Vous êtes sur la page 1sur 15

Homework 2

Multiple Choice Questions

1) D
Sample size 81 Distribution unknown
Mean 200
Stdev 18 Sample mean = 200
Sample Stdev = 2

2) A
Sample size 64
Mean 180
Stdev 24 Sample mean = 180
Sample Stdev = 3

P(183 < xbar < 186) = 0.135905 0.048968

3) D
Population (N) 500
Sample size (n) 225
n/N 0.45 >> 0.05 => Population is finite

Variance 900
Stdev 30 (Infinite Pop) Sample Stdev = 2.0
Finite Population Correction factor = 0.742363

Sample Stdev = 1.484725

4) B
Sample size 100
Mean 600
Stdev 50

At 95% confidence Z(0.95)= 1.96 1.959963985

Margin of Error = 9.8

5) C

6) B

7) A

8) A

9) A
10) B
Problem 1

Battery Producer - "D" size batteries

Expected life 87 hours


Stdev (population) = 9 hours

Management believes that due to improved production process there is an Increase in battery life.

Sample
Size (n) 36 batteries
Average life 88.5 hours

Sample mean
Stdev 1.5 hours

a) Null and Alternative Hypothesis


H0 mean<= 87 hours
Ha mean > 87 hours

b) Test Statistic
(xbar - mean0)/sample stdev

1.00 2.326348

c) At 99% confidene use Critical Value Approach to test management's belief


Upper tail test
Reject H0 if z >= z(alpha)

99% confidence interval => alpha = 0.01


1-alpha = 0.99

z(alpha)= 2.326348 2.326348

Since z < z(alpha), we do not reject the null hypothesis. (i.e., we reject the alternate bypothesis)
Conclusion: There is insufficient information to support the corporation's claim that average life > 87 hours 99% of th

d) What is the p-value associated with the sample results?

p-value = 0.1587 which is > alpha (0.01). (upper tail test)

Since p-value > alpha, we do not reject the null hypothesis. (i.e., we reject the alternate bypothesis)
Conclusion: There is insufficient information to support the corporation's claim that average life > 87 hours 99% of th
e life > 87 hours 99% of the time.

(Of course, the conclusion is the same irrespective of the method)


e life > 87 hours 99% of the time.
Problem 2

Price of Gasoline

Average price $ 2.20 per gallon


Stdev (population) = $ 0.12

Because of cost reduction measures, it is believed there has been a significant reduction in average price.

Sample
Size (n) 36 gas stations
Average price $ 2.14 per gallon

Sample mean
Stdev $ 0.02

a) Null and Alternative Hypothesis


H0 mean>= $ 2.20 per gallon
Ha mean < $ 2.20 per gallon

b) Test Statistic
(xbar - mean0)/sample stdev

-3.00

c) What is the p-value associated with the sample results?

p-value = 0.0013 which is > alpha (0.01). 0.00135

d) At 95% confidence, test the company's claim

At 95% confidence Alpha = 0.05 (lower tail test)

Since p-value < alpha (0.05), we reject the null hypothesis. (i.e., we accept the alternate hypothesis)
Conclusion: There is sufficient information to support the company's claim that average price < $2.20 at least 95% of the time.
erage price.

< $2.20 at least 95% of the time.


Problem 3

Account balances at credit card company

Stdev (population) = $ 200.00

Is the mean of all account balances (population mean) significantly different from $1000.

Sample
Size (n) 64 account balances
Average balance $ 1,040.00

a) Null and Alternative Hypothesis


H0 mean = $ 1,000
Ha mean NOT = $ 1,000

b) Test Statistic
(xbar - mean0)/sample stdev

1.60

c) What is the p-value associated with the sample results? 0.109599

p-value = 0.1096 There is alpha/2 area on both ends of the normal distribution curve
Find one area using NORMDIST and multiply by 2.
d) At 95% confidence, test the company's claim

At 95% conAlpha = 0.05

(Two tailed test)


Since p-value > alpha (0.05), we do not reject the null hypothesis. (i.e., we reject the alternate hypothesis)
Conclusion: There isn't sufficient evidence to say that the population mean is different from $1000.

e) Using the critical value approach at 95% confidence, test the hypotheses

Zcritical Two sided range -1.95996 to 1.959964

(Two tailed test)


Since the test statistic, z = 1.60, lies within the Zcritical range, we do not reject the null hypothesis.
Conclusion: There isn't sufficient evidence to say that the population mean is different from $1000.
normal distribution curve

e hypothesis)

(same as above)
Problem 4

Weight of Coffee in Cans

Weight marked on can = 12.00 ounces


Stdev (population) = UNKNOWN

We want to test whether the cans contain at least 12 ounces of coffee.

Sample
Size (n) 50 cans
Average weight 11.80 ounces
Stdev 0.50 ounces

Sample mean
Stdev 0.070711 ounces

a) Null and Alternative Hypothesis


H0 mean >= 12.00 ounces
Ha mean < 12.00 ounces

b) Test Statistic
(xbar - mean0)/sample mean stdev

-2.83 -2.83 -1.67655

c) What is the p-value associated with the sample results? 0.003365

p-value = 0.0034 Population Stdev unknown => Use t-distribution.


TDIST knows only how to calculate P(z>2.83) using one-tail, w
(Even if you use tables, T-distribution contains only the positi

Since p-value < alpha (0.05), we reject the null hypothesis. (i.e., we accept the alternate hypothesis)
Conclusion: There is insufficient information to support the claim that average weight of coffee per can is >= 12 ounces at least
Use t-distribution.
late P(z>2.83) using one-tail, which is the same as the desired P(z<-2.83) one-tail.
bution contains only the positive value i.e., right side of the curve. Use symmetry to figure out the left side if needed).

per can is >= 12 ounces at least 95% of the time.


if needed).
Problem 5

Length of bars cut by lathe.

Average length of bar = 6.00 cm


Stdev (population) = UNKNOWN

We are trying to see if the lathe is in perfect adjustment, i.e., producing bars of 6 cm average length.

Sample
Size (n) 121 bars
Average length 6.08 cm
Stdev 0.44 cm

a) Null and Alternative Hypothesis


H0 mean = 6.00 cm
Ha mean NOT = 6.00 cm

b) Test Statistic
(xbar - mean0)/stdev of sample means

-2.00
-1.97993
c) What is the p-value associated with the sample results? Conclusion?

p-value = 0.0478 Population Stdev unknown => Use t-distribution.


Need to use 2 in the TDIST tail specification to let the function know
Gives the area under the curve on each end.

Since p-value < alpha (0.05), we reject the null hypothesis. (i.e., we accept the alternate hypothesis)
Conclusion: There is insufficient information to support the claim that average length of a bar = 6.00 cm with 95% confidence.
on to let the function know this is a two tailed test.

cm with 95% confidence.

Vous aimerez peut-être aussi