Questions

Week 10

Y=Raw Score

1)Let’s assume the education is normally distributed. Using GSS data, we find the mean number of years of education is 13.47 with a
standard deviation of 3.1. A total of 1,496 respondents were included in the survey.

a)If you have 13.47 years of education, that is, the mean number of years of education, what is your Z score?

For an individual with 13.47 years of education, his or her Z score would be

b)If your friend is in the 60th percentile, how many years of education does she have?

Since our friend’s number of years of education completed is associated with the 60th percentile, we need to solve for Y.
However, we must first use the logic of the normal distribution to find Z. For any normal distribution, 50% of all cases will fall above the mean.
Since our friend is in the 60th percentile, we know that the area between the mean and our friend’s score is 0.10.
Similarly, the area beyond our friend’s score is 0.40.
We can now look in Appendix B column “B” for 0.10 or in column “C” for 0.40. We find that the Z associated with these values is 0.25. Now, we
can solve for Y:

c)How many people have between your years of education (13.47) and your friend’s years of education

Since we already know that the proportion between our number of years of education (13.47) and our friend’s number of years of education
(14.25) is 0.10,

We can multiply N (1,496) by this proportion.

=1,496*0,10

Thus, 149.6 people have between 13.47 and 14.25 years of education.
2)SAT scores are normed so that, in any year, the mean of the verbal or math test should be 500 and the standard deviation 100. Assuming this is
true, answer the following questions.
a)What percentage of students score above 625 on the math SAT in any given year?

The area beyond the Z is about 0.1056, so 10.56% of students should score above 625.

b)What percentage of students score between 400 and 600 on the verbal SAT?
The area between this score and the mean is 0.3413.

Questions 1
 c)A college decides to liberalize its admission policy. As a first step, the admissions committee decides to exclude only those applicants
scoring below the 20th percentile on the verbal SAT. Translate this percentile into a Z score. Then, calculate the equivalent SAT verbal test
score.

The 20th percentile is equivalent to a Z score of about –0.84 because about 0.20 of the area of the distribution lies below it. Then the SAT verbal
score equivalent is 416.

3)The number of hours people work each week varies widely for many reasons. Using the 2010 GSS, you find the mean number of hours
worked last week was 40.62, with a standard deviation of 15.26 hr, based on a sample size of 838.
a)Assume that distribution is normal. What is the probability that someone in the sample will work 60 hr or more in a week? How many
people in the sample of 838 have worked 60 hr or more?
The area between the value and the upper tail of the distribution is 0.1020. So, the probability that someone will work more than 60 hours per
week is 0.1020. This translates into approximately 85 (838 x 0.1020) respondents in the sample.

b)What is the probability that someone will work 30hr or fewer in a week? How many people does this represent in the sample?
The area between the value and the lower tail of the distribution is 0.2420. So, the probability that someone will work less than 30 hours per
week is 0.2420. This translates into approximately 203 (838 x 0.2420) respondents in the sample.

c)What number of hours worked per week corresponds to 60th percentile?

The 60th percentile corresponds to an area of 0.60; that’s 0.50 + 0 .10. An area of 0.10(z score değil karşısındaki değer Area Between Mean and
Z) in column B corresponds to a Z score of approximately 0.25.
So, a raw score of 44.44 acts as the 60th percentile in this distribution.

Week 11

1)Explain which of the following is a statistic and which is a parameter.

a)The mean age of Americans from the 2010 decennial census.

-Although there are problems with the collection of data from all Americans, the census is assumed to be complete, so the mean age would be a
parameter.

b)The unemployment rate for the population of U.S. adults, estimated by the government from a large sample

Questions 2
 -A statistic because it is estimated from a sample.

c)The percentage of Texans opposed to the health care reform bill from a poll 1,000 residents
-A statistic because it is estimated from a sample.
d)The mean salaries of employees at your school (e.g., administrators, faculty, maintenance)
-A parameter because the school has information on all employees.

e)The percentage of students at your school who receive financial aid

-A parameter because the school would have information on all its students.

2)

a) Calculate the mean and standard deviation of the population.

Mean = 22,766
SD = 14,687
b) Take 5 samples of size 4. (Use simple random sampling). Calculate the mean for each sample. Calculate the mean of means.

5 sample, n = 4
Rastgele 5 tane örnek seçelim (kitabın yaptığı gibi):

Sample 1
Cases: 1, 5, 8, 10

(11350+17458+18342+22545)/4=17,924

Questions 3
 Sample 2
Cases: 3, 7, 11, 15
(41654+15436+25345+18923)/4=25,340

Sample 3
Cases: 2, 6, 14, 18

(7859+8451+47567+16452)/4=20,082

Sample 4
Cases: 4, 9, 16, 20

(13445+19354+16456+25671)/4=18,732

Sample 5
Cases: 12, 13, 17, 19

(68100+9368+27654+23890)/4=32,253

Mean of Means
(17924+25340+20082+18732+32253)/5=22,866(17924 + 25340 + 20082 + 18732 + 32253)/5 = 22,866
(17924+25340+20082+18732+32253)/5=22,866

≈ 22,766

✔️ Population mean’e çok yakın.

c) Take 5 samples of size 7 (Use simple random sampling). Calculate the mean for each sample. Calculate the mean of means.

5 sample, n = 7

Sample 1
1, 4, 7, 10, 13, 16, 19
Mean = 17,848

Sample 2
2, 5, 8, 11, 14, 17, 20
Mean = 23,843

Sample 3
3, 6, 9, 12, 15, 18, 1

Mean = 25,355

Sample 4
4, 7, 10, 13, 16, 19, 2

Mean = 18,989

Sample 5
5, 8, 11, 14, 17, 20, 3
Mean = 29,880

Mean of Means
(17848+23843+25355+18989+29880)/5=22,983(17848 + 23843 + 25355 + 18989 + 29880)/5 = 22,983
(17848+23843+25355+18989+29880)/5=22,983

≈ 22,766

✔️ Yine population mean.

3)When talking a random sample from a very large population, how does the standard error of the mean change when

a)The sample size is increased from 100 to 1600?

Questions 4
 b)The sample size is decreased from 300 to 150?

c)The sample size is multiplied by 4?

Week 12

Questions 5
 If there is no standart deviation than he is asking for mean(şüpheli)
1)

(Sy nin üstünde çizgi var aradaki soru işaretleri de eşittir işareti)

Questions 6
Questions 7
Questions 8
 Week 13

BU SLAYT = tek örneklem Z hipotez testinin “full sınav makinesi” hali.

Şimdi bunu çözülebilir refleks şemasına çeviriyorum 👇

ADIM ADIM NE YAPMIŞ?

Questions 9
 1️⃣ Varsayımlar
Random sample ✔
Interval/ratio ✔
n > 50 ✔
→ Z testi yapılabilir.

2️⃣ Hipotez

(“Mean 28,895’ten büyük mü?” sorusu)

3️⃣ Güven & kritik değer

Level Zcritical

95% 1.96

(one-tailed)

4️⃣ Veriler
Değer

n = 100

Sample mean = 24,100

Population SD = 23,335

5️⃣ Z hesaplaması

6️⃣ Karar

➡️ H₀ reddedilir.
Yorum
Gerçek ortalama 28,895 USD’den anlamlı derecede düşüktür.

1)It is known that, nationally, doctors working for health maintenance organizations (HMOs) average 13.5 years of experience in their specialties,
with a standard deviation of 7.6 years. The executive director of an HMO in a Western state is interested in determining whether or not its doctors
have less experience than the national average. A random sample of 150 doctors from HMOs shows a mean of only 10.9 years of experience.

a)State the research and null hypotheses to test whether or not doctors in HMO have less experience than the national average.

b)Using an alpha level of .01, make this test.

Questions 10
 Bu slayt “population variance bilinmiyorsa yapılan tek örneklem t-testi”nin birebir sınav şablonu.

Yani Z değil → t-test.

Şimdi sana bunu yine otomatik refleks haline getiriyorum 👇

ADIM ADIM NE YAPIYOR?

1️⃣ Varsayımlar
Random sample ✔
Interval/ratio ✔
n > 50 ✔
→ t-test yapılabilir.

2️⃣ Hipotez

Questions 11
 (Yani “Türk seçmenlerin ortalaması 5.5’ten büyük mü?”)

3️⃣ Confidence & kritik

Level t critical

90% 1.65

4️⃣ Veriler
Değer

n = 1143

x̄ = 6.2

s = 2.42

d.f = 1142

5️⃣ Test statistic

6️⃣ Karar & yorum

Kural

t calc > t critical → Reject H₀

8.76 > 1.65

➡️ Reject H₀
Yorum (bunu yaz):
At the 10% significance level, we reject the null [Link] is sufficient statistical evidence that Turkish voters’ average left–right
positioning is significantly higher than 5.5.

2)One way to check on how representative a survey is of the population from which it was drawn is to compare various characteristics of the
sample with the population characteristics. A typical variable used for this purpose is age. The 2010 GSS of the American adult population found
a mean age of 49.28 and a standard deviation of 17.21 for its sample of 4857 adults. Assume that we know from census data that the mean age
of all American adults is 37.2. Use this information to answer these questions.
a)State the research and the null hypotheses for a two-tailed test.

b)Calculate t statistic and test the null hypothesis at the .001 significance level.

Questions 12
 a) Hipotezler (two-tailed)
Popülasyon ortalaması biliniyor:

(“GSS örnekleminin yaş ortalaması, gerçek ABD ortalamasından farklı mı?”)

b) t-istatistiği
Veriler:

Değer

( x(üssü) = 49.28 )

( s = 17.21 )

( n = 4857 )

( mean(salak işaret yukardaki) = 37.2 )

Questions 13
 İşte bu slayt proportion için Z-hipotez testinin birebir sınav şablonu.

Yani “% üzerinden hipotez testi.”

Şimdi bunu otomatik refleks haline getirelim 👇

ADIM ADIM NE YAPIYOR?

1️⃣ Varsayımlar
Random sample ✔
Proportion ✔
2️⃣ Hipotez

(Yani “Onay oranı %50.1’den büyük mü?”)

3️⃣ Confidence & kritik

Level Zcritical

95% 1.65

4️⃣ Veriler
Değer

Questions 14
 n = 1024

p̂ = 0.54

π₀ = 0.501

5️⃣ Test statistic

6️⃣ Karar & Yorum

➡️ Reject H₀
Yazılacak yorum:
At the 95% confidence level, we reject the null [Link] is sufficient statistical evidence that the percentage of Turkish citizens who
approve full EU membership is significantly greater than 50.1%.

Eğer ki

“Test this hypothesis at the 5% significance level.”

“At α = 0.01 …”

“Use a 95% confidence level.”

Questions 15
 gibi bir şey demiyorsa ve confidence levelı bana bıraktıysa 95 alabilirim

1️⃣ İlk baktığın kelime = TEST TİPİNİ SEÇTİRİR

Soruda gördüğün Beynin otomatik açtığı

“mean / average” Mean test

“percentage / proportion / %” Proportion test

“difference / changed / increased / decreased” Hypothesis testing

“estimate / how many” CI or projection

2️⃣ İkinci baktığın şey = KUYRUK SAYISI

Kelime Test

“different / changed / any difference” Two-tailed

“increased / more than / higher than” One-tailed right

“decreased / less than / lower than” One-tailed left

3️⃣ Üçüncü baktığın = SD var mı?

Görürsen Seç

σ (population SD) verilmiş Z-test (mean)

σ yok, s var t-test (mean)

% var Z-test (proportion)

4️⃣ Dördüncü = Confidence bilgisi var mı?

Durum Yap

Verilmiş Onu kullan

Yok 95%

5️⃣ SORUYU GÖRDÜĞÜN AN BEYNİNDE AÇILACAK OTOMATİK MENÜ

Sorunun şekli Yapacağın

% + artış / azalış Proportion Z test

Mean + karşılaştırma Mean Z / t test

“How many” Projection

“Between / percentile” Z / sampling distribution

6️⃣ Örnek (senin UCLA sorusu)

Soruda:

% var

“has increased?” var

eski oran verilmiş

yeni sample var

Beyin otomatik:

Proportion + one-tailed + Z-test

Questions 16