Biostatistics Quiz: Sampling Variability And Distributions
20 questions · exam conditions
0:00
Sampling Variability And DistributionsQuestion 1 of 20

A psychologist studies reaction times that are exponentially distributed with mean 0.8 seconds. She plans to analyze the sampling distribution of sample means from samples of size 49. Which statement correctly describes this sampling distribution?

Mean = 0.8 seconds, approximately normal shape, standard error dependent on the exponential parameter
Mean = 0.8 seconds, exponential shape, standard error = 0.8/7 seconds
Mean = 0.8 seconds, approximately normal shape, standard error = 0.8/7 seconds
Mean = 5.6 seconds, approximately normal shape, standard error = 0.8/7 seconds
Mean = 0.8 seconds, approximately normal shape, standard error = 0.8/49 seconds
← Back to quizzes

Biostatistics Quiz

Biostatistics Quiz: Sampling Variability And Distributions

Practice Sampling Variability And Distributions in Biostatistics with focused quiz questions that help you check what you know, review explanations, and build confidence with test-style prompts.

What this quiz covers

This quiz focuses on Sampling Variability And Distributions, giving you a quick way to practice the rules, question types, and explanations that matter most for Biostatistics.

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.

All questions

Question 1

A psychologist studies reaction times that are exponentially distributed with mean 0.8 seconds. She plans to analyze the sampling distribution of sample means from samples of size 49. Which statement correctly describes this sampling distribution?

  1. Mean = 0.8 seconds, approximately normal shape, standard error dependent on the exponential parameter
  2. Mean = 0.8 seconds, exponential shape, standard error = 0.8/7 seconds
  3. Mean = 0.8 seconds, approximately normal shape, standard error = 0.8/7 seconds (correct answer)
  4. Mean = 5.6 seconds, approximately normal shape, standard error = 0.8/7 seconds
  5. Mean = 0.8 seconds, approximately normal shape, standard error = 0.8/49 seconds
Explanation: This question tests your understanding of the Central Limit Theorem (CLT) and how it applies to sampling distributions, regardless of the original population distribution. When working with sampling distributions of means, you need to determine three key characteristics: the mean, the shape, and the standard error. For the mean, the sampling distribution of sample means always has the same mean as the original population, so it remains 0.8 seconds. For the shape, the CLT tells us that when sample size is sufficiently large (typically n ≥ 30), the sampling distribution becomes approximately normal regardless of the original distribution's shape. Since n = 49 > 30, the exponential shape transforms into a normal distribution. For the standard error, use the formula SE=σnSE = \frac{\sigma}{\sqrt{n}}. In an exponential distribution, the standard deviation equals the mean, so SE=0.849=0.87SE = \frac{0.8}{\sqrt{49}} = \frac{0.8}{7} seconds. Answer A incorrectly suggests the standard error depends on "the exponential parameter" without calculating the specific value. Answer B correctly identifies the mean and standard error but wrongly claims the sampling distribution retains the exponential shape—this ignores the CLT's power to normalize distributions with large samples. Answer D makes a calculation error for the mean, possibly confusing it with some multiple of the original mean (0.8 × 7 = 5.6). Remember: The CLT is your friend when dealing with sampling distributions. Large sample sizes (n ≥ 30) virtually guarantee normal sampling distributions, and always use σn\frac{\sigma}{\sqrt{n}} for standard error calculations.

Question 2

A researcher compares the sampling variability of two different statistics calculated from the same sample: the sample mean and the sample median. For samples from a normal distribution, which statement is most accurate?

  1. The sample mean and sample median have identical sampling variability for all sample sizes
  2. The sample median has lower sampling variability than the sample mean for all sample sizes
  3. The sample mean has lower sampling variability than the sample median for all sample sizes (correct answer)
  4. The relative sampling variability depends entirely on the population standard deviation
  5. The sampling variability is identical only when the sample size approaches infinity
Explanation: When comparing different statistics from the same sample, you're examining their efficiency - how much sampling variability each statistic has when estimating a population parameter. For normal distributions, this comparison reveals fundamental differences between the sample mean and median. The sample mean has lower sampling variability than the sample median when sampling from normal distributions. This happens because the mean uses all the information in the dataset, while the median only uses the middle value(s). The mean's standard error is σn\frac{\sigma}{\sqrt{n}}, while the median's standard error is approximately 1.25×σn1.25 \times \frac{\sigma}{\sqrt{n}} for normal distributions. This makes the mean about 25% more efficient than the median. Option A is incorrect because the sampling variabilities are not identical - the mean consistently has lower variability. Option B reverses the relationship; the median actually has higher sampling variability than the mean for normal distributions. Option D misses the point entirely - while the population standard deviation affects the absolute magnitude of both statistics' sampling variability, it doesn't determine their relative efficiency. The relationship between mean and median variability is determined by the population's distribution shape, not its spread. Study tip: Remember that for symmetric distributions (especially normal), the sample mean is the most efficient estimator of central tendency. However, this advantage disappears with heavily skewed or outlier-prone distributions, where the median's robustness becomes more valuable than the mean's efficiency. Focus on matching the estimator to the distribution type.

Question 3

Two researchers independently collect samples from the same population. Researcher A uses samples of size 16, while Researcher B uses samples of size 64. If both sampling distributions of the sample mean are approximately normal, how do their standard errors compare?

  1. Researcher B's standard error is exactly half of Researcher A's standard error (correct answer)
  2. Researcher B's standard error is exactly one-fourth of Researcher A's standard error
  3. Researcher B's standard error is exactly twice Researcher A's standard error
  4. Researcher B's standard error is exactly four times Researcher A's standard error
  5. The standard errors are equal since they sample from the same population
Explanation: When you encounter questions about sampling distributions and sample size, focus on the standard error formula and how sample size affects variability. The standard error of the sample mean equals the population standard deviation divided by the square root of the sample size: SE=σnSE = \frac{\sigma}{\sqrt{n}}. Since both researchers sample from the same population, they have identical population standard deviations (σ). The only difference is their sample sizes. For Researcher A with n = 16: SEA=σ16=σ4SE_A = \frac{\sigma}{\sqrt{16}} = \frac{\sigma}{4} For Researcher B with n = 64: SEB=σ64=σ8SE_B = \frac{\sigma}{\sqrt{64}} = \frac{\sigma}{8} To compare them: SEBSEA=σ/8σ/4=1/81/4=12\frac{SE_B}{SE_A} = \frac{\sigma/8}{\sigma/4} = \frac{1/8}{1/4} = \frac{1}{2} Therefore, Researcher B's standard error is exactly half of Researcher A's standard error, making A correct. B is wrong because it suggests the ratio is 1/4, which would require a sample size ratio of 16:1, not 4:1. C is incorrect because it claims B's standard error is larger, but larger samples always produce smaller standard errors. D makes the same directional error as C, incorrectly suggesting the researcher with the larger sample has a larger standard error. Study tip: Remember that standard error decreases with the square root of sample size. When sample size quadruples (16 to 64), standard error halves. This inverse square root relationship appears frequently in sampling distribution problems.

Question 4

A population has a right-skewed distribution with mean 40 and standard deviation 12. Which statement best describes what happens to the sampling distribution of the sample mean as the sample size increases from 4 to 100?

  1. The sampling distribution becomes more right-skewed and its standard deviation decreases from 6 to 1.2
  2. The sampling distribution becomes more normal and its standard deviation decreases from 6 to 1.2 (correct answer)
  3. The sampling distribution becomes more right-skewed and its standard deviation increases from 6 to 1.2
  4. The sampling distribution becomes more normal and its standard deviation increases from 6 to 1.2
  5. The sampling distribution maintains the same shape and its standard deviation decreases from 6 to 1.2
Explanation: This question tests your understanding of the Central Limit Theorem (CLT), which describes how sampling distributions behave as sample size increases. When you encounter questions about sampling distributions and changing sample sizes, focus on two key effects: the shape becomes more normal, and the spread decreases. The Central Limit Theorem tells us that regardless of the original population's shape, the sampling distribution of sample means approaches a normal distribution as sample size increases. Since we're going from n=4 to n=100, the sampling distribution will become much more normal despite the population being right-skewed. The standard error (standard deviation of the sampling distribution) equals σn\frac{\sigma}{\sqrt{n}}, where σ is the population standard deviation. For n=4: 124=122=6\frac{12}{\sqrt{4}} = \frac{12}{2} = 6. For n=100: 12100=1210=1.2\frac{12}{\sqrt{100}} = \frac{12}{10} = 1.2. So the standard deviation decreases from 6 to 1.2. Answer A incorrectly states the distribution becomes more right-skewed. The CLT guarantees it becomes more normal, not more skewed. Answer C makes the same error about skewness and also incorrectly claims the standard deviation increases rather than decreases. Answer D correctly identifies the normality trend and the standard deviation values, but wrongly states the standard deviation increases. Remember: larger sample sizes always lead to more normal sampling distributions (CLT) and smaller standard errors. The population's skewness becomes less influential as n increases, while the standard error formula σn\frac{\sigma}{\sqrt{n}} ensures decreasing variability.

Question 5

An epidemiologist studies the sampling distribution of sample proportions from a population where 30% of individuals have a certain condition. If samples of size 100 are repeatedly drawn, which statement correctly describes the expected characteristics of this sampling distribution?

  1. Mean = 0.30, Standard deviation = 0.21\sqrt{0.21}, approximately normal shape
  2. Mean = 0.30, Standard deviation = 0.046, approximately normal shape (correct answer)
  3. Mean = 0.70, Standard deviation = 0.046, approximately normal shape
  4. Mean = 0.30, Standard deviation = 0.21, approximately normal shape
  5. Mean = 0.30, Standard deviation = 0.046, right-skewed shape
Explanation: When you encounter questions about sampling distributions of proportions, you're dealing with one of the fundamental concepts in biostatistics that describes how sample statistics behave across repeated sampling. For a sampling distribution of sample proportions, three key characteristics define it: the mean equals the population proportion (p), the standard deviation equals p(1−p)n\sqrt{\frac{p(1-p)}{n}}, and the shape is approximately normal when sample size conditions are met. With p = 0.30 and n = 100, the mean of the sampling distribution equals 0.30. The standard deviation is 0.30×0.70100=0.21100=0.0021=0.046\sqrt{\frac{0.30 \times 0.70}{100}} = \sqrt{\frac{0.21}{100}} = \sqrt{0.0021} = 0.046. Since np = 30 and n(1-p) = 70 are both greater than 10, the Central Limit Theorem ensures the distribution is approximately normal. This confirms answer B is correct. Answer A incorrectly calculates the standard deviation as 0.21\sqrt{0.21} ≈ 0.458, which represents p(1−p)\sqrt{p(1-p)} but omits dividing by n. Answer C has the correct standard deviation but uses the wrong mean (0.70 instead of 0.30) – a common error of confusing p with (1-p). Answer D correctly identifies the mean but uses 0.21 as the standard deviation, which is actually the variance p(1-p), not the standard error. Remember the sampling distribution formula: mean = p, standard deviation = p(1−p)n\sqrt{\frac{p(1-p)}{n}}. The most frequent mistakes involve forgetting to divide by n or taking the square root incorrectly.

Question 6

In a large university, 40% of students live on campus. A researcher plans to take a random sample of students to estimate this proportion. For which sample size would the sampling distribution of the sample proportion definitely NOT be approximately normal?

  1. n=20n = 20 (correct answer)
  2. n=30n = 30
  3. n=50n = 50
  4. n=75n = 75
  5. n=100n = 100
Explanation: When you encounter questions about the sampling distribution of a sample proportion, you need to check whether the normal approximation conditions are met. For a sample proportion to have an approximately normal distribution, both np≥10np \geq 10 and n(1−p)≥10n(1-p) \geq 10 must be satisfied, where nn is the sample size and pp is the population proportion. Given that 40% of students live on campus, we have p=0.4p = 0.4 and (1−p)=0.6(1-p) = 0.6. Let's check each sample size: For option A (n=20n = 20): np=20×0.4=8np = 20 \times 0.4 = 8 and n(1−p)=20×0.6=12n(1-p) = 20 \times 0.6 = 12. Since np=8<10np = 8 < 10, the normal approximation condition fails. This sampling distribution would definitely NOT be approximately normal. For option B (n=30n = 30): np=30×0.4=12np = 30 \times 0.4 = 12 and n(1−p)=30×0.6=18n(1-p) = 30 \times 0.6 = 18. Both values exceed 10, so the normal approximation would apply. For option C (n=50n = 50): np=50×0.4=20np = 50 \times 0.4 = 20 and n(1−p)=50×0.6=30n(1-p) = 50 \times 0.6 = 30. Both conditions are satisfied. For option D (n=75n = 75): np=75×0.4=30np = 75 \times 0.4 = 30 and n(1−p)=75×0.6=45n(1-p) = 75 \times 0.6 = 45. Both conditions are well satisfied. Only option A fails to meet the normal approximation requirements, making it the correct answer. Study tip: Always check both np≥10np \geq 10 AND n(1−p)≥10n(1-p) \geq 10 for normal approximation questions. If either condition fails, the distribution won't be approximately normal, regardless of how large the other value might be.

Question 7

Two populations have the same mean but different standard deviations: Population 1 has σ1=12\sigma_1 = 12 and Population 2 has σ2=18\sigma_2 = 18. If equal-sized samples of n=36n = 36 are drawn from each population, what is the ratio of the standard error from Population 2 to the standard error from Population 1?

  1. 0.67
  2. 1.00
  3. 1.50 (correct answer)
  4. 2.25
  5. 3.00
Explanation: When you encounter questions about standard errors from different populations, remember that standard error measures how much sample means vary around the true population mean. The key relationship is that standard error equals population standard deviation divided by the square root of sample size. For Population 1: SE1=σ1n=1236=126=2SE_1 = \frac{\sigma_1}{\sqrt{n}} = \frac{12}{\sqrt{36}} = \frac{12}{6} = 2 For Population 2: SE2=σ2n=1836=186=3SE_2 = \frac{\sigma_2}{\sqrt{n}} = \frac{18}{\sqrt{36}} = \frac{18}{6} = 3 The ratio of standard errors is: SE2SE1=32=1.50\frac{SE_2}{SE_1} = \frac{3}{2} = 1.50 Looking at the wrong answers: Choice A (0.67) represents the inverse ratio - you'd get this if you calculated SE1SE2\frac{SE_1}{SE_2} instead of SE2SE1\frac{SE_2}{SE_1}. Choice B (1.00) would only be correct if both populations had identical standard deviations, since the sample sizes are equal. Choice D (2.25) might tempt you if you mistakenly squared the ratio of the standard deviations: (1812)2=1.52=2.25\left(\frac{18}{12}\right)^2 = 1.5^2 = 2.25, but standard error has a linear relationship with population standard deviation, not quadratic. The correct answer is C (1.50). Remember: standard error is directly proportional to population standard deviation when sample sizes are equal. If one population's standard deviation is 1.5 times larger, its standard error will also be 1.5 times larger. Always double-check which ratio the question asks for to avoid the inverse trap.

Question 8

A medical researcher collects blood pressure measurements from a population where systolic pressure is normally distributed with mean 120 mmHg and standard deviation 16 mmHg. She repeatedly takes samples of size 64 and calculates the sample mean each time. Approximately what percentage of these sample means will fall between 118 and 122 mmHg?

  1. 38.3%
  2. 68.3% (correct answer)
  3. 76.6%
  4. 95.4%
  5. 99.7%
Explanation: When you encounter a problem about sample means from a normally distributed population, you're dealing with the sampling distribution of the mean. The key insight is that sample means have less variability than individual observations due to the Central Limit Theorem. Here, individual blood pressure measurements follow a normal distribution with mean μ = 120 mmHg and standard deviation σ = 16 mmHg. However, when you take samples of size n = 64 and calculate their means, these sample means follow a normal distribution with the same mean (120 mmHg) but a smaller standard deviation called the standard error: SE=σn=1664=168=2 mmHgSE = \frac{\sigma}{\sqrt{n}} = \frac{16}{\sqrt{64}} = \frac{16}{8} = 2 \text{ mmHg} To find what percentage of sample means fall between 118 and 122 mmHg, you need to standardize these values:
  • Lower bound: z=118−1202=−1z = \frac{118-120}{2} = -1
  • Upper bound: z=122−1202=1z = \frac{122-120}{2} = 1
The area between z = -1 and z = 1 in a standard normal distribution is approximately 68.3%. Answer A (38.3%) represents roughly half of this interval. Answer C (76.6%) might result from incorrectly using z-scores of ±1.2. Answer D (95.4%) corresponds to the area between z = ±2, which would be correct if asking about the range 116-124 mmHg. Remember: always calculate the standard error when working with sample means, not the original population standard deviation. The "68-95-99.7 rule" becomes your friend once you've properly standardized the problem.

Question 9

An agricultural scientist measures crop yields that follow a normal distribution with mean 45 bushels per acre and standard deviation 6 bushels per acre. She randomly selects 9 plots and calculates the sample mean yield. What is the probability that this sample mean is more than 2 bushels above the population mean?

  1. 0.1587 (correct answer)
  2. 0.2266
  3. 0.3085
  4. 0.3694
  5. 0.6915
Explanation: When you encounter problems about sample means from normally distributed populations, you're working with the sampling distribution of the mean. This distribution has the same mean as the population but a smaller standard deviation called the standard error. The key insight is that while individual crop yields have standard deviation 6, the sample mean of 9 plots has standard error σxˉ=σn=69=2\sigma_{\bar{x}} = \frac{\sigma}{\sqrt{n}} = \frac{6}{\sqrt{9}} = 2. You want the probability that the sample mean exceeds the population mean by more than 2 bushels, so P(Xˉ>47)P(\bar{X} > 47). To find this probability, standardize using the z-score: z=47−452=1z = \frac{47 - 45}{2} = 1. From the standard normal table, P(Z>1)=0.1587P(Z > 1) = 0.1587. Looking at the wrong answers: Choice B (0.2266) corresponds to using the wrong critical value, possibly confusing this with a one-tailed probability at a different z-score. Choice C (0.3085) might result from incorrectly using the population standard deviation (6) instead of the standard error (2), giving z=26=0.33z = \frac{2}{6} = 0.33. Choice D (0.3694) could come from using an incorrect formula or misreading probability tables. The correct answer is A) 0.1587. Remember: whenever you're asked about sample means, always calculate the standard error first by dividing the population standard deviation by the square root of sample size. This is one of the most common errors in sampling distribution problems.

Question 10

A statistician examines two sampling scenarios: Scenario A uses simple random sampling, while Scenario B uses stratified random sampling with the same total sample size. Both scenarios target the same population parameter. Which statement about sampling variability is most accurate?

  1. Scenario A will always have lower sampling variability due to the simplicity of the design
  2. Scenario B will typically have lower sampling variability when strata are internally homogeneous (correct answer)
  3. Both scenarios will have identical sampling variability since the total sample size is the same
  4. The sampling variability depends only on the population variance, not the sampling method
  5. Scenario B will have higher sampling variability due to the complexity of the stratified design
Explanation: When you encounter questions comparing sampling methods, focus on how different designs affect the precision of your estimates. The key insight is that stratified sampling can reduce sampling variability when done strategically. Stratified sampling divides the population into homogeneous subgroups (strata) before sampling from each stratum. When strata are internally similar but different from each other, this design ensures your sample represents all important population segments. This representation typically reduces sampling variability compared to simple random sampling, which might by chance over-sample or under-sample certain population segments. Answer B correctly identifies that stratified sampling will typically have lower sampling variability when strata are internally homogeneous. The reduced variability occurs because you're guaranteeing representation from each important population subgroup, eliminating the random fluctuations that could occur with simple random sampling. Answer A incorrectly assumes design simplicity relates to lower variability. Simple random sampling is easier to implement, but this doesn't make it more precise. Answer C makes the common mistake of thinking identical sample sizes guarantee identical variability. While sample size affects precision, the sampling method's efficiency also matters significantly. Answer D oversimplifies by ignoring sampling design effects. While population variance does influence sampling variability, the sampling method's ability to capture that variance efficiently is equally important. Remember this pattern: stratified sampling reduces variability when strata are homogeneous within but heterogeneous between groups. Always consider both sample size AND sampling efficiency when comparing designs.

Question 11

A researcher compares two sampling strategies for estimating a population mean. Strategy X uses 25 samples of size 16 each. Strategy Y uses 16 samples of size 25 each. Both strategies result in the same total number of observations. Which statement about the sampling variability is correct?

  1. Strategy X will have lower sampling variability because it uses more samples
  2. Strategy Y will have lower sampling variability because it uses larger sample sizes (correct answer)
  3. Both strategies will have identical sampling variability since the total sample size is the same
  4. Strategy X will have lower sampling variability because smaller samples reduce bias
  5. The sampling variability cannot be compared without knowing the population variance
Explanation: When comparing sampling strategies, you need to focus on how sample size affects the precision of your estimates, not just the total number of observations collected. The key principle here is that sampling variability decreases as individual sample size increases. The standard error of the mean is calculated as σ/n\sigma/\sqrt{n}, where n is the size of each individual sample. Strategy Y uses samples of size 25, while Strategy X uses samples of size 16. Since 25=5\sqrt{25} = 5 is greater than 16=4\sqrt{16} = 4, Strategy Y will produce estimates with smaller standard errors and thus lower sampling variability. Looking at the wrong answers: Choice A incorrectly assumes that having more samples automatically reduces variability, but what matters is the size of each sample, not how many you take. Choice C falls into the trap of thinking total sample size determines sampling variability—this is incorrect because it's the individual sample sizes that affect the precision of each estimate. Choice D misunderstands both the relationship between sample size and variability (smaller samples actually increase variability) and confuses bias with variability (bias is about accuracy, while this question asks about precision). Remember this pattern: when comparing sampling strategies, larger individual sample sizes always win over more numerous smaller samples when it comes to reducing sampling variability. The square root in the standard error formula means you get diminishing returns, but the principle holds—focus on individual sample size, not total observations collected.

Question 12

A public health official knows that 25% of adults in a region have high blood pressure. She plans to survey random samples to estimate this proportion. For a sample of size 80, what is the approximate standard deviation of the sampling distribution of the sample proportion?

  1. 0.0484 (correct answer)
  2. 0.0543
  3. 0.1875
  4. 0.2500
  5. 0.4330
Explanation: When you encounter questions about sampling distributions of proportions, you're dealing with one of the fundamental concepts in statistical inference. The key insight is that when you repeatedly sample from a population, the sample proportions will form their own distribution with predictable characteristics. To find the standard deviation of the sampling distribution of a sample proportion, you use the formula: σp^=p(1−p)n\sigma_{\hat{p}} = \sqrt{\frac{p(1-p)}{n}}, where p is the true population proportion and n is the sample size. Here, the population proportion p = 0.25 and sample size n = 80. Substituting: σp^=0.25×0.7580=0.187580=0.002344=0.0484\sigma_{\hat{p}} = \sqrt{\frac{0.25 \times 0.75}{80}} = \sqrt{\frac{0.1875}{80}} = \sqrt{0.002344} = 0.0484 This makes A) 0.0484 correct. Looking at the wrong answers: B) 0.0543 likely results from a calculation error, possibly using an incorrect denominator. C) 0.1875 is actually the numerator p(1-p) before dividing by n and taking the square root – this represents forgetting the sample size adjustment entirely. D) 0.2500 is simply the population proportion p itself, showing a fundamental misunderstanding of what we're calculating. Remember this formula pattern: the standard error of a proportion always involves p(1-p) in the numerator and the square root of sample size in the denominator. The larger your sample, the smaller your standard error becomes, reflecting increased precision in your estimate.

Question 13

A clinical trial investigator wants to determine the minimum sample size needed so that the standard error of the sample mean is no more than 2.5 units. If the population standard deviation is known to be 15 units, what is the minimum required sample size?

  1. 6
  2. 25
  3. 36 (correct answer)
  4. 100
  5. 225
Explanation: When you encounter sample size questions involving standard error, you're working with the fundamental relationship between sample size and precision of estimates. The standard error of the sample mean tells us how much variability we expect in our sample means if we repeated the study multiple times. The formula for standard error of the sample mean is: SE=σnSE = \frac{\sigma}{\sqrt{n}}, where σ is the population standard deviation and n is the sample size. Since we want the standard error to be no more than 2.5 units, we set up the inequality: 15n≤2.5\frac{15}{\sqrt{n}} \leq 2.5 Solving for n: n≥152.5=6\sqrt{n} \geq \frac{15}{2.5} = 6, so n≥36n \geq 36 Therefore, the minimum sample size is 36, making C correct. Looking at the wrong answers: A) 6 represents the value of n\sqrt{n}, not n itself - this is a common algebraic mistake where students stop one step too early. B) 25 would give a standard error of 1525=3\frac{15}{\sqrt{25}} = 3, which exceeds our maximum allowable standard error of 2.5. D) 100 would work (giving SE = 1.5) but it's unnecessarily large and not the minimum required. Remember that standard error decreases as sample size increases, but the relationship involves a square root, so you need to square both sides when solving these inequalities. Always double-check your final answer by plugging it back into the standard error formula to confirm it meets the requirement.

Question 14

A quality control manager knows that the weights of manufactured parts follow a normal distribution with μ=200\mu = 200 grams and σ=8\sigma = 8 grams. If she randomly samples 16 parts and calculates the sample mean weight, what is the probability that this sample mean exceeds 203 grams?

  1. 0.0668 (correct answer)
  2. 0.1587
  3. 0.3536
  4. 0.4332
  5. 0.8413
Explanation: This question tests your understanding of sampling distributions and the Central Limit Theorem. When you're asked about the probability of a sample mean (rather than individual values), you need to recognize that sample means have their own distribution with different characteristics than the original population. The population has μ=200\mu = 200 and σ=8\sigma = 8, but when you take samples of size n=16n = 16, the sample means follow a normal distribution with the same mean (μxˉ=200\mu_{\bar{x}} = 200) but a smaller standard deviation called the standard error: σxˉ=σn=816=84=2\sigma_{\bar{x}} = \frac{\sigma}{\sqrt{n}} = \frac{8}{\sqrt{16}} = \frac{8}{4} = 2. To find P(xˉ>203)P(\bar{x} > 203), you standardize: z=203−2002=1.5z = \frac{203 - 200}{2} = 1.5. Using the standard normal table, P(Z>1.5)=1−0.9332=0.0668P(Z > 1.5) = 1 - 0.9332 = 0.0668. Looking at the wrong answers: B) 0.1587 is what you'd get if you mistakenly used the original population standard deviation (8) instead of the standard error, giving z=203−2008=0.375z = \frac{203-200}{8} = 0.375, which rounds to the probability for z=1.0z = 1.0. C) 0.3536 and D) 0.4332 likely come from calculation errors or misreading the z-table. Remember this key distinction: individual values use the population standard deviation, but sample means use the standard error (σn\frac{\sigma}{\sqrt{n}}). The correct answer is A) 0.0668.

Question 15

An environmental scientist measures pollutant concentrations that follow a distribution with mean 25 ppm and standard deviation 8 ppm. She wants the standard error of her sample mean to be exactly 1.6 ppm. If she increases her planned sample size from 25 to 100, how will this affect the standard error?

  1. The standard error will decrease from 1.6 ppm to 0.8 ppm, meeting her goal (correct answer)
  2. The standard error will decrease from 1.6 ppm to 0.4 ppm, exceeding her goal
  3. The standard error will decrease from 1.6 ppm to 1.6 ppm, maintaining her goal
  4. The standard error will increase from 1.6 ppm to 3.2 ppm, moving away from her goal
  5. The standard error will remain unchanged because it only depends on the population standard deviation
Explanation: When you encounter questions about sample size and standard error, you're dealing with one of the fundamental relationships in sampling theory: the standard error of the mean decreases as sample size increases according to the formula SE=σnSE = \frac{\sigma}{\sqrt{n}}, where σ is the population standard deviation and n is sample size. Let's calculate the standard errors for both sample sizes. With the original sample size of 25: SE=825=85=1.6SE = \frac{8}{\sqrt{25}} = \frac{8}{5} = 1.6 ppm. With the increased sample size of 100: SE=8100=810=0.8SE = \frac{8}{\sqrt{100}} = \frac{8}{10} = 0.8 ppm. The scientist wanted exactly 1.6 ppm, which she achieves with n=25. When she increases to n=100, the standard error drops to 0.8 ppm, which is actually better precision than her goal. Answer A correctly identifies this decrease from 1.6 to 0.8 ppm and notes she meets her goal (since lower standard error means better precision). Answer B miscalculates the final standard error as 0.4 ppm, which would require n=400, not 100. Answer C impossibly suggests the standard error stays the same despite quadrupling the sample size. Answer D defies the fundamental principle by suggesting standard error increases with larger sample size. Remember this key relationship: standard error is inversely proportional to the square root of sample size. Quadrupling sample size (25 to 100) halves the standard error. This appears frequently on biostatistics exams, so practice recognizing when sample size changes affect precision of estimates.

Question 16

A sociologist studies income data from a highly skewed population. She plans to construct confidence intervals for the population mean using sample means from samples of different sizes. For which sample size would the Central Limit Theorem provide the LEAST reliable justification for using normal-based confidence intervals?

  1. n=5n = 5 (correct answer)
  2. n=15n = 15
  3. n=30n = 30
  4. n=50n = 50
  5. n=100n = 100
Explanation: When you encounter questions about the Central Limit Theorem (CLT) and confidence intervals, focus on how sample size affects the reliability of normal approximations, especially with skewed populations. The CLT states that sample means approach a normal distribution as sample size increases, regardless of the population's shape. However, with highly skewed populations, you need larger sample sizes for this approximation to become reliable enough for valid confidence intervals. Answer A (n=5n = 5) is correct because it provides the least reliable justification. With only 5 observations from a highly skewed population, the sampling distribution of the sample mean will still closely resemble the original skewed shape. The CLT hasn't had sufficient opportunity to "kick in" and normalize the distribution, making normal-based confidence intervals inappropriate and potentially misleading. Answer B (n=15n = 15) is better than n=5n = 5 but still marginal for highly skewed data. Some statisticians consider this a borderline case requiring careful evaluation of the skewness severity. Answer C (n=30n = 30) represents the traditional rule-of-thumb minimum for invoking the CLT with moderately skewed data. While not perfect for highly skewed populations, it's substantially more reliable than smaller samples. Answer D (n=50n = 50) provides even stronger justification, as larger samples make the normal approximation increasingly robust against departures from normality in the original population. Study tip: Remember that greater population skewness requires larger sample sizes for reliable CLT application. When in doubt, "smaller sample size = less reliable" for questions about CLT effectiveness with non-normal populations.

Question 17

A manufacturing process produces items with weights that have mean 500g and standard deviation 20g. A quality inspector samples 16 items and finds a sample mean of 495g. Assuming the sampling distribution is approximately normal, what is the z-score for this sample mean?

  1. -5.00
  2. -1.00 (correct answer)
  3. -0.25
  4. 0.25
  5. 1.00
Explanation: When you encounter questions about sample means and z-scores, you're working with the sampling distribution of the mean, which requires using the standard error rather than the population standard deviation. To find the z-score for a sample mean, you need the formula: z=xˉ−μσxˉz = \frac{\bar{x} - \mu}{\sigma_{\bar{x}}} where σxˉ=σn\sigma_{\bar{x}} = \frac{\sigma}{\sqrt{n}} is the standard error. First, calculate the standard error: σxˉ=2016=204=5\sigma_{\bar{x}} = \frac{20}{\sqrt{16}} = \frac{20}{4} = 5 Then find the z-score: z=495−5005=−55=−1.00z = \frac{495 - 500}{5} = \frac{-5}{5} = -1.00 Choice A (-5.00) represents a common error where students use the population standard deviation (20g) instead of the standard error in the denominator: 495−50020=−0.25\frac{495-500}{20} = -0.25. Wait, that's actually choice C. Let me recalculate: Choice A (-5.00) occurs if you incorrectly use 495−5001=−5\frac{495-500}{1} = -5, perhaps confusing the calculation entirely. Choice C (-0.25) is the result of using the population standard deviation (20g) instead of the standard error: 495−50020=−0.25\frac{495-500}{20} = -0.25. Choice D (0.25) makes the same standard deviation error as C but also incorrectly treats the difference as positive. The correct answer is B (-1.00). Key strategy: Always remember that when dealing with sample means, divide the population standard deviation by n\sqrt{n} to get the standard error before calculating the z-score. This is the most common mistake on sampling distribution problems.

Question 18

A health survey finds that 60% of adults in a city exercise regularly. A researcher takes a random sample of 400 adults from this population. What is the approximate probability that the sample proportion of adults who exercise regularly is between 0.57 and 0.63?

  1. 0.4332
  2. 0.7745 (correct answer)
  3. 0.8664
  4. 0.9544
  5. 0.9973
Explanation: When you encounter questions about sample proportions from a known population, you're working with the sampling distribution of proportions, which follows a normal distribution for large samples. Here's how to solve this step-by-step. First, identify your parameters: population proportion p=0.60p = 0.60, sample size n=400n = 400, and you want P(0.57<p^<0.63)P(0.57 < \hat{p} < 0.63). The sampling distribution has mean μp^=p=0.60\mu_{\hat{p}} = p = 0.60 and standard error σp^=p(1−p)n=0.60×0.40400=0.0245\sigma_{\hat{p}} = \sqrt{\frac{p(1-p)}{n}} = \sqrt{\frac{0.60 \times 0.40}{400}} = 0.0245. Convert to z-scores: For 0.57: z=0.57−0.600.0245=−1.22z = \frac{0.57 - 0.60}{0.0245} = -1.22. For 0.63: z=0.63−0.600.0245=1.22z = \frac{0.63 - 0.60}{0.0245} = 1.22. Using the standard normal table, P(−1.22<Z<1.22)=P(Z<1.22)−P(Z<−1.22)=0.8888−0.1112=0.7776P(-1.22 < Z < 1.22) = P(Z < 1.22) - P(Z < -1.22) = 0.8888 - 0.1112 = 0.7776, which rounds to 0.7745. Answer A (0.4332) represents P(−0.67<Z<0.67)P(-0.67 < Z < 0.67), suggesting an incorrect standard error calculation. Answer C (0.8664) corresponds to P(−1.5<Z<1.5)P(-1.5 < Z < 1.5), indicating the boundaries were miscalculated. Answer D (0.9544) equals P(−2<Z<2)P(-2 < Z < 2), which would result from using 0.58 and 0.62 as boundaries instead. Study tip: Always verify your standard error calculation first—it's the most common source of error in sampling distribution problems. Remember the formula p(1−p)n\sqrt{\frac{p(1-p)}{n}} and double-check your z-score conversions.

Question 19

A medical researcher studies reaction times to a stimulus in milliseconds. The population distribution has an unknown shape with mean 250 ms and standard deviation 40 ms. The researcher is particularly interested in how sample means behave when drawn from this population.

If the researcher takes samples of size 25 and finds that the probability of obtaining a sample mean greater than 270 ms is approximately 0.006, what does this suggest about the sampling distribution, and what would be the approximate probability for samples of size 100?

  1. The sampling distribution appears normal as expected, and for n=100n = 100, the probability would be approximately 0.000003
  2. The sampling distribution shows expected behavior under the Central Limit Theorem, and for n=100n = 100, the probability would be essentially zero (correct answer)
  3. The result suggests the population distribution significantly affects the sampling distribution shape, and for n=100n = 100, the probability would be approximately 0.001
  4. The unusually low probability indicates potential issues with normality assumptions, and for n=100n = 100, the probability would be approximately 0.003
Explanation: When you encounter problems about sampling distributions, remember that the Central Limit Theorem (CLT) is your key tool for predicting how sample means behave, regardless of the original population's shape. Let's verify what we'd expect theoretically. For samples of size 25, the sampling distribution has mean μxˉ=250\mu_{\bar{x}} = 250 ms and standard error σxˉ=4025=8\sigma_{\bar{x}} = \frac{40}{\sqrt{25}} = 8 ms. The z-score for 270 ms is z=270−2508=2.5z = \frac{270-250}{8} = 2.5, giving P(Z>2.5)≈0.006P(Z > 2.5) \approx 0.006. This matches the observed probability perfectly, confirming the sampling distribution is behaving normally as the CLT predicts. For samples of size 100, the standard error becomes σxˉ=40100=4\sigma_{\bar{x}} = \frac{40}{\sqrt{100}} = 4 ms. The z-score for 270 ms is now z=270−2504=5z = \frac{270-250}{4} = 5, making P(Z>5)P(Z > 5) essentially zero (approximately 10−710^{-7}). This confirms answer B is correct. Answer A incorrectly suggests the probability would be 0.000003, which doesn't match the calculated value. Answer C wrongly implies the population distribution is interfering with normal behavior, but our calculations show the CLT is working perfectly. Answer D misinterprets the low probability as problematic, when it's actually exactly what theory predicts. Study tip: When sample sizes are reasonably large (typically n≥30n \geq 30, sometimes smaller), always assume the CLT applies and sampling distributions are normal. Calculate expected probabilities using z-scores to verify whether observed results match theory.

Question 20

A researcher collects samples of size n=25n = 25 from a population with mean μ=80\mu = 80 and standard deviation σ=15\sigma = 15. If the sampling distribution of sample means has a standard error of 3, but the researcher mistakenly believes the population standard deviation is 12, what would be the actual probability that a sample mean exceeds 83, given that the researcher calculates this probability as 0.159?

  1. The actual probability is 0.159 because the standard error calculation is independent of the population standard deviation assumption
  2. The actual probability is 0.239 because the researcher underestimated the population variability, leading to an underestimated probability
  3. The actual probability is 0.159 because the given standard error of 3 reflects the true population parameters regardless of the researcher's belief (correct answer)
  4. The actual probability is 0.097 because the researcher's miscalculation of population standard deviation affects the sampling distribution parameters
Explanation: The key insight is that the problem states the sampling distribution has a standard error of 3, which is the actual standard error. This means SE=σ/n=15/25=3SE = \sigma/\sqrt{n} = 15/\sqrt{25} = 3, confirming the true parameters. Since P(Xˉ>83)=P(Z>(83−80)/3)=P(Z>1)=0.159P(\bar{X} > 83) = P(Z > (83-80)/3) = P(Z > 1) = 0.159, the actual probability is 0.159. The researcher's mistaken belief about σ\sigma doesn't change the actual sampling distribution. Choice A is wrong because standard error does depend on σ\sigma. Choice B incorrectly assumes the researcher's error affects reality. Choice D incorrectly suggests the researcher's belief changes the actual probability.