Excel Self-Practices for Seminar 3 & 4
This document provides you the necessary practices to prepare for the class
activities in Seminar 3 & 4. Please make sure that you are familiar with the practices
prior to the seminar.
Please note that the guides provided in this document are based on Microsoft Excel
2010 version. If you are using other versions, the steps are quite similar with only
minor variations. You may do an online search for similar operation for your version if
you are unsure.
Practice 2.1: Confidence Interval when σ Unknown
A manufacturer of batteries for portable hand tools wishes to investigate the length of
time a battery will last between charges at a fixed load. A sample of 12 batteries
was selected and the lifespan of the batteries are as follows:
4.15, 4.3, 4.25, 4.3, 4.3, 4.5, 4.5, 4.25, 4.45, 4.2, 4.1, 4.55
Construct a 95% confidence interval for the population mean.
Compute Confidence Interval
Enter the data in Excel.
Click Data Data Analysis Descriptive Statistics OK.
The following dialog box will appear. Follow the steps indicated below.
1. Highlight the data range.
5. Click OK.
2. Tick “Labels”,
“Summary 4. Click anywhere
Statistics” and in the worksheet to
“Confidence Level” indicate the top
left cell for the
output.
3. Enter the desired
Confidence Level.
The following results will appear.
Lifespan
Mean 4.320833
Standard Error 0.042399
Median 4.3
Mode 4.3 Upper Limit of CI:
Standard Deviation 0.146874 4.32 + 0.09332 = 4.4133
Sample Variance 0.021572
Kurtosis -1.12413
Lower Limit of CI:
Skewness 0.256023
Range 0.45 4.32 - 0.09332 = 4.2269
Minimum 4.1
Maximum 4.55
Sum 51.85
Count 12
Confidence Level(95.0%) 0.093319
Therefore, the 95% confidence interval for the population mean is
from 4.2269 to 4.4133.
Practice 2.2: One-Sample t-Test when σ Unknown
The Bunting Brass & Bronze Company has a computer controlled machine that is
programmed to do precision cutting of a circular brass disc with a mean diameter of
6.125 inches. The shop foreman takes a random sample of 8 discs from the
production line. The diameters are as follows:
6.115 6.127 6.129 6.113 6.124 6.121 6.131 6.124
The foreman suspects that the machine is out of adjustment. Use the hypothesis
testing procedure, at 0.05 level of significance, to determine if the programmer needs
to make adjustments.
Perform One-Sample t-Test for Mean
Unfortunately, there is no direct way to perform a one-sample t-test in Excel. However, Excel is able
to perform paired t-test. Therefore, we need to do a small “trick” in order to carry out a one-sample
t-test.
Enter the data into a new data table. Create a dummy column of zeros.
Click Data Data Analysis t-Test: Paired Two Sample for Means OK.
The following dialog box will appear. Follow the steps indicated below.
1. Highlight the data range.
6. Click OK.
2. Tick “Labels”.
3. Enter the 4. Click anywhere in the 5. Enter the
significant level. worksheet to indicate hypothesized mean.
the top left cell for the
output.
The following results will appear.
t-Test: Paired Two Sample for Means
Diameters Dummy
Mean 6.123 0
Variance 4.08571E-05 0
Observations 8 8
Pearson Correlation #DIV/0!
Hypothesized Mean Difference 6.125
df 7
-
t Stat 0.884995358
P(T<=t) one-tail 0.202765452
t Critical one-tail 1.894578605
P(T<=t) two-tail 0.405530903
t Critical two-tail 2.364624252
For this case, since we are performing a two-tailed test, the p-value is 0.4055 > 0.05.
Therefore, we do not reject H0. The foreman’s suspicion that the machine is out of
adjustment cannot be substantiated with this sample.
Note: If we are doing a one-tail test, the appropriate p-value is 0.2028.
Practice 2.3: Pooled t-Test
USIM has conducted a graduate survey recently. The salaries for ten accounting
major and eight general business majors students are shown here. At the 0.05
significance level can we conclude that accounting major earn more?
Accounting major
$33,000 $29,000 $31,000 $30,000 $32,000
$28,000 $32,000 $27,000 $28,000 $30,000
General Business major
$30,000 $31,500 $29,000 $29,500
$28,000 $29,500 $28,000 $26,500
Perform Two-Sample pooled t-Test for Difference of Two Means
Enter the data into a new data table according to the following format.
Click Data Data Analysis Scroll Down t-Test: Two-Sample Assuming Equal
Variances OK.
The following dialog box will appear. Follow the steps indicated below.
1. Highlight the data range.
6. Click OK.
2. Tick “Labels”.
3. Enter the 4. Click anywhere in the 5. Enter the hypothesized
significant level. mean difference.
worksheet to indicate
the top left cell for the
output. Note: The Mean Difference is
Mean Var1 – Mean Var 2.
The following results will appear.
t-Test: Two-Sample Assuming Equal Variances
Accounting General Biz
Mean 30000 29000
Variance 4000000 2285714.286
Observations 10 8
Pooled Variance 3250000
Hypothesized Mean Difference 0
df 16
t Stat 1.169410692
P(T<=t) one-tail 0.129682486
t Critical one-tail 1.745883676
P(T<=t) two-tail 0.259364971
t Critical two-tail 2.119905299
Since we are investigating if mean salary for Accounting major is more than that for
General Business, we are performing a right-tailed test. As such, the respective p-
values is 0.1297. Since p-value > 0.05, we do not reject H0 and cannot accept H1.
Therefore, we cannot conclude that accounting majors earn more than general
business major.
Practice 2.4: Paired t-Test
The Dean of the College of Business at Kingsport Fall Spring
University wants to determine if the Grade Point Student Semester Semester
Average (GPA) of business college students A 2.7 3.1
decreases during the last semester (spring B 3.4 3.3
semester) of their senior year. A sample of six C 3.5 3.3
students is selected. Their GPAs for the fall and D 3.0 2.9
E 2.1 1.8
spring semester of their senior year are shown.
F 2.7 2.4
At the 0.05 significance level, can the Dean conclude that the GPA of graduating
seniors declined during their last semester?
Let d= Mean GPA of fall semester - Mean GPA of spring semester
Since we want to explore whether grades decrease, a one-tailed test is appropriate.
H0: d 0
H1: d> 0
Perform a paired t-Test for Difference of Two Means
Enter the data into a new data table according to the following format.
Click Data Data Analysis Scroll Down t-Test: Paired Two Sample for Means
OK
The following dialog box will appear. Follow the steps indicated below.
1. Highlight the data range.
6. Click OK.
2. Tick “Labels”.
3. Enter the
4. Click anywhere in the 5. Enter the hypothesized mean.
significant level.
worksheet to indicate
the top left cell for the Note: The Mean Difference is
output. Mean Var1 – Mean Var 2.
The following results will appear.
t-Test: Paired Two Sample for Means
Fall
Semester Spring Semester
Mean 2.9 2.8
Variance 0.268 0.352
Observations 6 6
Pearson Correlation 0.89860815
Hypothesized Mean Difference 0
df 5
t Stat 0.939336437
P(T<=t) one-tail 0.195342008
t Critical one-tail 2.015048373
P(T<=t) two-tail 0.390684017
t Critical two-tail 2.570581836
Since we are investigating if mean GPA for Fall is more than Spring, we will perform
a right-tailed test. As such, the respective p-values is 0.1953. Since p-value > 0.05,
we do not reject H0. Therefore, we cannot conclude that the GPA of graduating
seniors has significantly declined during their last semester.
Practice 2.5: ANOVA Test for Difference of Means
Tiedke's Department Store accepts three types of credit Master
cards, MasterCard, Visa, and their own store card. The sales Card Visa Store
manager is interested in finding out whether there is a $61 $85 $61
difference in the mean amounts charged by customers on the 28 56 25
42 44 42
three cards. A random sample of 18 credit card purchases
33 72 31
(rounded to the nearest dollar) revealed these credit card
51 98 29
amounts. At the 0.05 significance level, can we conclude
56 56
there is a difference in the mean amounts charged per 72
purchase on the three cards?
There are three populations involved, the three credit cards. The null and alternate
hypotheses are:
H 0 : 1 2 3
H1: the means are not all equal
Perform ANOVA Test for Difference of Means
Enter the data into a new data table according to the following format.
Click Data Data Analysis Anova: Single OK
The following dialog box will appear. Follow the steps indicated below.
1. Highlight the data range.
6. Click OK.
2. Tick “Labels”.
3. Enter the
significant level. 4. Click anywhere in the
worksheet to indicate
the top left cell for the
output.
The following results will appear.
Anova: Single Factor
SUMMARY
Groups Count Sum Average Variance
MasterCard 6 271 45.16667 170.9667
Visa 7 483 69 346.3333
Store 5 188 37.6 210.8
ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 3337.967 2 1668.983 6.629907 0.008648 3.68232
Within Groups 3776.033 15 251.7356
Total 7114 17
The p-values is 0.0086. Since p-value < 0.05, we reject H0 and accept H1. Therefore,
we conclude that mean amounts charged by Tiedke's Department Store customers
are not the same for the three credit cards.
