling error in other words the more people surveyed the less likely the poll will be inaccurate most public opinion polls are based on samples between 400 and 1500 respondents a properly drawn random sample of say 1200 respondents has a sampling error of roughly plus or minus 3 percent thus if 55 of such a sample say they intend to vote for the republican presidential candidate it's highly likely that somewhere between 52 percent and 58 that's 55 plus or minus 3 sampling error of the population actually intend to vote for the republican candidate now polls can go wrong for a lot of reasons poorly worded or slanted questions for example can confuse or mislead respondents as well pollsters can only approximate random selection in choosing their respondents most polls for example are conducted by telephone in this case pollsters are sampling telephone numbers rather than people directly nevertheless reputable polling firms have a solid track record the gallup organization for example has polled every presidential election since 1936 a total of 19 elections in all on average the final gallup poll's election prediction has come within two percentage points of the actual result now pollsters ask americans a great many things their feelings about the president their party identification their opinions on issues their religious affiliation and so on here we'll concentrate on policy opinions focusing on three of their attributes direction intensity and salience opinions have direction that is people can favor or oppose a policy a 2014 washington post abc news poll for example asks respondents whether undocumented immigrants living in the united states should or should not be given the right to live and work here legally 46 percent said they should 50 percent said they should not when as in this case the public is divided in its opinions the issue is likely to be a source of partisan conflict on the other hand when opinions are one-sided conflict is usually limited after the terrorist attacks on u.s soil of september 11 2001 the public was largely of one mind on the issue of taking the war to the terrorists there was relatively little debate over whether to invade afghanistan when polled at the time of the invasion whether it was the right thing to do eighty-nine percent said it was a mere nine percent felt the invasion was a mistake now intensity is another opinion attribute intensity is a question of how strongly people feel about an issue during the 2012 presidential campaign for example a pew research center poll asked respondents how important the abortion issue was to their vote very important somewhat important or not at all important the intensity level was much higher on the anti-abortion side of the issue compared with those who believed abortion should be legal in all circumstances those who felt it should be illegal in all circumstances were more than twice as likely to say that the issue was very important in their choice of a candidate intense opinions tend to get lawmakers attention intensity is a sign that people might act on what they believe opinions also have salience which refers to the importance of an issue relative to other issues salience is related to intensity the more strongly people feel about an issue the more likely they are to think it's important but intensity and salience are not the same thing for example people might have an intense opinion about genetically modified food but not see it as a salient issue it could be far down their list of important issues in polls issue salience is often measured by asking respondents what they regard as the country's most important problem here for example are the top responses to that question in a 2015 gallup survey as you can see pocketbook issues the economy jobs wages are at the top of the list virtually no one in the poll mentioned genetically modified food an issue salience matters to elected officials they don't risk much by neglecting low salience issues but they can get themselves in trouble by ignoring highly settled ones some analysts for example have concluded that barack obama made a strategic mistake in pursuing health care reform in the first months of his presidency a time when the economy was uppermost in americans minds obama paid for his decision in the 2010 midterms when democrats lost control of the house of representatives and saw their senate majority dwindle exit polls indicated that a majority of voters were upset with obama's handling of health care and the economy obama's strategy was a sharp contrast with that of democratic president franklin roosevelt during the great depression upon taking office in 1932 roosevelt concentrated on getting americans back to work and stayed on that issue for three years only then did he ask congress to enact social security in both the 1932 midterm and the 1936 presidential elections democrats increased the size of their house and senate majorities let's recap what we've noted so far opinion direction matters if the weight of opinion shifts in one direction officials are more likely to respond to that side but we also need to account for intensity officials are more likely to respond to intensely held opinions and then there's salience as the salience of an issue rises the likelihood that officials will respond increases let's look at the interplay of these three opinion attributes through the lens of a single policy issue gun control for starters let me give you a bit of background on guns and gun policy america has a lot of guns there are nearly as many guns in america as there are people that's the highest ratio in the world france has one gun for every three people england has won for every 16. full disclosure here i'm a gun owner it's a shotgun that i use for bird hunting while in high school in rural minnesota since then it's been locked up unused in my brother's house back home now there are a lot of hunters in the united states 10 million or so as well as uncounted number of skeet trap and target shooters the united states also has a lot of gun deaths about 40 000 per year roughly 15 000 of these deaths an average of one per hour are homicides the rest are suicides or accidental deaths now given the number of gun deaths it's not surprising that over the years attempts have been made to control guns the first major federal laws enacted in the 1930s limited the sale of machine guns a favorite of organized crime in the 1960s the sale of guns to felons and the mentally ill was restricted in the 1990s congress limited the sale of handguns and military-style assault weapons the assault weapons ban had a sunset provision the law would expire in 2004 unless renewed by congress since the 1990s congress has not enacted a single major gun control law in fact national policy has gone the other way when the assault weapons ban came up for renewal in 2004 congress voted it down later congress passed a law protecting gun sellers and manufacturers from being sued if their product was involved in a shooting in 2008 the supreme court got into the act citing the second amendment right to bear arms and overturning a ban on handguns the court did say however that some gun regulations were lawful though it did not list them in 2013 advocates of gun control in congress made a concerted effort to tighten the regulations including reinstating the ban on assault weapons congress voted down that proposal as well as all the others including stricter background checks on gun buyers now where does public opinion fit into this picture what part has it played in what's happened since the 1990s let's look first at the direction of public opinion where americans stand on gun control as you can see from this chart americans are split on the issue of whether gun control should take priority over gun ownership but the chart also shows that a larger number of americans at times twice as many have wanted stricter controls on guns so if we were considering only pro and con opinions we might have expected lawmakers to place additional restrictions on guns rather than fewer which has been the case so let's look at opinion intensity to see if it helps us understand what's happened and here we begin to get at the answer as you can see from this chart gun control opponents are far more willing to act on their opinions they are twice as likely as those favoring gun control to have written a public official or given money in support of their position now let's add opinion salience to the mix for starters gun violence is not a particularly salient issue in most gallup polls of the past two decades one percent or fewer of the respondents have named gun violence as the nation's biggest problem that would suggest that lawmakers can safely ignore the issue knowing that only a small number of people see it as a pressing problem on the other hand gun violence has sometimes become more salient in may 1999 for example 10 percent of gallup poll respondents cited gun violence as the top issue more recently february 2013 gun violence was mentioned by six percent that might seem small but only five issues were mentioned more frequently in the poll so what explains the occasional jump in the salience of gun violence it's mass killings in 1999 a shooting at columbine high school in colorado left 12 students and two teachers dead in late 2012 20 children and six adults were murdered in a shooting at the sandy hook elementary school in connecticut mass shootings get intense media coverage providing gun control advocates an opportunity to push for stricter laws after the shooting at the sandy hook school for example president obama set aside other issues to put pressure on congress to act yet as we emphasized in previous sessions legislation in the united states must work its way through a system of divided power where partisan differences can block action so let's take a look at how republican and democratic voters feel about guns here's the partisan split on the issue of assault weapons from a poll taken after the sandy hook school killings as you can see democrats were heavily in favor of banning assault weapons while only a minority of republicans favored it when the bill to ban assault weapons was considered in congress in 2013 it did not even come up for a vote in the republican-controlled house it was blocked before it reached the floor in the democratic-controlled senate there was a vote but the ban was defeated by 60 votes to 40. every senate republican but one voted against the ban they were joined by 15 democrats now that might surprise you what do you think most of these senate democrats had in common they were from rural states where hunting and gun ownership is widespread rural america as you can see from this figure ranks gun ownership rights ahead of gun control while the opposite is true of urban america let me note one final thing about the gun issue before we move to another topic you might recognize this chart from an earlier session it's based on the costs and benefits of a policy cost and benefits can either be concentrated targeted at a specific interest or diffused spread across society where would you place gun control on this chart it fits best in the diffuse benefits concentrated costs category the benefits of gun control are diffuse because they're hard to pinpoint fewer people will die or be injured if guns are more tightly regulated but it's impossible to know who those people are and since the odds of being a gun victim are small in the first place few individuals will think gun control will make them a whole lot safer on the other hand the costs of gun control are concentrated they restrict the actions of gun owners who do know who they are thus they individually feel the impact of stricter laws the distribution of the costs and benefits of gun control helps explain why intensity is much higher on the anti-gun control side individuals on that side of the issue feel they are bearing the costs of gun control policy it also explains why lawmakers from certain states and districts shy from the issue citizens who oppose gun control are more likely to vote on their opinion than those who favor controls now to this point in the session the discussion is aimed at explaining the attributes of opinion and how they affect the play of politics as yet unaddressed is the most basic question about public opinion how much influence does it have on government it clearly has some influence but what's the magnitude and what are the limits to it as it turns out that's a hard question to answer with any degree of precision policy decisions are rarely the result of a single influence lobbying partisanship media coverage events public opinion all these and more can play into the choice of one policy over another in addition public opinion has contradictory elements people's opinions don't always line up in a logical way for instance majorities routinely say they would like a large cut in their taxes but they also say they want more government services such as better schools and highways they can't easily have it both ways nevertheless considerable research has been done on the impact of public opinion and the findings support some clear conclusions one is that public opinion constrains officials it limits their choices as political scientist vlk expressed it officials must operate within the boundaries of what the public will find acceptable an example is the social security program which has such broad public support that has been called the third rail of american politics a metaphor for a policy that's likely to hurt any politician that tries to mess with it over the years a few have tried in 2005 for example president bush proposed privatizing part of the social security program like the others before him the public response was so strongly negative that bush had to back down during his 2016 presidential campaign donald trump vowed to cut federal domestic spending and promised that every domestic spending program would be subject to review yet when trump proposed his first set of budget cuts as president social security was not on his list the founder of social security president franklin roosevelt predicted that people would fight tooth and nail to protect the retirement benefits they had acquired by paying payroll taxes during their working years said roosevelt no damn politician can ever scrap my social security program now on the other hand there's a wide range of policy decisions most of them in fact where public opinion doesn't even come into play most policy decisions take place outside the public eye in the area of foreign affairs for example there are hundreds of policies on which the public is not remotely aware foreign aid for the ivory coast safety standard agreements with the european union the special tariff on wine from argentina most domestic policy decisions also take place outside of public view how many americans know or care for example that the government has funded research on dairy cow's methane gas emissions policies such as these rather than being decided in the context of public opinion are decided largely through the interest group system a subject we'll examine later in the course now what about those issues that do catch the public's eye those that get a lot of attention from the media and attract wide public interest are officials responsive to public opinion in these cases a number of studies have looked closely at that question they have examined trends in policy and opinion over time seeking to determine whether changes in public opinion are followed by corresponding changes in public policy these studies have found that when public opinion changes so in most cases does public policy for example in a study of 350 high-profile policy opinion relationships spanning a four-decade period political scientist bob shapiro and ben page found that changes in public opinion were normally followed by congruent changes in public policy they concluded when americans policy preferences shift it's likely that changes in policy will follow in other words opinion direction and salience matter as opinion direction on high visibility issues changes so typically does policy opinion intensity also matters as political scientist john kingdom found in his study he concluded that when public opinion is intense and unmistakable as to a preferred course of action politicians nearly always follow it now such findings should not be taken to mean that officials blindly follow public opinion at times public opinion trails policy as opposed to leading it for example the changes in race relations brought about by the 1964 civil rights act helped to change americans opinions on race moreover as political scientists jeff manza and faye lomax cook showed in their study politicians have room to maneuver even when they're responding to public opinion for example although republican and democratic lawmakers alike are responsive to public demands for policy action when the economy goes bad they have different approaches to fixing the economy that's a subject we'll examine in a later session nevertheless decades of research demonstrate that policy on high-profile policy issues usually change in the direction of public opinion rather than the reverse in a recent exhaustive study of the linkage between opinion and policy for example political scientists robert erickson michael mccune and james stimson flatly conclude public opinion influences policy okay let's wrap up this session with a summary we began with a brief look at early theories of public opinion and how the introduction of opinion polling changed the way public opinion is measured and studied we then examined the basics of polling the use of a sample of people as a means of estimating what the public as a whole is thinking noting the importance of random selection and sample size to the accuracy of polls we then studied attributes of opinion opinion direction as an indication of the distribution of opinions on an issue opinion intensity as an indicator of how strongly people feel about their position on an issue and opinions aliens as an indicator of how important people think an issue is relative to other issues all three opinion attributes affect whether officials are likely to take public opinion into account in making policy decisions and we close by reviewing research on the general relationship between public opinion and public policy noting that the link is fairly strong but largely limited to salient issues on which opinion is intense and clear in its direction