A simple random sample selects 150 of 750 units. What is each unit’s inclusion probability?
A 0.05
B 0.15
C 0.20
D 0.50
Answer: C. 0.20. For equal-probability sampling, the inclusion probability is n/N = 150/750 = 0.20.
A population SD is 12 and n is 36. What is the mean’s standard error?
A 2
B 3
C 6
D 12
Answer: A. 2. The standard error is Ο/βn = 12/6 = 2.
If sample size rises from 64 to 256, how does mean standard error change?
A It doubles
B It is unchanged
C It quadruples
D It is halved
Answer: D. It is halved. Standard error varies as 1/βn; quadrupling n halves the standard error.
For p=0.40 and n=100, what is the approximate proportion standard error?
A 0.020
B 0.049
C 0.060
D 0.240
Answer: B. 0.049. SE = β[p(1βp)/n] = β(0.24/100), about 0.049.
What does a sampling distribution describe?
A Statistic across repeated samples
B All population measurements
C One selected unit
D A survey questionnaire
Answer: A. Statistic across repeated samples. It is the probability distribution of a statistic over repeated samples of the same size.
What is the mean of the sample-mean distribution under random sampling?
A Sample range
B Population variance
C Population mean
D Sample size
Answer: C. Population mean. The expected value of the sample mean equals the population mean.
When is a sample-mean distribution normal for any sample size?
A Sample is convenient
B Population is normal
C Frame is large
D Response is complete
Answer: B. Population is normal. For a normally distributed population, the sample mean is normally distributed for any n under the usual random-sampling conditions.
Can a larger random sample correct a frame that excludes a whole region?
A Yes, always
B Only with replacement
C Only for means
D No, coverage remains missing
Answer: D. No, coverage remains missing. Increasing n does not repair undercoverage in the sampling frame.
What creates selection bias?
A Random sample fluctuation
B Larger sample size
C Nonrandom selection pattern
D Repeated measurements
Answer: C. Nonrandom selection pattern. Selection bias occurs when the selection process systematically favors or excludes units.
When is nonresponse bias a concern?
A Every unit answers
B Responders differ systematically
C Frames are current
D Samples are random
Answer: B. Responders differ systematically. Nonresponse can bias estimates when responders differ from nonresponders on relevant characteristics.
A faulty scale adds 2 kg to every reading. Which error is present?
A Measurement error
B Sampling error
C Coverage error
D Nonresponse error
Answer: A. Measurement error. The measuring instrument systematically records values incorrectly.
A clerk mistypes survey values during entry. Which error is this?
A Sampling error
B Coverage error
C Unit nonresponse
D Processing error
Answer: D. Processing error. Data-entry mistakes are processing errors.
A frame omits newly opened businesses. Which error source is this?
A Measurement error
B Sampling error
C Coverage error
D Item nonresponse
Answer: C. Coverage error. The frame fails to cover part of the target business population.
A frame lists the same household twice. What issue does this create?
A Undercoverage
B Randomization
C Item response
D Overcoverage
Answer: D. Overcoverage. Duplicate frame entries create overcoverage and can distort selection chances.
Why contact sampled nonrespondents again?
A Change target population
B Reduce nonresponse risk
C Remove sampling error
D Replace the sample mean
Answer: B. Reduce nonresponse risk. Follow-up can increase response and help assess or reduce nonresponse problems.
With unequal stratum sizes, how is the overall mean combined?
A Population-weighted stratum means
B Unweighted stratum average
C Largest stratum only
D Smallest stratum only
Answer: A. Population-weighted stratum means. The stratified mean weights each stratum mean by its share of the population.
What is the standard error of a statistic?
A SD of the population
B Size of the sample
C SD of its sampling distribution
D Largest observed value
Answer: C. SD of its sampling distribution. Standard error is the standard deviation of a statistic’s sampling distribution.
When can cluster sampling be less precise at equal sample size?
A Every stratum is sampled
B Cluster members are similar
C Units are weighted
D Frame is complete
Answer: B. Cluster members are similar. Similar units within clusters can increase variance compared with a simple random sample.
What can a finite-population correction reflect?
A Voluntary response
B Measurement bias
C Frame duplication
D Sampling without replacement
Answer: D. Sampling without replacement. For a sizable sampling fraction without replacement, the correction lowers estimated sampling variability.
What generally happens to margin of error when n increases?
A It generally decreases
B It always increases
C It remains identical
D It becomes nonresponse
Answer: A. It generally decreases. A larger sample generally gives a smaller standard error and margin of error at the same confidence level.
A survey gets 84 replies from 120 sampled units. What is the response rate?
A 60 percent
B 70 percent
C 72 percent
D 84 percent
Answer: B. 70 percent. The response rate is 84/120 = 0.70, or 70%.
Which error can occur even when every sampled person responds?
A Unit nonresponse
B Item nonresponse
C Measurement error
D Nonresponse bias
Answer: C. Measurement error. Respondents may still misunderstand or misreport answers, creating measurement error.
Why do repeated random samples have different means?
A Population changes each time
B Parameters are recalculated
C Frames become identical
D Random sampling variation
Answer: D. Random sampling variation. Different random samples contain different observations, so their means vary.
What is an unbiased estimator in repeated sampling?
A Expected value equals parameter
B Every estimate is exact
C Variance is always zero
D Sample size equals population
Answer: A. Expected value equals parameter. An estimator is unbiased when its long-run expected value equals the target parameter.
Which problem is not reduced just by increasing sample size?
A Random sampling variation
B Standard error
C Systematic selection bias
D Sampling fluctuation
Answer: C. Systematic selection bias. A larger sample cannot eliminate a systematic bias in how units are selected.