Running Head: PROBLEM SOLVING ASSIGNMENT

Problem Solving Assignment

Name.

Institutional Affiliation.

Date.
PROBLEM SOLVING ASSIGNMENT 2

Problem 1: Diameter size Quality Control

a) Step 1: Hypothesis

Ho=μ ≤ 6 ; vs

H 1=μ>6

Step 2: Set α

Reject Ho when the mean is less or equal to 6 otherwise fail to reject. Α is set to be at 3%

confidence interval.

Step 3: Data collection.

The standard deviation, otherwise denoted as δ, is at 0.26 cm. the recorded mean is at 5.9 cm.

the confidence interval is at 3%. Hence the data collected is as shown below;

δ =0.26

x=5.9

μ=6

α =0.03

n=40

Step 4: ‘z statistics

From the t tables, we determine the z tab for the above data. That is;

z tab =t { α , 40 }

From the t bales we find that the value is 2.021

x−μ
z calc =
δ

To calculate:

z= |5.9−6
0.26 |
PROBLEM SOLVING ASSIGNMENT 3

z=0.38

Representing this in a diagram,

Reject at 3%
Reject at 3%

Step 5: Decision rule

x−μ
When the value calculated from the test statistics, of z calc = is less than the value 2.021,
δ

we then reject the null hypothesis. In case the value is greater than 2.021, we fail to reject the

null hypothesisHo. Since0.38<2.021, we reject the null hypothesis. This means that we fail

to reject the alternative.

Step 6: Conclusion

The alternative decision is that the engine part is properly working hence no changes need to

be made.

p ( 1−p )
b) The p value is calculated as the square root to
n

The probability of chance in this example is;

5.9
¿
40

¿ √ of 0.1475

¿ 0.3841

This is the standard error.

To determine the p value we use the z tables,

z tab =α =0.03 , z=0.38

¿ 0.64803

Using the p value = 0.64803, the decision rule is that we reject the null hypothesis and fail to

reject the alternative hypothesis.

c) I would suggest the manager should not at any case replace the machine. This is because

despite the actual mean diameter being lower than the expected mean, the machine is still

working properly. We would reject the null hypothesis when the value of z calculated is less

than the value of z tabulated. With this we would comply with the decision that his belief is

correct.

d) Statistically this is a type II error as it is the alternative hypothesis

Problem Two: Working hours of lectures.

a) Probability of more than 58 hours per week;

S
μ=
√n
2
S
n= ()
μ

n=378

58
¿
378

¿ 0 .1535

b) Randomly selected lectures

50
¿
378
PROBLEM SOLVING ASSIGNMENT 5

¿ 0.1323

¿ 0.1323∗3

¿ 0.3968

c) Five lectures:

At 54 hours;

54
378

¿ 0.1428

At 60 hours:

60
¿
378

¿ 0.1587

For five lectures:

0.1428 ≤ 0.1535≤ 0.1587

d) Longest time :

1−Prob

1−0.0.1587

¿ 0.8413

¿ 0.8413∗378

¿ 318.01

¿ 318.01∗1 %

¿ 3.18
PROBLEM SOLVING ASSIGNMENT 6

Problem Three:

a) Dependent variable is number of crimes per day whereas the independent variable is the

number of police. This is because the number of crimes will depend on the number of police

in each region. Also through computation, the x variable, which is the independent variable

sums up to 344, suggesting the number of police. The y variable on the other hand sums up to

245, which is basically the dependent variable, hence the number of crime cases per day.

b) Sx2 is calculated as;

¿ √¿ ¿

12,742
¿
344

¿ 289.59

Sy2 is calculated as:

√¿ ¿

7,135
¿
245

¿ 29.12

Sxy is calculated as:

∑ Xi Y i 2
∑ Xi∗∑Yi

7,5092
¿
344∗245

¿ 669.02

c) Sample correlation coefficient:

Sxy
r=
√ Sx∗Sy

7,509
¿
√ 344∗245

¿ 0.2984

d) Correlation coefficient r measures the linear relationship of the variables given. The value is

more close to zero than one. This shows that the two variable, number off crime cases and

number of police do not perfectly correlate. It shows that the number of crime cases might

not be so much dependable on the number of police.

e) Slope coefficient:

y=a+bx

∑ Y =na+b ∑ x
∑ y2 =a ∑ XY +b ∑ X 2
This can be written as:

245=10 a+344 b

7,135=7,509 a+ 12,742b

Let us solve simultaneously and eliminate a:

127,420 b+75,090 a=71,350

2,583,096 b+75,090 a=1,839,705

−2,455,676 b=−1,768,355

b=0.72

f) This means that there is a marginal increase of 0.72 number of cases in a change in number

of police in the region.

g) Calculating the value of a;

b=0.72

By substitution:

¿ {127,420∗0.72 }+75,090 a=71,350

¿ 75,090 a=−20,392.4

a=−0.27

Regression equation;

y=a+bx

y=0.72 x−0.27

h) Y intercept, x=0 hence;

y=0.27

It means when there is no police the crimes number is relatively zero.

X intercept, y=0 hence;

x=0.375

This means when there is no crime the number of police is relatively at zero.

i) Regression equation;

y=a+bx

y=0.72 x−0.27

j) If 45 police are employed;

y= { 0.72∗45 }−0.27

y=33.48

The number of crimes are:

¿ 33
PROBLEM SOLVING ASSIGNMENT 9

k) Step 1: Hypothesis

Ho=μ 1 ≤ μ 2

H 1=μ 1> μ2

Step 2: Decision statistics

We reject the null hypothesis if the value of the mean is less than one, otherwise we fail to

reject Ho.

Step 3: Set α

Alpha is at 1%. This is the confidence interval for one-tailed normal population.

Step 4: Test statistic

x−μ
We use the test statistics z calc = in evaluating our hypothesis. Calculated z calc. is at
δ

-6.06.

Step 5: T tables

We find the value of z when α is at 1% and n=10.

¿ 2.764

Step 6: Decision Rule;

The decision rule is that since the z calculated is less than the value of z tab., we reject the

null hypothesisHo. This suggest that there is no negative correlation that exist between the

two variables.

l) The coefficient of determination formula is given as:

¿n¿¿

The coefficient of determination is:

10∗7,509−344∗245
¿ 2 2
√ {10∗12,742−344 }∗{10∗7,135−245 }
PROBLEM SOLVING ASSIGNMENT 10

−9,190
¿
√ 9,084∗11325
¿−0.9061

m) This coefficient of determination is a measurement used to explain the variability of a factor

related to another one. The value recorded for R2 is -0.9061 which shows that the model is

doing worse than the mean value in terms of relationship between the two variables.

Problem Four: Funding a statistic course.

a) The formula I would apply to this problem is :

1 +1
√
( x 1−x 2 ) ± zSp ( ¿ )¿
n1 n2

The reason I would use this formula is because it has values which are greater than 30 for

both n1 and n2, that is;

n 1>30 , n 2> 30

b) Confidence intervals:

n 1=105

n 2=395

( 395−1 ) 67.152 + ( 105−1 ) 17.852

Sp=60.28

x 1=3,236 , x 2=443

z=@ 0.025 ,105 , 395

z=1.960
PROBLEM SOLVING ASSIGNMENT 11

Confidence interval:

1 +1
( 3,236−443 ) ±1.960∗60.28 (
√ ¿
105 395
)¿

Confidence interval;

¿ 2,793 ±12.97

2,780 ≤ μ ≤ 2,805

c) The range or interval presented above shows that the mean between the two groups only

ranges from 2,780 to 2,805 grandaunts who pursue statistics in their field of work. There are

no extremes beyond or under the range stated.

d) No, the Dean should not allocate funding. This is because the range or the interval presented

is lower than the expected 17% of graduates.

e) The width of the confidence interval would increase when a lower sample is used. This is

because there would be an increase in the standard error in the sample size.