--%>

Compute two sample standard deviations

Consider the following data for two independent random samples taken from two normal populations.

Sample 1 14 26 20 16 14 18

Sample 2 18 16 8 12 16 14

a) Compute the two sample means and the two sample standard deviations.

b) What is the point estimate of the difference between the two population means?

c) Assuming α = .10, conduct p-value based and critical-value based hypothesis tests for the equality

of means of the two populations.

d) What is the 90% confidence interval estimate of the difference between the two population means?

How do the results compare in all the three approaches to hypothesis testing?

 

E

Expert

Verified

Mean sample 1 = X1-bar = (14+26+20+16+14+18)/6 = 18

Mean sample 2 = X2-bar = (18+16+8+12+16+14)/6 = 14

Sample 1 SD = SD1

X1

X1-X1-bar

(X1-X1-bar)2

14

-4

16

26

8

64

20

2

4

16

-2

4

14

-4

16

18

0

0

Sum of (X1-X1-bar)2 = 104

S12 = 104/6-1

        = 20.8

SD1 =  = 4.56

Sample 2 SD = SD2

X2

X1-X1-bar

(X1-X1-bar)2

18

4

16

16

2

4

8

-6

36

12

-2

4

16

2

4

14

0

0

 

Sum of (X2-X2-bar)2 = 64

S22 = 64/6-1

        = 12.8

SD2 =  = 3.58

(b)

Point estimation of difference b/w two means = 18 - 14 = 4

(c)

t-test will be applied because sample size is small.

Hypothesis Formation

Null Hypothesis H0:    µ1 - µ2 = 0

Alternative Hypothesis H1:    µ1 - µ2 ≠ 0

t Statistic

t-statistic = (X1-bar  - x2-bar)/Sp

Where SP =

                  = 2.016

Critical value

Critical value of t with df=10 at 0.1 significance level = 1.812

Critical Region

Reject null hypothesis in favor of alternative if t is greater than t critical value of 1.812 or less than -1.812.

Computation

t-statistic = (18 - 14)/2.016

   = 5.95

Decision

Null hypothesis is rejected in favor of alternative as Z value is greater than Z critical value.

(d)

90% CI of difference between means = (18-14) - 1.812*2.016

                                                                    = 4 - 1.22 < µ < 4 + 1.22

                                                                    = 2.78< µ< 5.22

   Related Questions in Basic Statistics

  • Q : Problem on queuing diagram Draw a 

    Draw a queuing diagram for the systems below and describe them using Kendall’s notation: A) Single CPU system <

  • Q : Simplified demonstration of Littles Law

    Simplified demonstration of Little’s Law:

    Q : Data Description 1. If the mean number

    1. If the mean number of hours of television watched by teenagers per week is 12 with a standard deviation of 2 hours, what proportion of teenagers watch 16 to 18 hours of TV a week? (Assume a normal distribution.) A. 2.1% B. 4.5% C. 0.3% D. 4.2% 2. The probability of an offender having a s

  • Q : Use the NW corner rule to find an

      (a) Use the NW corner rule to find an initial BFS, then solve using the transportation simplex method. Indicate your optimal objective function value. (b) Suppose we increase s1 from 15 to 16, and d3 from 10 to 11. S

  • Q : Variance and standard error A hospital

    A hospital treated 412 skin cancer patients over a year. Of these, 197 were female. Give the point estimate of the proportion of females seeking treatment for skin cancer. Give estimates of the

  • Q : Compute the stoke statistics Please do

    Please do the following and submit your results in the table format in a word file on canvas: a)      Go to Yahoo finance/Investing/Stocks/Research tools/Historical quotes/Historical prices and download adjusted monthly closing prices for the period 1/1/2006 to 31

  • Q : Designing a system What are the

    What are the questions that comes into mind when designing a system?

  • Q : Model Checking Approach Model Checking

    Model Checking Approach: • Specify program model and exhaustively evaluate that model against a speci?cation        –Check that properties hold   

  • Q : Derived quantities in Queuing system

    Derived quantities in Queuing system: • λ = A / T, Arrival rate • X = C / T, Throughput or completion rate • ρ =U= B / T, Utilization &bu

  • Q : What is Interactive Response Time Law

    Interactive Response Time Law: • R = (L/X) - Z• Applies to closed systems.• Z is the think time. The time elapsed since&nb