Public opinion is the collection of attitudes and preferences citizens hold about government, policies, and public figures. Because no one can interview an entire nation, we estimate opinion by surveying a small slice and generalizing. Whether the estimate is any good depends entirely on the method.
A poll is scientific when it is built so a small group accurately represents a much larger one.
Because a poll surveys a sample, its result is an estimate. The margin of error (sampling error) measures how much the sample result could differ from the true population value purely by the chance of who was drawn.
A poll reporting "52% approval, margin of error ±3 points" means the true value most likely lies between 49% and 55%. Bigger samples shrink the margin — but slowly, with the square root of sample size, so cutting the margin in half takes about four times the sample. Crucially, the margin captures only sampling error. It says nothing about a biased sample, a leading question, or respondents who lie.
Is a "lead" real? If Candidate A polls 49% and Candidate B polls 46% with a ±3 margin, apply the margin to each: A ranges 46–52%, B ranges 43–49%. The ranges overlap, so this is a statistical tie. Rule of thumb: if the gap between two figures is within the margin of error, treat the race as tied.
A non-scientific poll abandons random sampling. Self-selected online or call-in polls let people choose to participate, so only the motivated respond — a large biased sample is still biased.
Sampling error vs. sampling bias: sampling error is random, shrinks with a bigger sample, and is measured by the margin. Sampling bias is a systematic skew from a non-representative sample; a bigger sample does not fix it.
Public opinion influences policymakers, who track polls to gauge what voters will reward or punish. But opinion is often uninformed, unstable, or contradictory (people want lower taxes and more services), and the framers built a republic, not a direct democracy, so elected representatives filter opinion rather than obey each week's poll.
Most American ideological disagreement runs along two separate dimensions.
The two dimensions are independent — someone can want an active government in the economy but minimal government in personal life, or the reverse.
| Position | Economic dimension | Social dimension |
|---|---|---|
| Modern liberal | Active government | Limited government (permissive) |
| Modern conservative | Limited government | More active (traditional values) |
| Libertarian | Limited government | Limited government |
| Centrist | Mixed | Mixed |
| Populist | Varies (a style of "people vs. elites," left or right) | Varies |
[GRAPH: two-axis diagram with the economic dimension on the horizontal axis (limited vs. active government) and the social dimension on the vertical axis, plotting liberal, conservative, and libertarian positions in separate quadrants to show the axes are independent.]
Fiscal policy (the government's taxing and spending, set by Congress and the president) is where ideology shows most directly — the Keynesian-versus-supply-side argument is the tax-and-spend debate. Monetary policy (control of the money supply and interest rates) is set by the Federal Reserve, an independent body, and is insulated from day-to-day ideology by design, though debate persists over whether the Fed should prioritize fighting inflation or maximizing employment.
Adult context. When Congress debates whether to cut taxes and trim spending or raise taxes and expand programs, that is fiscal policy and it maps straight onto the economic dimension. When the Fed raises interest rates to cool inflation, that is monetary policy — a different lever, pulled by a different, independent institution.
1. C. Fix: A scientific poll requires a random (probability) sample so results can generalize. (A) size cannot rescue a non-random method. (B) one party is unrepresentative. (D) an open form is self-selected. (E) landline-only misses many voters. Rule: no random sample, no science — regardless of size.
2. D. Fix: A click-to-vote poll is self-selected, so only motivated visitors respond — a biased sample. (A) and (B) treat size and margin as cures for bias; they are not. (C) strong feelings do not make a sample representative. (E) the flaw is the sampling method, not the question. Rule: self-selected respondents = biased poll, no matter how many.
3. E. Fix: With a ±3 margin, the true value most likely falls within 51–57% (54 ± 3). (A) ignores sampling uncertainty. (B) and (D) fall outside the band. (C) wrongly denies that samples can estimate. Rule: true value ≈ reported figure ± margin.
4. C. Fix: The 2-point gap is within the ±3 margin, so the ranges overlap — a statistical tie. (A) and (B) overstate a within-margin gap. (D) closeness does not make a poll unscientific. (E) a large sample would help, not hurt. Rule: gap smaller than the margin = statistical tie.
5. A. Fix: Precision depends on the number sampled, not the population share, so ~1,000 suffices for any large population. (B) is false; views vary widely. (C) is arithmetically false (1,000 is far under 1% of 330 million). (D) no such law exists. (E) larger samples reduce error, not add bias. Rule: it is the count sampled, not the fraction, that sets precision.
6. B. Fix: "Don't you agree that wasteful spending should be cut?" is a leading question with loaded framing. (A), (C), (D), and (E) use neutral, balanced wording. Rule: loaded adjectives and "don't you agree" signal a biased question.
7. B. Fix: A systematic skew that more respondents cannot cure is sampling bias. (A) sampling error is random and does shrink with size. (C) the margin measures error, not bias. (D) random selection is a method, not a flaw. (E) a benchmark poll is an early baseline poll. Rule: if more people won't fix it, it's bias, not error.
8. D. Fix: Limited government on both the economic and social dimensions is the libertarian signature. (A) liberals want active economic government. (B) conservatives take a more active social stance. (C) populism is a "people vs. elites" style, not this package. (E) centrists hold mixed positions. Rule: "out of the boardroom and the bedroom" = libertarian.
9. E. Fix: Lower taxes, lighter regulation, and restrained spending to spur investment is supply-side, limited-government economics. (A) Keynesian reasoning favors government spending. (B) modern liberalism favors active economic government. (C) this is an economic stance, not social policy. (D) this is fiscal, not monetary, policy. Rule: cut taxes and spending to grow the private economy = supply-side.
10. A. Fix: Increasing government spending to boost demand in a downturn is Keynesian / demand-side reasoning. (B) supply-side favors tax cuts, not spending. (C) monetary tightening is the Fed's tool, not fiscal spending. (D) laissez-faire opposes intervention. (E) a period effect is an opinion concept, not a fiscal policy. Rule: government spends to lift demand in a slump = Keynesian.
11. D. Fix: Monetary policy — interest rates and the money supply — is set by the Federal Reserve, an independent body. (A) Ways and Means handles taxation (fiscal policy). (B) the president cannot set rates by order. (C) the Court does not make monetary policy. (E) states do not control the money supply. Rule: interest rates and money supply = the Fed, not Congress.
12. E. Fix: An active-economics, traditional-values voter shows the two dimensions are independent and can diverge. (A) not every voter fits one pole. (B) the axes do not always align — this voter proves it. (C) the voter clearly holds beliefs. (D) libertarians are not a majority of either party. Rule: economic and social dimensions are separate axes that can point opposite ways.
1. C. Fix: A scientific poll requires a random (probability) sample so results can generalize. (A) size cannot rescue a non-random method. (B) one party is unrepresentative. (D) an open form is self-selected. (E) landline-only misses many voters. Rule: no random sample, no science — regardless of size.
2. D. Fix: A click-to-vote poll is self-selected, so only motivated visitors respond — a biased sample. (A) and (B) treat size and margin as cures for bias; they are not. (C) strong feelings do not make a sample representative. (E) the flaw is the sampling method, not the question. Rule: self-selected respondents = biased poll, no matter how many.
3. E. Fix: With a ±3 margin, the true value most likely falls within 51–57% (54 ± 3). (A) ignores sampling uncertainty. (B) and (D) fall outside the band. (C) wrongly denies that samples can estimate. Rule: true value ≈ reported figure ± margin.
4. C. Fix: The 2-point gap is within the ±3 margin, so the ranges overlap — a statistical tie. (A) and (B) overstate a within-margin gap. (D) closeness does not make a poll unscientific. (E) a large sample would help, not hurt. Rule: gap smaller than the margin = statistical tie.
5. A. Fix: Precision depends on the number sampled, not the population share, so ~1,000 suffices for any large population. (B) is false; views vary widely. (C) is arithmetically false (1,000 is far under 1% of 330 million). (D) no such law exists. (E) larger samples reduce error, not add bias. Rule: it is the count sampled, not the fraction, that sets precision.
6. B. Fix: "Don't you agree that wasteful spending should be cut?" is a leading question with loaded framing. (A), (C), (D), and (E) use neutral, balanced wording. Rule: loaded adjectives and "don't you agree" signal a biased question.
7. B. Fix: A systematic skew that more respondents cannot cure is sampling bias. (A) sampling error is random and does shrink with size. (C) the margin measures error, not bias. (D) random selection is a method, not a flaw. (E) a benchmark poll is an early baseline poll. Rule: if more people won't fix it, it's bias, not error.
8. D. Fix: Limited government on both the economic and social dimensions is the libertarian signature. (A) liberals want active economic government. (B) conservatives take a more active social stance. (C) populism is a "people vs. elites" style, not this package. (E) centrists hold mixed positions. Rule: "out of the boardroom and the bedroom" = libertarian.
9. E. Fix: Lower taxes, lighter regulation, and restrained spending to spur investment is supply-side, limited-government economics. (A) Keynesian reasoning favors government spending. (B) modern liberalism favors active economic government. (C) this is an economic stance, not social policy. (D) this is fiscal, not monetary, policy. Rule: cut taxes and spending to grow the private economy = supply-side.
10. A. Fix: Increasing government spending to boost demand in a downturn is Keynesian / demand-side reasoning. (B) supply-side favors tax cuts, not spending. (C) monetary tightening is the Fed's tool, not fiscal spending. (D) laissez-faire opposes intervention. (E) a period effect is an opinion concept, not a fiscal policy. Rule: government spends to lift demand in a slump = Keynesian.
11. D. Fix: Monetary policy — interest rates and the money supply — is set by the Federal Reserve, an independent body. (A) Ways and Means handles taxation (fiscal policy). (B) the president cannot set rates by order. (C) the Court does not make monetary policy. (E) states do not control the money supply. Rule: interest rates and money supply = the Fed, not Congress.
12. E. Fix: An active-economics, traditional-values voter shows the two dimensions are independent and can diverge. (A) not every voter fits one pole. (B) the axes do not always align — this voter proves it. (C) the voter clearly holds beliefs. (D) libertarians are not a majority of either party. Rule: economic and social dimensions are separate axes that can point opposite ways.