Gregory Walker

MHA 610 Introduction to Biostatistics

Prof Jean Benzel-Lindley

May 6th, 2013

7.9 How many people would need to be studied to have 80% power
under the assumptions in Problem 7.7

N= (.54)²(0.84+1.96) = 230
(0.1)²

7.46 Perform a significance test to assess whether there is an

association between mortality from cirrhosis of the liver and
duration of employment in the group hired after 1945.Report a
p-value
P value = 2*[1(0.0334+0.1135+0.1929+0.2186)]

=2*0.4416

=0.8833

Therefore, there is no significant association between mortality from cirrhosis of the liver and
duration of employment in the group hired after 1945.

7.63 Suppose we want to compare the observed and expected

number of events among the group with <250 treatments.
Perform an appropriate significance test, and report a two-tailed
p-value.
H0:=3.7 vs. H1:3.7

Number of observed events , x=5

Because x=5; 0=3.7, the p value is given by

P value = 2

K! = 2*[1(0.0247+0.0915+0.1692+0.2087+0.1931)]

This study source was downloaded by 100000855936879 from CourseHero.com on 03-18-2023 12:14:56 GMT -05:00

https://www.coursehero.com/file/8036526/MHA-610-Week-4-Assignment/
 K=0

= 2*(10.6872)

= 0.6256

7.64 Provide a 95% CI for the expected number of events in the

group with ≥ 250 treatments.
Here, the observed number of events, x=6

The two sided 95% CI based on x

=6 is (2.20, 13.06)

7.108 What is the 25th percentile of a χ 2 distribution with 20

degrees of freedom? What symbol is used to denote this value?
From, Table 6 in the Appendix, the 25th percentile of a

X distribution with 20 degrees of freedom is given by,

20, 0.25 =15.45

7.109 Suppose we wish to test the hypothesis H0: µ = 2 vs. H1: µ ≠ 2.

We find a two-sided p-value of .03 and a 95% CI or µ of (1.5,
4.0). Are these two results possibly compatible? Why or why not?
Given that two sided p value is 0.03.

Since this p value is less than 0.05, we have sufficient evidence to reject the null hypothesis.

8.5 What is the p-value corresponding to your answer to Problem

8.4?
X1 = 6.56; s1=0.64, n1=25, x2=6.80, s2=0.76, n2=40

(n1 1)s12 + (n2 1)s2

=[(251)*(0.64²)+(401)*(0.76²)]/(25+402)=0.5136

n1+n2 2

s=0.7167

t=(6.566*80)/[(0.7167* sqrt(1/25+1/40)] = 1.3135

p value = 2*p(t63, 1.315)

This study source was downloaded by 100000855936879 from CourseHero.com on 03-18-2023 12:14:56 GMT -05:00

https://www.coursehero.com/file/8036526/MHA-610-Week-4-Assignment/
 =0.1938

8.6 Compute a 95% CI for the difference in means between the two
groups.
The 95% confidence interval for the difference in means between the two groups is given by

n1 n2

=(0.242*0.7167*Sqrt(1/25+1/40),0.24+2*0.7167*Sqrt(1/25+1/40))

=(0.240.3654,(0.24+0.3654)

=(0.6054,0.1254)

8.59 Conduct the test mentioned in Problem 8.58, and report a p-

value.

8.60 Provide a 95% CI for the true mean change in overnight

sodium excretion over a 1-week period
95% confidence interval is given by:

Xbardt(/2,n1)*sd/n

1.14t(0.025,7)*12.22/9

1.142.365*12.22/9

1.149.63

= 8.49, 10.77

8.112 Implement the test mentioned in Problem 8.111, and report a

p-value (two-tailed).
Null Hypothesis (Ho):1=2

Alternative Hypothesis (Ha):12

WehaveXbar1=5.3, Xbar2=5.1, s1=1.4, s2=0.8, n1=356andn2=1081

Z= (Xbar1Xbar2)/s1/n1+s2/n2

= (5.35.1)/(1.4)/356+(0.8)/108

=0.2/0.1069

This study source was downloaded by 100000855936879 from CourseHero.com on 03-18-2023 12:14:56 GMT -05:00

https://www.coursehero.com/file/8036526/MHA-610-Week-4-Assignment/
 =1.87

Using Z tables, the p value is

P [Z1.87] =0.0615

Since p value is greater than 0.05, we have sufficient evidence to reject the null hypothesis.

8.164 Perform the test in Problem 8.163, and report a two tailed p-
value
Null Hypothesis (Ho):d=0

Alternative Hypothesis (Ha):d0

We have Xbard=0.64,sd=1.44andn=10

t= (Xbardd)/(sd/n)

= (0.640)/(1.44/10)

=1.405

Degrees of freedom=n1

=101

=9

Using T table, the p value is

P [t91.405]=0.1936

8.165 Provide a 95% CI for the mean difference in HgbA1c before

minus the mean HgbA1c after use of the insulin pump.
95% confidence interval is given by:

Xbardt(/2,n1)*sd/n

0.64t(0.025,9)*1.44/10

0.642.262*1.44/10

0.641.03

=0.39, 1.67

This study source was downloaded by 100000855936879 from CourseHero.com on 03-18-2023 12:14:56 GMT -05:00

https://www.coursehero.com/file/8036526/MHA-610-Week-4-Assignment/
 10.1 How many participants need to be enrolled in each group to
have a 90% chance of detecting a significant difference using a
two-sided test with α = .05 if compliance is perfect?
We have 1=0.10, 2=0.05, p1=1/325=0.003, p2=0.997

p1*=(11)p1+1p1

= (10.10)(0.003)+0.10*0.997

=0.1024

p2*= (12)p2+2p1

= (10.05)(0.997)+0.05*0.003

=0.9473

p*=(112)|p1p2|

=0.8449

The sample size is given by:

N= [2p*q(Z1/2)+p1*q1*+p2*q2(Z1)]/(|p1p2|)

= [2*0.8449*0.1551*1.96+0.1024*0.8976+0.0527*0.9473(0.84)]/(0.994)

=16302.5

=16303 patients need to be enrolled

10.16 Compute the Kappa statistic and its standard error regarding
reproducibility of the diagnosis of dyspnea in this clinic.
po=(7+82)/100=89/100=0.89

a1=15/100=0.15

a2=85/100=0.85

b1=10/100=0.10

b2=90/100=0.90

pe=a1b1+a2b2=0.15*0.10+0.85*0.90=0.78

Thus,

Kappa = (pope)/(1pe)=(0.890.78)/(10.78)=0.11/0.22=0.50

This study source was downloaded by 100000855936879 from CourseHero.com on 03-18-2023 12:14:56 GMT -05:00

https://www.coursehero.com/file/8036526/MHA-610-Week-4-Assignment/
 The standard error of kappa statistic is given by,

=Sqrt{[1/100*(10.78)2]*(0.78+0.7821.3425)}

=0.0974

10.38 Implement the test procedure in Problem 10.37, and report a

p-value.
We have Xbard =0.40, sd=2.81 and n=398

t=(Xbardd)/(sd/n)

=(0.40)/(2.81/398)

=2.84

Degrees of freedom=n1

=3981

=397

Using t tables, the p value is

P[t&lt;2.84]=0.0024

10.62 Implement the test in Problem 10.61, and report a two-tailed

p-value
X=(OE)/E

=2.40

Degrees of freedom=(r1)(c1)

=4

Using chi square tables, the p value is

P[X42&gt;2.40]=0.6627

10.69 Implement the test in Problem 10.68, and report a two-tailed

p-value.
Here,n1=133,x1=15,n2=138,x2=2

p1 =x1/n1=15/133=0.1128, p2 =x2/n2=2/138=0.0145

This study source was downloaded by 100000855936879 from CourseHero.com on 03-18-2023 12:14:56 GMT -05:00

https://www.coursehero.com/file/8036526/MHA-610-Week-4-Assignment/
 Now, p =(x1+x2)/(n1+n2)=(15+2)/(133+138)=17/271=0.0627

Since n1 pq =133*0.0627*0.9373=7.82&gt;5and n2 pq =138*0.0627*0.9373=8.11&; 5

We can use test for Binomial proportion using Normal theory.

The test statistic is given by,

={|0.11280.0145|(1/2*133+1/2*138)}/Sqrt[0.0627*0.9373*(1/133+1/138)]

=0.0909/0.0295

=3.0853

The p value is given by,

P value= 2[1 (3.0853)] =2*(10.9990)=0.0020

10.121 What is the sensitivity and specificity of the test?

Sensitivity = 80/(80+5) = 0.941

Therefore, this can be considered a reliable test because the sensitivity is close to 100%

Specificity= 15/(15+5) = 0.75

Because this is a positive result, the test can be considered reliable

This study source was downloaded by 100000855936879 from CourseHero.com on 03-18-2023 12:14:56 GMT -05:00

https://www.coursehero.com/file/8036526/MHA-610-Week-4-Assignment/
Powered by TCPDF (www.tcpdf.org)