Sampling Distributions

̂
Sampling Distributions: Sample Proportions, 𝒑 ̅
Sampling Distributions: Sample Means, 𝑿
• The mean of the sampling distribution of 𝑝̂ is 𝜇𝑝̂ = 𝑝 • The mean of the sampling distribution of 𝑋̅ is 𝜇𝑥̅ = 𝜇
• The standard deviation of the sampling distribution of 𝑝̂ is: • The standard deviation of the sampling distribution of 𝑝̂ is:
𝜎
𝑝(1 − 𝑝) 𝜎𝑥̅ =
𝜎𝑝̂ = √ √𝑛
𝑛
as long as the 10% condition is satisfied: 𝑛 ≤ 0.10𝑁. as long as the 10% condition is satisfied: 𝑛 ≤ 0.10𝑁.

• The sampling distribution of 𝑝̂ is approximately Normal as long as • The sampling distribution of 𝑋̅ is approximately Normal as long as
the Large Counts Condition is satisfied: either:
𝑛𝑝 ≥ 10 𝑎𝑛𝑑 𝑛(1 − 𝑝) ≥ 10 o Population is Normal or approximately Normal
o The central limit theorem applies (𝑛 ≥ 30)

Sample Problem Sample Problem

A mail order company advertises that it ships 90% of its orders within A company’s cereal boxes advertise that each box contains 9.65 ounces of
three working days. You select an SRS of 100 of the 5000 orders received cereal. In fact, the amount of cereal in a randomly selected box follows a
in the past week for an audit. The audit reveals that 86 of these orders Normal distribution wit mean 𝜇 = 9.70 ounces and standard deviation
were shipped on time. If the company really ships 90% of its orders on 𝜎 = 0.03 ounces.
time, what is the probability that the proportion in an SRS of 100 orders is
0.86 or less? a) What is the probability that a randomly selected box of cereal contains
less than 9.65 ounces?
𝑝 = 𝑡𝑟𝑢𝑒 𝑝𝑟𝑜𝑝𝑜𝑟𝑡𝑖𝑜𝑛 𝑜𝑓 𝑜𝑟𝑑𝑒𝑟𝑠 𝑠ℎ𝑖𝑝𝑝𝑒𝑑 𝑜𝑛𝑡𝑖𝑚𝑒 𝑋 = 𝑤𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑐𝑒𝑟𝑎𝑙 𝑏𝑜𝑥
9.65 − 9.70
𝑃(𝑋 < 9.65) = 𝑃 (𝑧 < ) = 0.0478
𝜇𝑝̂ = 𝑝 = 0.90; Since 100 is less than 10% of the 5000 orders received: 0.03
0.9(0.1)
𝜎𝑝̂ = √ 100
= 0.03. Because 𝑛𝑝 = 100(. 9) = 90 ≥ 10 𝑎𝑛𝑑 𝑛(1 − b) Now take an SRS of 5 boxes. What is the probability that the mean
𝑝) = 100(. 1) = 10 ≥ 10, sampling distribution of 𝑝̂ is approximately amount of cereal in these boxes is less than 9.65?
normal.
0.86 − 0.9 𝜇 = true mean weight of cereal box.
𝑃(𝑝̂ ≤ 0.86) = 𝑃 (𝑧 ≤ ) = 0.0912 Since the population is Normal, the sampling distribution of 𝑥̅ is
0.03
approximately normal with 𝜇𝑥̅ = 𝜇 = 9.70 ounces. Because the sample
0.03
size 5 is less than 10% of all boxes of cereal, 𝜎𝑥̅ = = 0.0134.
√5
9.65 − 9.70
𝑃(𝑋̅ < 9.65) = 𝑃 (𝑧 < ) = 0.000095
0.0134
 Sampling Distributions: Difference of Two Proportions, 𝒑 ̂𝟏 − 𝒑
̂𝟐 Sampling Distributions: Difference of Two Means, 𝒙 ̅𝟏 − 𝒙 ̅𝟐
• The mean of the sampling distribution of 𝑝̂1 − 𝑝̂ 2 is 𝜇𝑝̂1 −𝑝̂2 = 𝑝1 − 𝑝2 • The mean of the sampling distribution of 𝑥̅1 − 𝑥̅2 is 𝜇𝑥̅1 −𝑥̅2 = 𝜇1 − 𝜇2
• The standard deviation of the sampling distribution of 𝑝̂1 − 𝑝̂ 2 is: • The standard deviation of the sampling distribution of 𝑥̅1 − 𝑥̅2 is:
𝑝1 (1 − 𝑝1 ) 𝑝2 (1 − 𝑝2 ) 𝜎12 𝜎22
𝜎𝑝̂1 −𝑝̂2 = √ + 𝜎𝑥̅1 −𝑥̅2 = √ +
𝑛1 𝑛2 𝑛1 𝑛2
as long as the 10% condition is satisfied: 𝑛 ≤ 0.10𝑁. as long as the 10% condition is satisfied: 𝑛 ≤ 0.10𝑁.

• The sampling distribution of 𝑝̂1 − 𝑝̂ 2 is approximately Normal as long • The sampling distribution of 𝑥̅1 − 𝑥̅2 is approximately Normal as long
as the Large Counts Condition is satisfied: as either:
𝑛1 𝑝1 ≥ 10 𝑎𝑛𝑑 𝑛1 (1 − 𝑝1 ) ≥ 10 o Both Population are Normal or approximately Normal
𝑛2 𝑝2 ≥ 10 𝑎𝑛𝑑 𝑛2 (1 − 𝑝2 ) ≥ 10 o The central limit theorem applies (𝑛 ≥ 30) to Both
Populations
Sample Problem Sample Problem

In a very large high school, the junior class has 800 students, 54% of The distribution of heights for U.S. adult males follows a Normal
whom are female. The senior class has 700 students, 49% of whom are distribution with mean 69.3 inches and standard deviation 2.94 inches.
female. The student council selects a random sample of 40 juniors and For adult females, the distribution of heights follows a Normal
a separate random sample of 35 seniors. Let 𝑝̂𝑗𝑟 − 𝑝̂𝑠𝑟 be the difference distribution with a mean of 63.8 inches and standard deviation of 2.80
in the sample proportions of females. Estimate the probability that the inches. Suppose that a researcher selects an SRS of 30 males and an SRS
proportion of females in the sample from the junior class is less than of 50 females. Let xM − xF be the difference in the sample mean height
the proportion of females in the sample from the senior class. between the sample of males and the sample of females. Estimate the
𝑝 = 𝑡𝑟𝑢𝑒 𝑝𝑟𝑜𝑝𝑜𝑟𝑡𝑖𝑜𝑛 𝑜𝑓 𝑓𝑒𝑚𝑎𝑙𝑒𝑠, 𝐽𝑅 = 𝑗𝑢𝑛𝑖𝑜𝑟𝑠, 𝑆𝑅 = 𝑠𝑒𝑛𝑖𝑜𝑟𝑠 probability that the difference in sample means is within 1 inch of the
true difference in means.
Approximately Normal, because are all ≥ 10.
𝜇 = 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑚𝑒𝑎𝑛 ℎ𝑒𝑖𝑔ℎ𝑡𝑠, 𝑀 = 𝑚𝑎𝑙𝑒, 𝐹 = 𝑓𝑒𝑚𝑎𝑙𝑒
40(0.54) = 21.6 40(0.46) = 18.4
35(0.49) = 17.15 35(0.51) = 17.85
Approximately Normal, because both population distributions are Normal.
 pˆ jr − p
ˆ sr = 0.54 − 0.49 = 0.05 𝜇𝑥̅𝑀 −𝑥̅𝐹 = 69.3 − 63.8 = 5.5 𝑖𝑛𝑐ℎ𝑒𝑠
Because 30 < 10% of all U.S. adult males and 50 < 10% of all U.S. adult
Because 40 < 10% or all juniors and 35 < 10% of all seniors, females:
2.942 2.802
0.54(0.46) 0.49(0.51) x = + = 0.667
 pˆ jr − p
ˆ sr = + M − xF
30 50
40 35
= 0.1155 4.5 − 5.5 6.5 − 5.5
𝑃(4.5 < 𝑥̅𝑀 − 𝑥̅𝐹 < 6.5) = 𝑃 ( <𝑧< ) = 0.8662
0.667 0.667
0 − 0.05
𝑃 ((𝑝̂𝑗𝑟 − 𝑝̂𝑠𝑟 ) < 0) = 𝑃 (𝑧 < ) = 0.3325
0.1155