A teacher takes many random samples of n=36 test scores from a population with μ=75 and σ=18. Sketch in words the sampling distribution of xˉ.
Example 3
medium
A factory claim says mean bolt length is μ=50 mm with σ=0.4 mm. A sample of n=64 gives xˉ=50.1 mm. Is 0.1 mm unusual?
Example 4
medium
In one random sample of n=50, the sample mean is xˉ=12. Why is it wrong to claim 'the population mean is exactly 12'?
Example 5
hard
A population has σ=15. You want the standard error of xˉ to be at most 1.5. What is the minimum n?
Example 6
hard
A sample of n=36 from a population with μ=120, σ=24 gives xˉ=132. How many standard errors above μ is xˉ?
Example 7
hard
The standard error of xˉ is 3 for n=25. What is it for n=225, assuming the same population?
Example 8
hard
Population standard deviation is σ=6. For random samples of n=9, find a typical range that should contain about 95% of sample means around μ=50.
Example 9
challenge
A factory's bolts have μ=5.00 cm, σ=0.10 cm. A quality engineer rejects a shift if xˉ from n=25 bolts differs from 5.00 by more than 0.04 cm. What is the rejection probability for an in-control shift?
Example 10
challenge
A population has σ unknown. From n=100 data values, the sample SD is s=8 and xˉ=52. Estimate the standard error of xˉ and a 95% interval for μ.
Practice Problems
Try each one, then open the solution to compare your method.
Example 1
easy
Two honest random samples of 50 voters give 48% and 52% support for a measure. Does this difference prove one sample was biased?
Example 2
easy
A population has mean μ=100. Three random samples give means 98, 101, and 103. What explains the spread of these sample means?
Example 3
easy
True or false: if a random sample's statistic does not equal the population parameter, the sample must be biased.
Example 4
easy
Polling 30 students gives a mean study time of 4.2 hours; another 30 gives 3.9 hours. What kind of difference is this most likely?
Example 5
easy
Which decreases sampling variability of the sample mean: a larger sample or a smaller sample?
Example 6
easy
A scientist remeasures the same object 5 times with a scale and gets slightly different readings. Is this sampling variability?
Example 7
easy
Fill in: sampling variability is the natural ____ difference seen when taking repeated random samples.
Example 8
easy
A claim states: 'Our one sample of 1000 gives the exact population average.' What is wrong with this?
Example 9
medium
A population of test scores has μ=70, σ=10. You take many samples of size n=25. Roughly how much should sample means typically vary around 70?
Example 10
medium
A survey of 100 people estimates 60% approval. A repeat survey of 100 gets 56%. A third gets 63%. What single concept best describes this pattern of differing results?
Example 11
medium
If you quadruple the sample size, by what factor does the typical sampling variability of the mean change?
Example 12
medium
A teacher says: 'Class A averaged 82 and Class B averaged 80 on a random quiz sample, so A is the stronger class.' Why is caution warranted?
Example 13
medium
Population proportion is p=0.5. For samples of n=100, the standard deviation of the sample proportion is p(1−p)/n. Compute it.
Example 14
medium
Why do two honest pollsters using proper random sampling sometimes report different percentages?
Example 15
medium
A sample of 36 items from a population with σ=18 gives mean 50. What is the typical sampling variability of this mean, and is a value of 53 surprising?
Example 16
medium
Among these, which is NOT a source of sampling variability: (a) which people land in the sample, (b) random selection, (c) a miscalibrated scale used on every unit?
Example 17
medium
You take 1000 random samples of size 40 and plot all 1000 sample means. The plot's spread reflects what?
Example 18
challenge
A population has σ=20. A researcher wants the typical sampling variability of the mean to be at most 2. What minimum sample size is needed?
Example 19
challenge
Sample A (n=25) and sample B (n=100) come from the same population with σ=15. By what factor is sample A's typical variability larger than sample B's?
Example 20
challenge
A pollster reports a result with no margin of error, claiming their single random sample of 500 gives the precise population value. Explain the conceptual error and what they should report.
Example 21
easy
Two random samples of size 40 from the same school give mean GPAs of 3.10 and 3.18. What concept best explains the gap?
Example 22
easy
A population proportion is p=0.5. A random sample of 100 gives p^=0.46. Is the difference 0.04 surprising?
Example 23
easy
Which sample produces less sampling variability for the sample mean: n=25 or n=400?
Example 24
easy
A radio host claims one sample of 500 adults gives the 'exact' national approval rating. What is the flaw?
Example 25
medium
A population has σ=12. To cut the standard error of the sample mean from 4 to 1, by what factor must the sample size grow?
Example 26
medium
A poll of 400 voters shows 52% support. The standard error of p^ is about 0.025. Roughly how far can p^ stray from p?
Example 27
medium
Why is sampling variability not a 'mistake' that good statisticians can avoid?
Example 28
medium
A population has μ=60, σ=10. Compare the standard error of xˉ for n=25 vs n=100.
Example 29
medium
A coin is fair (p=0.5). Out of 100 flips, you get 58 heads. Is this strong evidence the coin is biased?
Example 30
hard
Two polls each survey 1000 adults randomly. Poll A reports 42% approval, Poll B reports 46%. Is the 4-point gap likely real?
Example 31
hard
A sociologist insists 'my one sample shows kids today watch more TV than last year.' What flaw of reasoning does this overlook?
Example 32
hard
A class simulates taking 1000 random samples of n=40 and records each sample's mean. What does the spread of those 1000 means estimate?
Example 33
hard
A scientist wants to compare two diets using random samples of n=10 each, then n=200 each. Which design will show smaller differences as significant?
Example 34
challenge
A poll of n=1000 adults estimates support at 51%. The margin of error (at 95%) is about ±3.1%. Why can't the poll conclude that 'a majority supports' the measure?
What to focus on
Core idea:Every random sample catches a different set of individuals, so its statistic wobbles around the population value by a predictable amount that shrinks like n1; judge any difference between samples against that wobble.
Common stuck point:Students often know a procedure related to sampling variability but skip the recognition step: Could chance alone explain why fair random samples from the same population gave different results? That leads to a calculation or graph that looks reasonable but answers a different question.
Ask yourself:Could chance alone explain why fair random samples from the same population gave different results?
Concept recap
Sampling variability is the natural sample-to-sample difference that appears when we take repeated random samples from the same population. Even good random samples do not all produce identical statistics.
A basketball player who makes 60% of her free throws shows sampling variability every practice: in sets of 20 shots she might make 10, 13 or 14, though her skill is the same. Longer sets give steadier percentages. A real player has good and bad days; a population stays fixed, so all of its wobble comes from which individuals a sample catches.