AP Statistics Quiz: Introducing Statistics Worry About Error

What this quiz covers

This quiz focuses on Introducing Statistics Worry About Error, giving you a quick way to practice the rules, question types, and explanations that matter most for AP Statistics.

How to use this quiz

Try each quiz question before looking at the correct answer. Use the explanations to review missed ideas, then come back to similar questions until the pattern feels familiar.

Question 1

1. Because if the new design worked in the sample, it must work for all future visitors with no exceptions
2. Because the observed difference could be partly due to chance, and a wrong rollout decision could affect revenue and user experience at scale (correct answer)
3. Because random assignment removes all uncertainty from the results
4. Because a confidence interval is the probability that the new design is better
5. Because Type I and Type II errors only apply to medical studies, not business decisions

Explanation: This question examines understanding of experimental error in business decision-making. The company must decide whether to roll out a new checkout design based on A/B test results. Option B correctly recognizes that the observed difference in purchase rates could be partly due to chance (sampling variability) rather than a true effect, and rolling out an inferior design could reduce revenue while missing a superior design means lost opportunities. Success in the sample doesn't guarantee success for all users (eliminating A), random assignment reduces bias but not sampling variability (eliminating C), confidence intervals quantify uncertainty about parameters not probabilities of superiority (eliminating D), and Type I/II errors apply to all hypothesis testing including business experiments (eliminating E). In high-stakes business decisions affecting revenue and user experience, companies must acknowledge that A/B test results are subject to sampling error and might not reflect true population effects.

Question 2

1. Because inference from a sample is uncertain, and approving an ineffective or unsafe drug (or delaying an effective one) can have major health consequences (correct answer)
2. Because once a study is randomized, Type I error cannot occur
3. Because a statistically significant result proves the drug will work for every patient
4. Because the $p$-value is the probability that the drug is ineffective
5. Because considering error is unnecessary if the sample size is larger than 30

Explanation: This question examines understanding of clinical trial error in drug approval decisions. Regulators must decide whether to approve a drug based on trial results showing apparent effectiveness. Option A correctly recognizes that inference from the trial sample is uncertain - approving an ineffective or unsafe drug (Type I error) exposes patients to risks without benefits, while rejecting an effective drug (Type II error) denies patients a helpful treatment. Randomization reduces bias but doesn't eliminate Type I error risk (eliminating B), statistical significance doesn't guarantee universal effectiveness (eliminating C), p-values measure evidence against the null not probabilities of ineffectiveness (eliminating D), and sample size requirements don't eliminate the need to consider error (eliminating E). In pharmaceutical regulation where decisions affect millions of patients, regulators must balance the risks of approving harmful drugs against denying beneficial treatments, recognizing that sample-based trial results might not reflect true drug effects.

Question 3

1. Because an estimate from a sample is exact, so error is not a concern
2. Because the estimate could be too high or too low, and either unnecessary restrictions or insufficient protection could result from a wrong inference (correct answer)
3. Because restricting access will automatically increase the sample size and remove uncertainty
4. Because sampling error and measurement error are the same thing and can be ignored
5. Because statistical inference cannot be used for populations that move

Explanation: This question tests understanding of estimation uncertainty in conservation policy. The wildlife agency must decide on hiking restrictions based on a population estimate from sample sightings. Option B correctly identifies that the estimate could be too high (leading to unnecessary access restrictions) or too low (risking inadequate habitat protection), with both errors having consequences for conservation and public recreation. Estimates from samples are never exact due to sampling variability (eliminating A), restricting access doesn't affect the statistical uncertainty of past estimates (eliminating C), sampling error and measurement error are different concepts and neither can be ignored (eliminating D), and statistical inference can be applied to mobile populations using appropriate methods (eliminating E). Conservation decisions balancing species protection with public access must acknowledge that population estimates based on limited sightings contain uncertainty that could lead to over- or under-protection.

Question 4

1. Because the sample mean can vary from sample to sample, and a wrong decision has financial and customer-safety consequences (correct answer)
2. Because the sample mean equals the population mean whenever the sample size is at least 30
3. Because the only important error is rounding error from the calculator output
4. Because random sampling eliminates all uncertainty, so the decision can be treated as a guarantee
5. Because sampling error and a Type I error are the same thing, so considering both would be double-counting error

Explanation: This question evaluates understanding of why error consideration is crucial when using sample means for business decisions in manufacturing. The decision context involves balancing the risks of shipping defective batteries, which could harm customers, against unnecessarily discarding good ones, which wastes resources. The correct choice is A, emphasizing that sampling variability causes sample means to fluctuate, potentially leading to erroneous shipping decisions with real consequences. Distractor B is misleading because while the central limit theorem applies for large samples, it doesn't make the sample mean exactly equal to the population mean; variability still exists. For a mini-lesson on uncertainty, sampling error arises because a sample is only a subset of the population, so its statistics are estimates that can deviate from true values by chance; acknowledging this helps avoid overconfidence in decisions based on samples.

Question 5

1. Because finding no contamination in the sample guarantees the full shipment is contamination-free
2. Because a small sample can miss rare contamination, so the decision should account for the chance of error and the serious consequences of a wrong conclusion (correct answer)
3. Because error is only a concern when contamination is found, not when none is found
4. Because sampling error can be eliminated by testing only the cleanest-looking packages
5. Because the inspector should ignore uncertainty since random samples always produce the same results

Explanation: This question in AP Statistics illustrates worrying about error in contamination testing, stressing its relevance to food safety and economic decisions. The store's plan to sell the shipment after no contamination in 25 packages risks customer illness if rare issues were missed, or unnecessary waste if overly cautious. Considering potential error is critical because small samples may not detect low-level problems, leading to serious consequences if inferences are wrong. Choice A distracts by asserting no findings guarantee safety, but sampling error allows for undetected issues. In a mini-lesson on uncertainty, absence in a sample doesn't prove absence in the population due to variability, and statistical power assesses detection reliability. This promotes informed risk management in inspections.

Question 6

1. Because the sample defect rate could differ from the true defect rate, and either action (shutdown or no shutdown) has costly consequences if the inference is wrong (correct answer)
2. Because a random sample eliminates all uncertainty about the day's true defect rate
3. Because if 4 defects were found, exactly 4 defects exist in the entire day's production
4. Because sampling error is the same as a machine defect, so recalibration always removes it
5. Because considering error is unnecessary unless the sample size is at least 1,000

Explanation: This AP Statistics question focuses on worrying about error in sample-based defect rates, illustrating how uncertainty influences manufacturing decisions. Management's choice to shut down or continue production involves costs and safety risks if the sample of 60 batteries with 4 defects doesn't reflect the true rate. Considering potential error is key because sampling variability could mean the true defect rate is higher or lower, leading to faulty inferences with consequences like financial loss or customer harm. Choice B distracts by suggesting random samples remove all uncertainty, but they only provide estimates prone to variation. As a mini-lesson on uncertainty, sample statistics like defect rates vary from the population parameter due to chance selection, and statistical inference helps evaluate the likelihood of errors. This approach prevents overreactions or oversights in high-stakes scenarios.

Question 7

1. Because the sample proportion is always exactly equal to the population proportion when the sample is random
2. Because the margin of error reflects possible sampling variability, and being wrong could waste money and damage credibility (correct answer)
3. Because any difference between the sample and population must be caused by biased questions, not chance
4. Because if the sample size is at least 30, the estimate is guaranteed to be correct
5. Because confidence intervals eliminate the possibility of an incorrect decision

Explanation: This question examines understanding of sampling variability in political polling contexts. The city council must decide whether to place a tax measure on the ballot based on a sample survey estimate. The correct answer (B) properly identifies that the margin of error reflects sampling variability - different random samples would yield different estimates of support. If the true support is below 50% but the sample estimate suggests it's above, the council might proceed with a doomed measure, wasting resources and damaging public trust. Choice A incorrectly claims sample and population proportions are always equal, while D wrongly suggests sample size alone guarantees accuracy. Statistical inference always involves uncertainty because we're using partial information (the sample) to make conclusions about the whole (all voters).

Question 8

1. Because any observed difference in sample acceptance rates must reflect a real change in the population
2. Because inference from sample data includes uncertainty, and a wrong conclusion could lead to harmful policy choices affecting students (correct answer)
3. Because the acceptance rate is a percentage, and percentages cannot have sampling error
4. Because if the sample was collected carefully, there is no need to consider the possibility of Type I or Type II error
5. Because considering error only matters after the policy has already been implemented

Explanation: This question assesses understanding of inference uncertainty in educational equity contexts. The university uses sample data to estimate whether a new admissions rubric affects acceptance rates for first-generation students. The correct answer (B) recognizes that inference from sample data includes uncertainty - the observed difference in sample acceptance rates might not reflect the true population change due to sampling variability. Type I error (adopting a rubric that doesn't actually improve equity) fails to help students who need support, while Type II error (rejecting a rubric that would improve access) misses an opportunity to increase educational equity. Choice A wrongly assumes any sample difference reflects real population change, while D incorrectly claims careful sampling eliminates error possibility. Statistical inference involves uncertainty, and when decisions affect student access to education, we must consider the serious consequences of potential errors.

Question 9

1. Because a $p$-value below 0.05 proves the program works for every student in the district
2. Because sampling and random assignment remove all uncertainty from conclusions
3. Because statistical conclusions can be wrong due to chance, and the costs of a wrong decision affect students and budgets (correct answer)
4. Because if the sample mean is higher, the population mean must be higher by the same amount
5. Because confidence levels measure how confident the researchers feel, not uncertainty in the process

Explanation: This question assesses understanding of statistical uncertainty in educational program evaluation. The school district uses sample data to infer whether a tutoring program improves math scores. The correct answer (C) recognizes that statistical conclusions can be wrong due to chance - even with proper randomization, the sample might not perfectly represent the true program effect. Type I error (concluding the program works when it doesn't) wastes funds, while Type II error (concluding it doesn't work when it does) harms students who would benefit. Choice A incorrectly interprets p-values as proof for every student, while B wrongly claims randomization eliminates uncertainty. Statistical inference involves making probabilistic statements, not certainties, and decision-makers must weigh the consequences of potential errors.

Question 10

1. Because the estimated probability is a guarantee of what will happen to this specific applicant
2. Because the estimate can be wrong due to sampling variability and model uncertainty, leading to costly approval or rejection mistakes (correct answer)
3. Because error matters only when the probability is above 50%, not near 5%
4. Because using a probability threshold eliminates the possibility of any decision error
5. Because any wrong decision would be a Type I error, so the bank should ignore Type II errors

Explanation: This question tests awareness of error in model-based probability estimates for financial decisions like loan approvals. In the bank's context, errors could lead to losses from bad loans or missed profits from rejecting good ones, with a 5% threshold amplifying risks. B is correct, as it addresses sampling variability and model uncertainty that could make the estimate inaccurate, causing approval mistakes. Distractor A errs by treating the estimate as a personal guarantee, but probabilities are population-based and uncertain for individuals. Mini-lesson on uncertainty: Predictive models from samples inherit sampling error and may not capture all factors, so considering confidence in estimates prevents overreliance and supports better risk management.

Question 11

1. Because sampling variability could mean true support is below 50%, and acting on an incorrect inference could waste resources or delay needed funding (correct answer)
2. Because if the sample is random, the estimate must equal the population value exactly
3. Because nonresponse bias is impossible whenever a poll is conducted by phone
4. Because the only possible mistake is a Type II error, and it occurs whenever the council holds an election
5. Because a confidence interval is a guarantee that the tax will pass if it is on the ballot

Explanation: This question examines understanding of sampling error in political polling and decision-making. The city council must decide whether to place a tax measure on the ballot based on a sample survey showing majority support. Option A correctly recognizes that sampling variability means the true population support could differ from the sample estimate - if actual support is below 50%, the measure would fail, wasting ballot costs and delaying wildfire prevention efforts. Random sampling doesn't guarantee the sample estimate equals the population value (eliminating B), phone polls can still have nonresponse bias (eliminating C), option D mischaracterizes error types and when they occur, and confidence intervals quantify uncertainty but don't guarantee outcomes (eliminating E). In policy decisions with real costs and consequences, we must consider that our sample-based estimate of public support might overstate or understate true population sentiment due to random sampling variation.

Question 12

1. Because a sample average guarantees the new plan reduces commute times for the entire city
2. Because the observed decrease could be due to sampling variability, and being wrong could worsen commutes and safety outcomes (correct answer)
3. Because error only occurs when measuring commute times with a stopwatch, not when using averages
4. Because if the sample mean decreased, the population mean must also decrease by exactly the same amount
5. Because Type I error and Type II error are the same thing, so the city does not need to distinguish consequences

Explanation: This question addresses the AP Statistics concept of worrying about error in estimates, particularly how sampling variability affects inferences from data like average commute times. The city's decision to implement a new traffic plan citywide carries risks, such as increased accidents if commutes actually worsen despite the sample showing a decrease. Considering potential error is vital because the 2.5-minute drop in the sample of 40 mornings might result from random chance rather than a true improvement. Choice A is a distractor that appeals by suggesting samples guarantee population outcomes, but this ignores that samples provide estimates, not certainties. In a mini-lesson on uncertainty, sampling error arises because different samples from the same population can yield varying results, and without accounting for this, decisions could lead to negative outcomes like safety hazards. Statistical tools like hypothesis tests can assess whether observed changes are likely due to chance or real effects.

Question 13

1. Because statistical significance means the tutoring caused the improvement, regardless of how classes were selected
2. Because the observed improvement could be partly due to chance or study design issues, and a wrong decision could waste funds or remove a beneficial program (correct answer)
3. Because if the sample mean increased, the population mean must have increased by the same amount
4. Because a Type I error occurs when you fail to reject the null hypothesis, and that always saves money
5. Because inference is only for experiments, so no error is possible in this situation

Explanation: This question examines understanding of statistical error in educational program evaluation. The district must decide whether to expand a tutoring program based on a pilot study showing statistically significant improvement. Option B correctly recognizes that the observed improvement could be partly due to chance (sampling variability) or study design issues (like selection bias if classes weren't randomly assigned), and either expanding an ineffective program or cutting an effective one wastes resources and harms students. Statistical significance doesn't prove causation, especially with potential selection issues (eliminating A), sample and population means can differ due to sampling variability (eliminating C), option D incorrectly defines Type I error (it's rejecting a true null, not failing to reject), and inference applies to observational studies too (eliminating E). Educational decisions affecting budgets and student outcomes require acknowledging that pilot results might not reflect true program effectiveness due to statistical uncertainty.

Question 14

1. Because the sample mean is always within 0.01 units of the true mean when the sample is random
2. Because sampling variability means the estimate may be higher or lower than the true mean, so the advisory decision could be incorrect (correct answer)
3. Because any discrepancy between sample and population is a Type II error, not a sampling issue
4. Because the pollutant concentration is fixed, so there is no uncertainty in the estimate
5. Because potential error only matters when the data are collected with a convenience sample

Explanation: This question examines why error matters in environmental sampling for public advisories on pollutants. The context involves economic harm from unnecessary advisories versus health risks from failing to issue them, making accurate estimation critical. Correct answer B points out that sampling variability can make the sample mean deviate from the true mean, risking wrong decisions. A distractor such as A falsely promises precision like being within 0.01 units, but random sampling doesn't guarantee such exactness; variability persists. For uncertainty, a mini-lesson: Samples provide estimates, not exact values, due to chance selection; understanding standard errors and confidence levels allows us to gauge how much trust to place in these estimates for high-stakes choices.

Question 15

1. Because the sample mean is guaranteed to equal the companywide mean if the stores were randomly selected
2. Because potential error accounts for sampling variability, so the true mean wait time could differ enough to change whether the staffing plan is warranted (correct answer)
3. Because error is only important when measuring proportions, not means
4. Because the decision to change staffing is not statistical, so uncertainty should be ignored
5. Because sampling error and measurement error are identical, so there is no reason to think about them separately

Explanation: This question explores why error consideration is key in using sample means for operational changes, like pharmacy staffing. The decision context involves costs of new plans versus customer satisfaction and health if waits are misjudged. Correct choice B notes that sampling variability could alter whether the true mean justifies the plan. Distractor A is wrong because random selection doesn't guarantee equality; variability remains. Mini-lesson on uncertainty: Sample means are estimates prone to error from random variation; calculating standard deviations and intervals quantifies this, ensuring decisions account for possible inaccuracies.

Question 16

1. Because any observed difference in sample means must be caused by the website version and cannot be due to chance
2. Because statistical evidence is never certain, and acting as if the sample result is a guarantee can lead to a costly rollout decision (correct answer)
3. Because error only matters when the two versions have exactly equal sample means
4. Because randomization makes Type I and Type II errors impossible
5. Because potential error refers only to computer coding mistakes, not to inference from samples

Explanation: This question evaluates the role of error in A/B testing for business strategies, such as website changes. In this context, rolling out Version B could boost or harm revenue if the sample difference was chance-based. The correct answer B stresses that statistical results aren't certain, so treating them as guarantees risks costly errors. Distractor A wrongly assumes observed differences are always real in randomized tests, ignoring possible chance effects. Mini-lesson on uncertainty: Even in controlled experiments, random assignment doesn't eliminate variability; p-values and significance tests quantify the chance of error, guiding whether to act on results.

Question 17

1. Because a random sample guarantees the estimate will match the true rider proportion exactly
2. Because the survey estimate is subject to sampling variability, so the true proportion could be on the other side of the 40% cutoff (correct answer)
3. Because sampling error is the same as response bias, and random sampling removes both completely
4. Because potential error only affects the mean, not a proportion
5. Because if the agency repeats the survey, the same estimate must occur, so error is not relevant

Explanation: This question focuses on considering error in survey estimates for service expansions, like transit routes. The decision context weighs costs of adding unneeded routes against denying useful services to riders. Choice B correctly identifies that sampling variability might push the true proportion across the 40% threshold, unlike the sample. A distractor like A mistakenly asserts exact matching from random sampling, but variability ensures samples only approximate the population. As a mini-lesson, uncertainty stems from the fact that proportions in samples fluctuate due to random selection; using margins of error helps determine if an estimate reliably exceeds a cutoff for decisions.

Question 18

1. Because with a sample of 400, the poll result cannot differ from the true parent support in the district
2. Because the decision rule (>60%) converts the poll into a certainty, so error is irrelevant
3. Because the sample proportion is subject to sampling variability, so the true support might be above or below 60% even if the sample is not (correct answer)
4. Because any error would be caused only by biased wording, not by random sampling
5. Because sampling error occurs only when individuals refuse to respond, not when the sample is random

Explanation: This question assesses the skill of appreciating sampling error in polling for policy adoption, such as school attendance rules. In this context, the district risks protests or missing beneficial changes if the sample misrepresents true parent support, especially with a threshold like 60%. Choice C is correct, as it explains how sampling variability could place the true proportion on the opposite side of the cutoff from the sample. A distractor like A wrongly claims that a sample of 400 eliminates all differences, but no sample size completely removes variability, though larger sizes reduce it. As a mini-lesson, uncertainty in statistics comes from sampling variability, where random samples can differ from each other and the population; tools like confidence intervals help quantify this to inform whether a result is reliable for decisions.

Question 19

1. Because random assignment guarantees that the observed difference must be a real treatment effect
2. Because even in a randomized study, chance variation can produce differences, and acting on a wrong conclusion affects patient outcomes (correct answer)
3. Because the only meaningful error is a data-entry mistake, not statistical uncertainty
4. Because if the sample size is large enough, the study result becomes a guarantee about all future patients
5. Because considering error is only necessary for observational studies, not experiments

Explanation: This question probes the need to consider error in experimental results for medical decisions, like treatment choices. The decision context includes potential cost increases and patient harm if the switch to Treatment A is based on a flawed conclusion from the sample. B is the correct answer, noting that even randomized studies have chance variation, so observed differences might not reflect true effects, impacting outcomes. Distractor A is incorrect because randomization reduces bias but doesn't guarantee that differences are real; chance can still play a role. In a mini-lesson on uncertainty, statistical inference involves recognizing that sample results include random noise; hypothesis testing helps evaluate if differences are likely due to chance or actual effects, preventing misguided actions.

Question 20

1. Because the estimate could differ from the true proportion due to sampling variability, and over- or under-reacting could misallocate limited resources (correct answer)
2. Because a point estimate is always exact when the population is large
3. Because the margin of error indicates the maximum possible bias in the sampling method
4. Because considering error is unnecessary once data have been collected
5. Because confidence intervals guarantee the agency will make the correct policy choice

Explanation: This question tests understanding of estimation error in public health resource allocation. The agency must decide whether to increase restaurant inspections based on a sample estimate of violation rates. Option A correctly identifies that the sample proportion could differ from the true population proportion due to sampling variability - if the estimate is too high, resources are wasted on unnecessary inspections; if too low, violations go unchecked. Point estimates are never exact due to sampling variability (eliminating B), margin of error reflects sampling variability not bias (eliminating C), considering error remains important after data collection for decision-making (eliminating D), and confidence intervals quantify uncertainty but don't guarantee correct decisions (eliminating E). When allocating limited public health resources, agencies must recognize that sample-based estimates of violation rates are uncertain and could lead to over- or under-investment in enforcement.