04 - Public Opinion

Public opinion is the distribution of individual attitudes toward a particular issue, candidate, or political institution. Although the definition is simple enough, public opinion encompasses the attitudes of millions of diverse people from a myriad of racial, ethnic, age, and regional groups. As a result, the study of American public opinion is especially complex, but also very important. For American government to operate effectively, the opinions of the American public must reach and become an integral part of the political process.

MEASURING PUBLIC OPINION

The measurement of public opinion is a complex process that involves careful interviewing procedures and question wording. To complicate the task further, people are often not well informed about the issues, and may comment on topics they know little about. Public opinion polls must be constructed and executed carefully in order to accurately reflect the attitudes of the American public.

Public opinion polling is a relatively new science, first developed by George Gallup, who did some polling for his mother-in-law, a candidate for secretary of state in Iowa in 1932. Gallup founded a firm that spread from its headquarters in Princeton, New Jersey throughout the democratic world. Today, other well-known private firms conduct polls, and big television networks, magazines and newspapers, such as CNN, Time, and The New York Times, conduct their own polls. Pollsters are also hired by political candidates to determine their popularity, and the results of their polls often shape the direction of political campaigns. The national government even sponsors opinion polls of its own.

Accurate opinion polls must be based on several important principles:

• The sample of those interviewed must be representative of the entire population. Every citizen cannot be polled regarding his or her opinion on a whole range of issues, but those selected must allow the pollster to make accurate assessments of public opinion. The most common technique employed is random sampling, which gives everyone in the population an equal probability of being selected. Most national surveys draw a sample of between a thousand and fifteen hundred persons. The pollster most commonly makes a list of groups, using criteria such as region, age, ethnic and racial groups, gender, and religion. From these groups, people are selected randomly for interviews.
• People must have some knowledge of the issues they are asked about. If the issue is complex (such as American policy toward Bosnia), people should be allowed to say "I don't know." Still, pollsters always must allow for the fact they people often pretend to know things that they don't.
• The structure and wording of the question is very important in obtaining an accurate response. "Loaded" or emotional words should not be used, and the pollster must not indicate what the "right" answer is. The categories of answers also determine the results of the poll. A yes or no question, such as, "Do you think the president is doing a good job?" will give very different results that a question that gives the interviewee a chance to rank the president's performance (excellent, very good, good, average, poor, very poor).
• The same poll conducted with a different random sample almost certainly will produce slightly different results. These slight variations are known as sampling errors. A typical poll of about fifteen hundred usually has a sampling error of + or - 3 percent. This means that 95% of the time the poll results are within 3 percentage points of what the entire population thinks. If 60% of the population supports a candidate for office, in actuality, 57-63% of the population supports him or her. Usually, the larger the sample in proportion to the population, the smaller the sampling error.

FACTORS THAT INFLUENCE POLITICAL ATTITUDES

When pollsters divide people into groups before they conduct random samples, they are acknowledging a well-proven fact: group identifications often influence political attitudes. Several important factors follow:

• The family is probably the most important shaper of political attitudes, particularly of party identification. Polls show that the majority of young people identify with their parents' political party. The process begins early in life (by the age of ten or eleven), and even though an individual becomes more independent as they grow older, the correlation between adult party identification and the parents' party is still very high. A parallel trend, however, is a tendency for this correlation to be lower than it has in the past. This trend may be related to another trend: the growing number of voters who call themselves "independents" rather than Democrats or Republicans.
• A person's gender also influences political views. For example, more women consider sexual harassment in the workplace to be a serious problem than do men, and more men than women tend to support military actions and spending in foreign affairs. Party identification is also affected by gender, but the relationship has shifted through the years. In the 1920s when women first began to vote, they were more likely to support the Republican party than were men. Some experts explain this correlation by pointing out that the Republicans tended to be more the party of "hearth and home" in the 20s. Whatever the explanation, the tendency for women to vote for Republicans continued through the 1950s (Although most women supported the Democrat Franklin Roosevelt over his Republican opponents, the percentage of women supporters was lower than the percentage of men who supported Roosevelt). Since the late 1960s, the correlation has reversed: women are more likely than men to vote for Democrats. This reversal has been explained by the advent of the modern women's rights movement and the Democrats' tendency to support points of view women support: equal opportunity for women, abortion rights, and welfare programs. On the other hand, some experts argue that Republicans are more concerned about defense issues, and thus they attract more men to their party. A more recent gender-related issue has to do with male vs. female support for women political candidates. Although common sense may tell us that women would be more likely to support women candidates, the research does not show a clear correlation.
• An individual's religion is a factor in determining his or her political attitudes. Although the relationships are not as strong as they once were, Protestants are more conservative on economic matters than are Catholics and Jews. Jews tend to be more liberal on both economic and social issues than are Catholics or Protestants. Some special research on "born-again" Christians indicates that they are more conservative than any other religious groups on social issues (such as abortion, civil rights for minorities, and women's rights), but not necessarily on economic issues (such as government services and job guarantees).
• A person's level of education also affects political attitudes, with college education generally influencing the individual to have more liberal social and economic attitudes. Countless studies dating back about fifty years document the correlation between college education and liberalism. The longer students stay in college and the more prestigious the institution they attend, the more liberal they become. The reasons for the correlation are unclear, but some experts believe that the liberal attitudes of professors may influence students. Others believe that the differences lie not in the schooling itself, but in the characteristics of people who attend college vs. those that don't.
• A number of years ago, the relationship between social class and political attitudes was clear: the higher the social class, the more conservative the individual, and the more likely he or she was to belong to the Republican party. Today, that relationship is much less clear, perhaps partly because of the correlation cited above between college education and liberalism. Even though the broad affiliations between blue collar workers and the Democratic party and businessmen and the Republican party still have some credibility, those relationships are much weaker than they once were.
• Much research has focused on the relationship between an individual's race and ethnicity and his or her political attitudes. The oldest and largest numbers of studies focus on black Americans, who tend to identify with the Democratic party and are still the most consistently liberal group within that party. Some research indicates that this relationship may be weakening, but the correlation is still strong. Much less research has been conducted with Hispanic Americans, but preliminary results indicate that they too tend to be liberal and affiliate with the Democratic party, although the correlation appears to be weaker than with black Americans. A very limited amount of research among Asian Americans indicates that they are more conservative than blacks or Hispanics, although attitudes of the various nationalities of Asians fluctuate widely. For example, preliminary research indicates that Korean Americans are more liberal than are Japanese Americans.
• The geographic region in which an individual lives almost certainly affects political attitudes. Stereotypically, white southerners are more conservative than are whites from other regions. However, the relationship is much more complex than it appears to be. For example, although some research indicates that white southerners tend to be less liberal than others on social issues, such as aid to minorities, legalizing marijuana, and rights of those accused of crimes, southern attitudes on economic issues (government services, job guarantees, social security) are very similar to those from other regions. Although there is some evidence that southerners are more conservative than they were forty years ago, political views today of white southerners are less distinct from those in other regions than they used to be. Clearer evidence concerning party affiliations can be seen in the shift since the 1950s of southerners away from the Democratic party. From the time of Reconstruction until the 1950s, the Solid South always voted Democratic. Virtually all representatives, senators, governors, and local officials in the South belonged to the Democratic party. Since the 1950s, more and more political leaders have affiliated with the Republicans, so that today, in most Southern states, both parties have viable contenders for public office. Some experts explain this phenomenon by the support of the Democratic party for black civil rights starting in the 1950s, with the result that many white southerners changed their party affiliation.

POLITICAL IDEOLOGY: LIBERALS AND CONSERVATIVES

A political ideology is a coherent set of values and beliefs about public policy. In U.S. politics, ideologies generally are thought to fall into two opposite camps: liberal and conservative. While there are general guidelines for determining the nature of liberalism and conservativism, the differences between the two are not always obvious.

How Ideological are American Citizens?

The classic study of the 1950s, The American Voter, investigated the ideological sophistication of the American electorate. The authors created four classifications of voters:

• ideologues - 12 % of the people connected their opinions and beliefs to policy positions by candidates and parties. In other words, only 12% of the American voting populations voted along primarily ideological lines.
• group benefits voters - 42% of the people voted for parties based on which one they thought would benefits groups they belonged to or supported. ("Democrats are more supportive of labor union members like me.")
• nature of the times voters - 24% of the people linked good times or bad times (usually based on economics) to one political party or the other and vote accordingly. ("The Republicans can get us out of this recession.")
• no issue content - 22% of the people could give no issue-based or ideological reasons for voting for a party or a candidate. ("_____________is better looking than the other candidate.")

Follow up studies conducted through 1988 reveal some variation in percentages among the groups, with ideologues faring somewhat better than they did in the 50s, but they are still a relatively small group (18% in 1988).

Even though the terms liberal and conservative are more meaningful for the political elite than for the typical voter, the concepts are roughly, if inconsistently, understood by most Americans.

The following table summarizes some of the political beliefs likely to be preferred by liberals and conservatives:

ISSUE	LIBERALS .	CONSERVATIVES
Health Care	Health Care should be more widely available to ordinary people and not necessarily tied to work Tendency to support a national health care system	Health care is best handled by private insurance companies and are most logically tied to work place benefits.
Crime	Cure the economic and social reasons for crime.	Stop coddling criminals and punish them for their crimes.
Business Regulation	Government should regulate businesses in the public interest	Businesses should be allowed to operate under free market conditions
Military Spending	Spend less.	Spend more.
Taxes	The rich should be taxed more; the government is responsible for reducing economic inequality.	Taxes should be kept low.
Welfare State	The government is responsible for helping the poor find employment and relieving their misery.	People are responsible for their own well-being; welfare takes away the incentive to take care of themselves