Scribd
Upload
96 views

Tutorial 6

This tutorial checks understanding of hypothesis testing for population means when the population standard deviation is known and unknown. It provides two examples testing claims about order dispatch times and average work hours using sample data and the t-test. The second example also constructs a 95% confidence interval and tests an additional hypothesis about average hours worked.

Uploaded by

Quỳnh Nga
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
96 views

Tutorial 6

This tutorial checks understanding of hypothesis testing for population means when the population standard deviation is known and unknown. It provides two examples testing claims about order dispatch times and average work hours using sample data and the t-test. The second example also constructs a 95% confidence interval and tests an additional hypothesis about average hours worked.

Uploaded by

Quỳnh Nga
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

This tutorial checks your understanding of testing hypothesis about the population

mean when population standard deviation is known and unknown.

Question 1

A mail order firm claims that new procedures enable it to dispatch orders more
quickly than previously. Records show that the previous average delay was 20 days.
A random of 17 orders using the new procedure showed a delay time of 18 days with
a sample standard deviation of 2.5 days. Is the claim made for the new procedures
supported at the 5% level?

n = 17 x̄ = 18 s= 2.5 α = 0.05

H0: μ = 20

HA: μ < 20 (chọn nhỏ hơn vì đề bài nói quickly than previously) → gud :>

t = (x̄ - μ) / (s/√n) = (18-20) / (2.5/√17) = -3.3

(cách 2) P(t<-3.3) = 0.0005 = 0.05% < 5% → Reject null hypothesis

Question 2

In the lecture you tested the null hypothesis that college graduate students work
longer than 40 hours per week on average. Now use gss.sav file and test the
hypothesis that all full-time workers work for 40 hours per week on average
(variable hrs1).

a. What are the null and alternative hypotheses?

H0 : μ = 40

HA : μ # 40
b. Test the hypothesis and write a brief summary of your conclusions.

Step 1: Chạy bảng frequencies cho biến wrkstat xem trong mẫu có những phân loại
nào.

- Analyze -> Descriptive Statistics -> Frequencies


- Add biến wrkstat
- OK

=> Kiểm tra mã của WORKING FULL TIME ở cột Value trong Variable View. (mã 1)

Step 2: Lọc dữ liệu

- Data -> Select Cases -> If wrkstat =1 -> Continue -> OK


- Chạy lại bảng frequencies cho biến wrkstat

Step 3: Kiểm tra giả thuyết H0 với 774 quan sát

- Vào Analyze -> Compare Means -> One sample T-test


- Test Variable: Nhập biến cần kiểm định (hrs1)
- Test Value: Nhập giá trị của giả thuyết H0 (40)
- OK
c. Explain the difference between the standard deviation of your sample
and the standard error of the mean.

→ điều chỉnh HA là μ > 40

p-value ở cột 2-tailed nhỏ hơn alpha → H0 sai

d. If your sample size were doubled, how would you expect the value of
S.D to change? How would the value of the standard error of the mean
change? Estimate both the S.D and the standard error for a sample
twice as large.

https://www.sfu.ca/personal/archives/richards/Zen/Pages/Chap9.htm

có vẻ hơi tương tự nhg t chưa đọc kĩ, đang xem lại bài 1 đã

t = (x̄ - μ) / (s/√n)

● n tăng gấp đôi → s tăng thành s * căn 2 (check coi ) :)

σM = σ/√n

● n tăng gấp đôi → SE giảm thành căn SE / căn 2


e. What is the 95% CI for the average number of hours per week by full-
time workers? Give your conclusion on the hypothesis testing using the
95% CI.

k hiểu cái t mũ n-1 alpha/2 (sao l kí hiệu thôi m ơi, confidence degree của nó là n, ncl
m hiểu cả cái cụm đấy là 1.96 tức giá trị để 95%ại mũ nhỉ)

à thế đánh lại đi m ko cần đâu để ntn mn hiểu đc mà

f. Based on the 95% CI for the mean difference, can you reject the null
hypothesis that the average population value for hours worked is 43
hours? Explain.

chắc là dựa vào kết luận cái trên → reject

chuẩn (43 không trong khoảng chả reject)

Question 3

Use the gss.sav data file to perform the appropriate analyses to test whether the
average number of hours of daily television viewing (tvhours) is the same for men
and women. Include appropriate charts to illustrate your findings. Be sure to look at
the distribution of hours of television viewed separately for men and women.
1. Based on the results you observed, is it reasonable to conclude that
in the population, men and women watch the same amount of
television?

2. If you found a statistically significant difference between average


hours watched by men and women, would you necessarily conclude
that men and women do not watch the same amount of television?
What other nonstatistical explanations are possible for your findings?

You might also like