{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

14 Sampling Distributions Part 3

# 3 construct and interpret a 95 confidence interval

This preview shows pages 12–20. Sign up to view the full content.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 3. Construct and interpret a 95% confidence interval for the population mean error rate in customer support. p = 176/621 = ̂ 0.283 Ma rg in o f e rro r = 1.96 *√(.283*(1-.283)/621) 9 5 % CI: (.2 4 8 , . 3 8 1 ) p ̂ + m a rg in o f e rro r = .2 8 3 + 0 .0 3 5 = 0 .0 3 5 12 Summary: Computing (1-α)*100% confidence intervals Proportion • = sample proportion • np>5; n(1-p)>5 Mean, σ unknown • s = sample standard deviation • t has n-1 degrees of freedom • n ≥ 30 if population not normally distributed Mean, σ known • n ≥ 30 if population is not normally distributed 13 n p p z p ) ˆ 1 ( ˆ ˆ ) 2 / (- ± α n s t x ) 2 / ( α ± n z x σ α ) 2 / ( ± ˆ p Determining sample size Before collecting data, need to determine the sample size n for estimating the population mean or proportion. One method is to collect enough data so that the 95% confidence interval will have a predetermined margin of error 14 Margin of Error (e): Our tolerance level for sampling error. Margin of error mples: σ = 25, n=300, and 95% conf. level: e = σ = 25, n=50, and 95% conf. level: e = 1.96 * 25 / 300 = 2.829 6.930 Determining sample size: Margin of error 15 n σ z e α/ ) 2 ( = n /2) ( σ α z x ± Margin of Error (e): Confidence Interval 1. Data variation, σ or s : e as σ 2. Sample size, n : e as n 3. Level of confidence, 1 - : e if 1 - or or or Which is bigger? 1. s = 10, = 0.05, e when n =50 vs e when n =500 2. n = 50, = 0.05, e when s =50 vs e when s =75 3. n = 50, s = 250 e when =0.01vs e when =0.05 Determining sample size: Margin of error 16 n σ z x ) (α 2 / ± n σ z e ) (α 2 / = Just rearrange margin of error to get sample size Required sample size, σ known: What if σ is unknown? Estimate σ- Select a pilot sample and estimate σ with the sample standard deviation s- If range, R, is known, may use σ = R/6 (Recall Empirical Rule, ±3σ contains virtually all data) Example Always round-up! Determining sample size, n 17 2 ) 2 / ( = e z n σ α 5 100 96 . 1 ) 2 ( = = = e z σ α = n 1537 5 100 * 96 . 1 2 = Example: Communications expenditures Want an estimate of mean monthly expenditure on communications Desire a margin of error of \$5 for the mean, at a 95% confidence level. From other work, estimate S = \$30 What should you use as a sample size? n = (1.96 * 30/ 5)2 = 138.3 use 139 18 2 ) 2 / ( = e z n σ α Margin of Error (e): Confidence Interval Determining sample size for proportions : Margin of error Finding sample size: Do the same thing for finding required sample size for proportions....
View Full Document

{[ snackBarMessage ]}

### Page12 / 27

3 Construct and interpret a 95 confidence interval for the...

This preview shows document pages 12 - 20. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online