STATISTIKA
KELOMPOK 3
DODI PRIANTO LUMBANTOBING (13)
EKA RISKIA NABILA (14)
FITRI AYU NINGTIYAS (15)
GUSTI AYU PUTU DELLA A.D. (16)
HANDIKA PUTRA PRIHANDOYO (17)
HANNY EKA SETYANINGRUM (18)
WAWAN ARWANI (39)
REVIEW EXERCISE
question answer
1.◎True/False
Characterize each of the following a. False, if the value of P is high, the null
statements as being true or false. hypothesis (and not the alternative
a. In a hypothesis test, a very high P-value indicates hypothesis) is supported.
strong support of the alternative hypothesis.
b. The Student t distribution can be used to test a b. True, for in order to use a Student t
claim about a population mean whenever the distribution to test a claim, the data must be
sample data are randomly selected from a normally randomly selected, and the population must
distributed population. be normally distributed.
c. When using a distribution to test a claim about a
population standard deviation, there is a very loose c. False, for the requirement of a normal
requirement that the sample data be from a distribution is not "loose."
population having a normal distribution.
3
question answer
1. True/False Characterize each of the following d. False, for people who are more likely to
statements as being true or false. respond also may be more likely to have a
d. When conducting a hypothesis test about the passport, meaning that a larger sample size
claimed proportion of adults who have current will not help.
passports, the problems with a convenience sample
can be overcome by using a larger sample size. e. False, because the sample size is
e. When repeating the same hypothesis test with important for the results
different random samples of the same size, the
conclusions will all be the same.
4
question answer
2. Leisure Time In a Gallup poll, p◎
= 2/3
, p 2/3, = 0.01, x = 657, n = 1010,
1010 adults were randomly q = 1-2/3 = 1/3
selected and asked if they were
satisfied or dissatisfied with the
= = = 0.6505
amount of leisure time that they
had. Of this sample 657 said that z = = = -1.09
they were satisfied and 353 said P = 0.1379+(1-0.8621) = 0.2758
that they were dissatisfied. Use a P-value , fail to reject the null hypothesis
0.01 significance level to test the
claim that 2/3 of adults are satisfied
So, there is not sufficient evidence to reject that 2/3 of
with the amount of leisure time that adults are satisfied with the leisure time they have.
they have.
5
question answer
3. To gauge the popularity of a recently :◎
p = 0.5,
: p0.5, = 0.05, = 0.6, n = 900
released movie, its producer conducts a
z= = =6
poll among people who have watched
the movie. In the poll, out of 900 Critical value, z = 1.645
respondents, 60% said that they liked P- value 0
the movie. Test the claim, at a 0.05 level
The null hypothesis is rejected. Thus, there is
of significance, that the movie is liked
sufficient evidence to support the claim that more than
by more than 50% of the people who
50 % people who watched the movie like it. As the
watched it. What can you say about the
responses through e-mail were voluntary, the data is
validity of the result if the opinions of
not reliable and the result are not necessarily valid.
1000 people were sought through e-
mail and only 900 responded?
6
question answer
4. Birth Weight A simple random ◎
= 3369
g
sample of 1862 births of Chinese babies
< 3369 g
resulted in a mean birth weight of 3171
g and a standard deviation of 428 g Test statistics : t = = = -19,962
(based on “Comparison of Birth Weight Critical value : t = -2,328 dan P-value <0,005
Distributions between Chinese and
Karena P-value < , yaitu 0,005 < 0.01 dan test
Caucasian Infants,” by Wen et al.,
statistic berada di dalam critical region maka ditolak
American Journal of Epidemiology, Vol.
dan mendukung bahwa rata-rata berat bayi China
172, No 10). Use a 0.01 significance
yang lahir kurang dari 3369 g.
level to test the claim that the mean
birth weight of Chinese babies is less
than the mean birth weight of 3369 g
for Caucasian babies.
7
question answer
5. A manufacturer of electric bulbs =◎1 tahun
claims that the life of the electric bulbs
1 tahun
produced by his company have a
standard deviation of 1 year. Fifteen of Test statistics : = = = 18,515
these electric bulbs are tested and the Critical value : = 31,319 dan 4,075
standard deviation of bulbs life is found
0,2<P-value <0,4
to be s = 1,15 year. Assuming that the
times are normally distributed, test the Karena P-value > , yaitu P-value > 0.01 dan test
claim of the manufacturer at a 0,01 level statistic tidak berada di dalam critical region maka fail
of significance to reject dan dan tidak memenuhi bukti yang
mendukung bahwa standar deviasi tidak sama
dengan 1 tahun.
8
question
6. Monitoring Lead in Air Listed below are measured amounts of lead (in micrograms per
cubic meter, or mg/m3, in the air. The Environmental Protection Agency has established an air
quality standard for lead of 1.5 mg/m3. The measurements shown below constitute a simple
random sample of measurements recorded at Building 5 of the World Trade Center site on
different days immediately following the destruction caused by the terrorist attacks of
September 11, 2001. After the collapse of the two World Trade Center buildings, there was
considerable concern about the quality of the air. Use a 0.05 significance level to test the
claim that the sample is from a population with a mean greater than the EPA standard of 1.5
mg/m3.
5.40 1.10 0.42 0.73 0.48 1.10
9
answer
◎ g/
= 1,5
> 1,5 g/
= = 1,5383 g/
S = = 1,9142
Test statistics : t = = = 0,049
Critical value : t = 2,015 dan P-value > 0,10
Karena P-value > , yaitu 0,10 > 0.05 dan test statistic tidak berada di dalam critical region
maka fail to reject dan dan tidak memenuhi bukti yang mendukung bahwa rata-rata lead in
air lebih dari 1,5 g/.
10
question answer
7. Pennsylvania Lottery In the H◎
0 : =
25
Pennsylvania Match 6 lottery, six H1 25
numbers between 1 and 49 are
randomly drawn. To simulate the The value of the test statistic is then :
number selection process, a TI- t = -0.567
83/84 Plus calculator was used to The P-value is the corresponding probability which can be
randomly generate 100 numbers found in the table A-3 with df = n – 1 = 100 – 1 = 99:
between 1 and 49 inclusive. The
sample has a mean of 24.2 and a P > 0.2
standard deviation of 14.1. Use a If the P-value is less than the significance level, then reject
0.01 significance level to test the the null hypothesis:
claim that the sample is selected P > 0.01 -> fail to reject H0
from a population with a mean
equal to 25, which is the mean of There is not sufficient evidence to reject the claim that the
the population of all drawn sample is selected from a population with a mean equal to
numbers. 25.
11
question answer
8. Type I Error and Type II a. The Type I error is the probability of rejecting the null
Error hypothesis when the null hypothesis is true.
a. In general, what is a type I error? The Type II error is the probability of failing to reject the
In general, what is a type II error? null hypothesis, when the null hypothesis is false.
b. For the hypothesis test in b. The Type I error is the probability of rejecting the null
Exercise 7, write a statement that hypothesis when the null hypothesis is true. We reject the
would constitute a type I error, and claim that the sample is selected from a population with a
write another statement that would mean equal to 25, when the population mean is equal to
be a type II error. 25
The Type II error is the probability of failing to reject the
null hypothesis, when the null hypothesis is false. We do
not reject the claim that the sample is selected from a
population with a mean equal to 25, when the population
mean is not equal to 25.
12
question answer
9.◎Pennsylvania
Lottery and Test ◎ test has some requirements:
The
Assume that we want to use the same 1. The sample is a simple random sample
sample data from Exercise 7 to test
2. The population has a normal distribution
the claim that the standard deviation
of all drawn numbers is less than In Exercise 7, the numbers are selected in
15.0. Why can’t we use a test with such a way that they are all equally likely, so
the methods described in Section 8- the population has a uniform distribution
4? instead of the required normal distribution.
Because the requirements are not all
satisfied, the test shouldn’t be used.
13
question
10. The diastolic blood pressure levels (mm Hg) for a random sample of eight adults from a
community are:
91 88 89 72 73 74 83 70
Assume that the above data are from a normal distribution. Test the claim, at a 0.05
significance level, that the mean of the blood pressure of all adults in the community is less
than 80 mm Hg.
answer
n = 8
= = 80
s = = 8.652
: 80 mm Hg
: 80 mm Hg
14
answer
◎ statistic
Test
t= = =0
The P-value is the corresponding probability which can be found in table A-3 with df 8-1
=7
P > 0.5
If the P-value is less than the significance level, then reject the null hypothesis:
P > 0.05 = Don’t reject
There isn’t sufficient evidence to support the claim that the mean of blood pressure of all
adults in the community is less than 80 mm Hg.
15
CUMULATIVE REVIEW
EXERCISE
question answer
1. dictionary Words A simple ◎ diurutkan menjadi
Jika
random sample of pages from
34 36 39 43 51 53 62 63 73 79
merriam-Webster’s Collegiate
- Mean =
Dictionary, 11th edition, is
obtained. Listed below are the - Median =
numbers of words defined on - Variance = [ ]/10(10-1) = 245,1
those - Standar Deviasi = 15,66
pages. Find the values of the - Range = 79-34
indicated statistics.
= 45
51 63 36 43 34 62 73 39 53 79
a. Mean b. Median c. Standard
deviation d. Variance e. Range
17
question answer
2. dictionary Words Refer to the Answer :
sample data in Exercise 1.
a. Ratio, Karena nilainya dapat dikategorikan
a. What is the level of dan mempunyai makna .
measurement of the data (nominal,
b. Diskrit, karena berdasarkan hasil
ordinal, interval, ratio)?
perhitungan bukan berdasarkan hasil
b. Are the values discrete or pengukuran.
continuous?
c. Sampel diambil secara acak dari populasi
c. What does it mean to state that
the sample is a simple random
sample?
18
question answer
3. Confidence Interval for df◎= n-1
= 10-1 = 9
dictionary Words Use the sample
t = t0,025 = 2,262
values given in Exercise to
construct a 95% confidence Maka E = t x
interval estimate of the population =2,262 x
mean. Assume that the population =11,2
has a normal distribution.
Sehingga confidence interval adalah
Mean – E < μ < Mean + E
= 53,3 – 11,2 < μ < 53,3 + 11,2
= 42,1 < μ < 64,5
19
question answer
4. Hypothesis Test for dictionary Words Refer to Karena confidence intervalnya =
the sample data given in Exercise 1. 42,1 < μ < 64,5
Given that the dictionary has 1459 pages with Maka tidak bisa menjamin bahwa
defined words, the claim that there are more setiap halaman terdiri dari lebih
than 70,000 defined words is the same as the claim dari 48 halaman
that the mean number of defined words
on a page is greater than 48.0. Use a 0.05
significance level to test the claim that the mean
number of defined words on a page is greater than
48.0. What does the result suggest about
the claim that there are more than 70,000 defined
words in the dictionary?
20
question answer
5. Designing Cars The sitting height (from a.◎P(X>38,8). Maka Z>
seat to top of head) of drivers must be
considered in the design of a new car model. .
Men have sitting heights that are normally P untuk Z<2 adalah 0,9772
distributed with a mean of 36.0 in. and a
standard deviation of 1.4 in. (based on Sehingga P untuk Z>2 = 1 – 0,9772
anthropometric survei data from Gordon, = 0,0228 atau 2,28%
Churchill, et al.).
a. One car is designed to accommodate
sitting heights of 38.8 in. or less. Find the
percentage of men with sitting heights greater
than 38.8 in.
21
question answer
5. Designing Cars The sitting height (from
seat to top of head) of drivers must be b.◎Untuk
P = 0,98 . Z scorenya adalah 2,05
considered in the design of a new car model. maka X = μ + (z x σ )
Men have sitting heights that are normally
distributed with a mean of 36.0 in. and a = 36 + (2,05 x 1,4) = 38,87
standard deviation of 1.4 in. (based on c. Z=
anthropometric survei data from Gordon,
Churchill, et al.).
Z=
b. If the sitting height is to be changed so that Z = 1,43
98% of men will be accommodated, what P untuk Z < 1,43 adalah 0,9236
isthe new sitting height?
c. If a car is occupied by four men, find the
probability that they have a mean sitting
height less than 37.0 in.
22
question answer
6. Left-Handedness Among Americans,
97% of males are left-handed and 12,5% of a. We find that the probability is:
females are left-handed. = (.125)3 = 0.00195
a. if three females are randomly selected, This is below 1 percent, so this
find the probability that they are all left- event is unlikely.
handed. When randomly selecting three b. We find that the probability is:
females, is it unlikely that all pof them are = (.125)(.097) = 0.0121
left-handed. Why or why not ?
c. We first find the probability that
b. if one male is randomly selected and one they are all right handed:
female is randomly selected, find the
= .8755 = 0.513
probability that they are both left-handed.
Thus, the probability that there is at
c. if five females are randomly selected, find least one that is left handed is:
the probability that at least one of them is
= 1−.513 = 0.487
left-handed
23
question
7. Normal Rainfall The following histogram is obtained from the daily rainfall
amounts (in) in Boston for a year. If we want to conduct a hypothesis test of a
claim about the mean, there is a requirement that the population of rainfall
amounts must have a normal distribution. Based on the histogram, does it appear
that the population satisfies the requirement of a normal distribution? Why or why
not?
answer
The distribution is approximately
normally distributed if it has a roughly
bell-shaped, but here this doesn't apply
hence it is not normal, hence the
requirement is not satisfied.
24
question
8. Analyzing Graph The accompanying graph depicts results from a recent year
in which there were 8878 male graduates and 8203 female graduates from
medical schools in the United States. Does the graph depict the data in a way that
is fair and objective, or is it somehow deceptive? Explain.
answer
The bars of the graphs should start at 0
instead of 7000, hence the answer is
no.
25
question answer
9. Gun Survey In a recent Gallup poll,
1003 randomly selected adults in the a. The number of respondents who
United States answered “yes” is the result of the
were asked if they have a gun in their proprtion and the sample size
home, and 37.2% of them answered
“yes.” answer = 373 respondents
a. What is the number of respondents
who answered “yes”?
26
question
b. Construct a 95% confidence interval estimate of the percentage of all adults
who would answer “yes” when asked if they have a gun in their home.
answer
b. Determine the critical value using table A-2 :
Margin of error :
The confidence interval :
Result 34.21% < p < 40.19%
27
question
c. Based on a hypothesis test, can we safely conclude that less than 50% of adults
answer “yes” when asked if they have a gun in their home? Why or why not?
answer
c.
The value of the test statisic :
Determine the corresponding probability using table A-2 :
p = p(Z < -8.11) = 0.0001
If the P-value is less than the significance level, then reject the null hypothesis :
P < 0.01
There is sufficient evidence to support claim that less than 50% of adults answer “yes”
28
question answer
9. Gun Survey In a recent Gallup poll,
1003 randomly selected adults in the d. The required minimum sample
United States size is not dependent on the
were asked if they have a gun in their population size, but is dependent on
home, and 37.2% of them answered the confidence level and the sample
“yes.” proportion.
d. What is a sensible response to the
criticism that the Gallup poll cannot
provide good results
because the sample size is only 1003
adults selected from a large population
of 241,472,385
adults in the United States?
29
question
10. hypothesis Test for Gun Survey Refer to the survey results from Exercise 9 and use a
0.01 significance level to test the claim that fewer than 50% of Americans say that they have
a gun in their home.
answer
10.
The value of the test statisic :
Determine the corresponding probability using table A-2 :
P = P(Z < -8.11) = 0.0001
If the P-value is less than the significance level, then reject the null hypothesis :
P < 0.01
There is sufficient evidence to support claim that less than 50% of adults answer “yes”
There is sufficient evidence to support the claim that fewer than
50% of Americans say that they have a gun in their home
30
