Scribd
Upload
10 views

Lesson 19

This document provides instructions on how to perform a sign test in Minitab to analyze whether data comes from a population with a median greater than a specified value. It includes an example that analyzes credit card charges, finds a p-value of 0.8062 which is greater than 0.01 so it fails to reject the null hypothesis that the median is greater than $300. It also provides a practice problem analyzing class sizes and asks the student to conclude if the median is greater than 25 based on a sample of data.

Uploaded by

zaenalkmi
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
10 views

Lesson 19

This document provides instructions on how to perform a sign test in Minitab to analyze whether data comes from a population with a median greater than a specified value. It includes an example that analyzes credit card charges, finds a p-value of 0.8062 which is greater than 0.01 so it fails to reject the null hypothesis that the median is greater than $300. It also provides a practice problem analyzing class sizes and asks the student to conclude if the median is greater than 25 based on a sample of data.

Uploaded by

zaenalkmi
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Return to Cover Page

LESSON 19 - SIGN TEST


In this lesson we will learn how to use Minitab to do a sign test. We will use Problem 3
on page 618 as an example. We will use the P-value method, but since the text tells us to use a
1% level of significance, we will write a decision and conclusion.
Open Minitab and clear everything below the date/time stamp. Type your name, Lesson
19 and Example on separate lines. Define the variable, X = amount of a new credit card charge
for the previous month. Enter the data into C1, label the column and display the data. Now click
on Stat > Nonparametrics > 1-Sample Sign. Select C1 into the "Variables:" box, click on the
button for "Test median", type 300 into that box, then change the "Alternative:" to "greater than"
and click "OK". The results, along with the decision and conclusion that we must write, are
shown in the figure below.

————— 8/12/2009 8:16:57 PM ————————————————————

Jeonghun Kim
Lesson 19
Example

X = amount of a new credit card charge for the previous month

Data Display

Charge
346.71 382.59 255.03 202.17 309.80 265.88 299.41 270.83
296.54 316.46 245.92 309.47

Sign Test for Median: Charge

Sign test of median = 300.0 versus > 300.0

N Below Equal Above P Median


Charge 12 7 0 5 0.8062 298.0

Decision: Since the P-value = 0.8062 > 0.01, we fail to reject H0.

Conclusion: There is insufficient evidence at the 1% level of significance to


support the claim that the median amount of new credit card charges
for the previous month was more than $300.

Notice that if we had chosen to use the classical approach, the number of plus signs
would be given by what Minitab has labeled "Above", 5 in this case. If we had chosen to use the
classical method, our critical value for the number of +'s would have been 1 (see Table 8 in our
text).
84
MINITAB ASSIGNMENT 19

See instructions on page 8.

1. Podunk University claims that their median class size is no greater than 25. A group of
students believe that the median is greater than that. The data below represents class sizes of a
random sample of 15 classes at PU. Enter and display this data. Using the P-value method and a
10% level of significance, what conclusion can the students draw from this data?

32 19 26 25 28 21 29 22 27 28 26 23 26 28 29

2. Do Problem 6 on page 618. Enter and display the data. Using the P-value method, what
conclusion can the meteorologist draw from this data? Type your answer in the session window.

Return to Cover Page

85

You might also like