Week 2 Assignment

Chapter 8

21. Sampling error is the difference between a sample statistic and its corresponding population parameter. Typically we would not expect for the sampling error to be zero unless the sample is an exact representation of the population. If the sampling error is zero then it is true of the population.

22. The reasons for sampling are because to capture the entire population would be physically impossible, time consuming, costly, destructive. Also sampling is proven to adequate of the population. Some populations are infinite and it would be impossible to contact everyone. We could not physically contact everyone in the world because people are constantly dying and being born. The cost to observe the entire population would be too expensive and time consuming, companies may not have the resources to send everyone a questionnaire on the experience they had.

34. c. u=110,000, s=40,000

a. n=50

SE = s/sqrt(n) = 40000/sqrt(50) = 5656.85425

b. x=112,000

z=(x-u)/SE = (112000-110000) / 5656.85425 = 0.3536

c. z=(x-u)/SE = (112000-110000) / 5656.85425 = 0.3536

d. z=-1.7678

so our probability is 1-0.0385 = 0.9614

e. The probability is then 0.6381 - 0.0385 = 0.5996

Chapter 9

32.

a. 3.01 pounds

b. mean - z*sd/sqrt(N) to mean + z*sd/sqrt(N)

z (from a table) is 1.96

N = 36

sd = 0.03

mean = 3.01

3.01 - 1.96*0.03/sqrt(36) to 3.01 + 1.96*0.03/sqrt(36)

3.0002 to 3.0198

34. mean - z*sd/sqrt(N) to mean + z*sd/sqrt(N)

z = 1.96 (from a table)

N = 50

sd = 6.2

mean = 26

26 - 1.96*6.2/sqrt(50) to 26 + 1.96*6.2/sqrt(50)

24.281 to 27.719

The value of 28 weeks in not inside that confidence interval, so it is not reasonable that the population mean is 28 weeks.

46. The sample proportion is 14/220 = 0.0636

The confidence interval is:

.0636 +/- 1.96 * sqrt(.0636*.9364/220) = (0.021, 0.106)

Since the interval includes 10%, there is not enough evidence to suggest that the population proportion is not...

Chapter 8

21. Sampling error is the difference between a sample statistic and its corresponding population parameter. Typically we would not expect for the sampling error to be zero unless the sample is an exact representation of the population. If the sampling error is zero then it is true of the population.

22. The reasons for sampling are because to capture the entire population would be physically impossible, time consuming, costly, destructive. Also sampling is proven to adequate of the population. Some populations are infinite and it would be impossible to contact everyone. We could not physically contact everyone in the world because people are constantly dying and being born. The cost to observe the entire population would be too expensive and time consuming, companies may not have the resources to send everyone a questionnaire on the experience they had.

34. c. u=110,000, s=40,000

a. n=50

SE = s/sqrt(n) = 40000/sqrt(50) = 5656.85425

b. x=112,000

z=(x-u)/SE = (112000-110000) / 5656.85425 = 0.3536

c. z=(x-u)/SE = (112000-110000) / 5656.85425 = 0.3536

d. z=-1.7678

so our probability is 1-0.0385 = 0.9614

e. The probability is then 0.6381 - 0.0385 = 0.5996

Chapter 9

32.

a. 3.01 pounds

b. mean - z*sd/sqrt(N) to mean + z*sd/sqrt(N)

z (from a table) is 1.96

N = 36

sd = 0.03

mean = 3.01

3.01 - 1.96*0.03/sqrt(36) to 3.01 + 1.96*0.03/sqrt(36)

3.0002 to 3.0198

34. mean - z*sd/sqrt(N) to mean + z*sd/sqrt(N)

z = 1.96 (from a table)

N = 50

sd = 6.2

mean = 26

26 - 1.96*6.2/sqrt(50) to 26 + 1.96*6.2/sqrt(50)

24.281 to 27.719

The value of 28 weeks in not inside that confidence interval, so it is not reasonable that the population mean is 28 weeks.

46. The sample proportion is 14/220 = 0.0636

The confidence interval is:

.0636 +/- 1.96 * sqrt(.0636*.9364/220) = (0.021, 0.106)

Since the interval includes 10%, there is not enough evidence to suggest that the population proportion is not...