Scribd
Upload
28 views10 pages

One-Sample t-Test for Coca-Cola Data

This document outlines an in-class activity focused on conducting a one-sample hypothesis test for means, specifically a one-sample t-test, using a dataset related to plastic items from The Coca-Cola Company. Students will learn to calculate test statistics, P-values, and interpret results in context, while also preparing for future activities comparing two populations. The activity includes group work, individual calculations, and discussions to reinforce understanding of hypothesis testing steps.

Uploaded by

serkan özel
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
28 views10 pages

One-Sample t-Test for Coca-Cola Data

This document outlines an in-class activity focused on conducting a one-sample hypothesis test for means, specifically a one-sample t-test, using a dataset related to plastic items from The Coca-Cola Company. Students will learn to calculate test statistics, P-values, and interpret results in context, while also preparing for future activities comparing two populations. The activity includes group work, individual calculations, and discussions to reinforce understanding of hypothesis testing steps.

Uploaded by

serkan özel
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Introductory Statistics First Edition (2021)

Instructor Edition

In-Class Activity 13.B


One-Sample Hypothesis Test for Means
Overview and student objectives
In-Class Activity Length: 25 minutes
Prior In-Class Activity: In-Class Activity 13.A, “Introduction to Hypothesis Testing for
Means”
Next In-Class Activity: In-Class Activity 13.C, “Comparing Two Populations” (25
minutes)
Constructive Perseverance Level: 3
Content Learning Outcomes: DP.3.D, DP.4.B
DCMP Learning Goals: Problem Solving, Reasoning, Technology

Overview
In this in-class activity, students will learn how to conduct and interpret the results from
a one-sample hypothesis test for means using the t Distribution, commonly referred to
as a one-sample t-test. Students will use the “plastics” dataset from the previous in-
class activity (13.A) to conduct the test about the mean number of plastic items from
The Coca-Cola Company products found in communities in 2020. Students will
calculate the value of the test statistic and use the DCMP Data Analysis Tools to
calculate the P-value and interpret the results. The activity concludes with students
working in groups to conduct another one-sample t-test for another company from the
dataset.

Objectives
Students will understand:
• The steps of hypothesis testing can be applied to a one-sample hypothesis test
for means, also known as a one-sample t-test.
Students will be able to:
• Calculate the value of a test statistic for a one-sample t-test.
• Calculate a P-value for a one-sample t-test.
• Interpret the results of a one-sample t-test in context.
Suggested resources and preparation
Materials and technology
• Computer, projector, document camera
• Preview Assignment 13.B
• Student Pages for In-Class Activity 13.B

Copyright © 2021, The Charles A. Dana Center at The University of Texas at Austin
Introductory Statistics First Edition (2021)
Instructor Edition

• Practice Assignment 13.B


• Access to the DCMP Inference for a Population Mean tool
• “plastics” dataset
Prerequisite assumptions
Students should be able to:
• Identify the null and alternative hypotheses for a one-sample test for means.
• Identify that they collect sample data after stating the null and alternative
hypotheses.
• Identify that they check test assumptions after collecting sample data.
• Identify that they summarize the data using a test statistic after checking test
assumptions.
• Identify that they calculate a P-value after calculating the value of the test
statistic.
• Identify that they interpret the test results in context after calculating the P-value.
Making connections
This activity:
• Connects back to one-sample tests for proportions.
• Connects forward to comparing two sample means.
Background context
None
Suggested instructional plan
Frame the activity (3 minutes)
Resources Instructor Suggestions
and Structure
Think-Pair- Question 1
Share
• Have students answer Question 1 individually, and then have them
share their answers with elbow partners.
• Briefly discuss answers as a class. Use student answers to let
students know that they will continue exploring the number of plastic
products found in waste and the corporations those plastic products
came from.

Group Work • Transition to the in-class activity by briefly discussing the Objectives
for the activity.

Copyright © 2021, The Charles A. Dana Center at The University of Texas at Austin
Introductory Statistics First Edition (2021)
Instructor Edition

Activity flow (20 minutes)

Resources Instructor Suggestions


and Structure
• If necessary, reorient the students to the “plastics” dataset that was
used in In-Class Activity 13.A. This is the same dataset that they will
be using in this activity.
o Spreadsheet link:
[Link]
mi15bH87dUIz8X1o60WcsukoLlFFo/edit?usp=sharing
o An Excel version of the spreadsheet can be found in the
spreadsheet DCMP_STAT_13A_Plastics.
• The filtered Coca-Cola 2020 spreadsheet can be found in the
spreadsheet DCMP_STAT_13A_Plastics_Coca_Cola.
• If necessary, introduce students to the layout of the DCMP Inference
for a Population Mean tool at
[Link]
• For this activity, students will need to select “Significance Test” under
“Type of Inference.”
Pairs • Have students answer Question 2 in pairs to re-orient themselves to
the data.
• Students will need to remember the research question from In-Class
Activity 13.A for the one-sample test. If students don’t remember,
you can provide them with the following research question:
o “For the products reported from The Coca-Cola Company, is
there evidence that the mean total plastics count found in
various countries in 2020 differs from a claimed value of 275
items?”

• The distribution of the t-statistic was talked about in In-Class Activity


12.B.
• If necessary, walk through the test statistic for a one-sample
hypothesis test for means with students:
𝑥̅ − 𝜇!
𝑡=
𝑠/√𝑛
• You may want to display the components of the test statistic:
𝑥̅ = sample mean
𝜇! = hypothesized population mean (from the null hypothesis)

Copyright © 2021, The Charles A. Dana Center at The University of Texas at Austin
Introductory Statistics First Edition (2021)
Instructor Edition

𝑠 = sample standard deviation


𝑛 = sample size

Pairs • Have students work together in pairs to answer Questions 3 and 4


about calculating the test statistic by hand.
Individually • Have students answer Questions 5 and 6 on their own and then
have them compare their answers with elbow partners.
Pairs • Have students work in pairs to answer Questions 7–10 about
conducting another one-sample mean test using a different company
in the dataset. If short on time, assign this section for homework.
• Students can decide whether to use just data for 2019 or 2020.
• The instructions for filtering the dataset are included in the instructor
resources for In-Class Activity 13.A. If the student struggle with
filtering the data is unproductive, you can use the PepsiCo 2019
filtered dataset found in spreadsheet DCMP_STAT_13B_PepsiCo.

Wrap-up/transition (2 minutes)

Resources Instructor Suggestions


and Structure
Wrap-up • Have students share the interpretations for the hypothesis test(s)
they completed in class and briefly discuss.
• Have students refer back to the Objectives for the activity and
check the ones they recognize. Alternatively, they may check the
objectives throughout the activity.
Transition • Preview the next activity, which will be about comparing two
populations using a two-sample hypothesis test.

Suggested assessment, assignments, and reflections


• Give Practice Assignment 13.B.
• Give the preview assignments, if any, for the activities you plan to complete in
the next class meeting.

Copyright © 2021, The Charles A. Dana Center at The University of Texas at Austin
Introductory Statistics First Edition (2021)
Instructor Edition

In-Class Activity 13.B One-Sample Hypothesis Test for


Means ANSWERS
1) How many plastic products do you think
you use each week? How do you think
this compares to your neighbors?
Answers will vary.

Credit: iStock/piotr_malczyk

Objectives for the activity


You will understand:
¨ The steps of hypothesis testing can be applied to a one-sample hypothesis test for
means, also known as a one-sample t-test.
You will be able to:
¨ Calculate the value of a test statistic for a one-sample t-test.
¨ Calculate a P-value for a one-sample t-test.
¨ Interpret the results of a one-sample t-test in context.

In this in-class activity, you will be revisiting the “plastics” dataset used in In-Class
Activity 13.A. Recall that this dataset is a sample of plastic products collected by
community volunteers in countries around the world during a brand audit to see what
types of plastics are found in waste and from which companies the plastics found came
from.
The dataset can be accessed at
[Link]
ukoLlFFo/edit?usp=sharing.

2) In In-Class Activity 13.A, you used a subset of the “plastics” dataset that contains
counts of plastics from The Coca-Cola Company in 2020.

Part A: What was the research question you were trying to answer using the Coca-
Cola subset of data?
Answer: For the products reported from The Coca-Cola Company, is there evidence
that the mean total plastics count found in various countries in 2020 differs from a
claimed value of 275 items?

Copyright © 2021, The Charles A. Dana Center at The University of Texas at Austin
Introductory Statistics First Edition (2021)
Instructor Edition

Part B: What were the null and alternative hypotheses you wrote to answer the
research question?
Answer: 𝐻! (null): µ = 275; 𝐻" (alternative): µ ≠ 275

Part C: Is this a one-tailed or two-tailed one-sample test?


Answer: This is a two-tailed test.

Part D: What is the sample mean count of all plastic products reported in 2020 that
are from The Coca-Cola Company?
Answer: The sample mean count of all plastic products reported in 2020 that are
from The Coca-Cola Company is 276.7 items.

Part E: What is the sample standard deviation of all plastic products reported in 2020
from The Coca-Cola Company in the subset?
Answer: The sample standard deviation is 726.3 for all types of plastic in 2020 from
The Coca-Cola Company.

3) Calculate the value of the test statistic by hand using the formula introduced in In-
Class Activity 12.B:
𝑥̅ − 𝜇!
𝑡=
𝑠/√𝑛

#$%.$'#$(
Answer: Test statistic: 𝑡 = $#%.)/√(!
= 0.01655

4) Let’s say that we got a new sample of the total plastics count for products from The
Coca-Cola Company in the sample of countries around the world. Would the value
of the test statistic that we calculated using this new sample change from the test
statistic value we calculated in Question 3? Explain.
Answer: Yes, the test statistic would change with this new sample dataset because
the sample mean (𝑥̅ ) and the sample standard deviation (𝑠) would be different. Also,
the sample size might be different as well, depending on the new sample.

5) Use the DCMP Inference for a Population Mean tool at


[Link] to calculate the P-value
with a significance level of 0.05.

Copyright © 2021, The Charles A. Dana Center at The University of Texas at Austin
Introductory Statistics First Edition (2021)
Instructor Edition

Sample output:

Part A: What is null value you entered into the data analysis tool?
Answer: 275

Part B: What is the P-value?


Answer: The P-value is 0.9869.

Part C: Does the value of the test statistic from the data analysis tool match your
answer to Question 3?
Answer: Yes, the test statistic in the data analysis tool is 0.0166.

6) How would you interpret the result of this test in context? Is there evidence to
suggest that the claim by Coca-Cola is false?
Answer: I conclude that the P-value is larger than 0.05, so I fail to reject the null
hypothesis. There is not enough evidence to conclude that the mean total count of
plastics from The Coca-Cola Company found in various countries in 2020 is different
from 275 items.

The following is a list of companies that have a large enough sample size to meet the
conditions for a hypothesis test:

Copyright © 2021, The Charles A. Dana Center at The University of Texas at Austin
Introductory Statistics First Edition (2021)
Instructor Edition

2019 2020
Nestle Nestle
PepsiCo PepsiCo
Unilever
Mondelez International
Mars, Incorporated

Choose a company from the list. Create a subset of data for the company you chose by
filtering by the parent_company variable and the year. This is the subset of data you will
use to answer Questions 7–10.
In Questions 7–10, use a significance level of 0.05 and a hypothesized value of 90
items. The variable that you will be focused on is the total plastics count found in
various countries.

7) Which company did you choose?


Answers will vary.
Sample answer: PepsiCo, 2019

8) What is your research question? What are the null and alternative hypotheses?
Answers will vary.
Sample answer: Is there sufficient evidence to conclude that the average total
plastics count from PepsiCo in 2019 is different from 90?
𝐻! (null): µ = 90
𝐻" (alternative): µ ≠ 90

9) Calculate the test statistic and P-value using the DCMP Inference for a Population
Mean tool.

[Continued on the next page.]

Copyright © 2021, The Charles A. Dana Center at The University of Texas at Austin
Introductory Statistics First Edition (2021)
Instructor Edition

Sample output:

Answers will vary.


Sample answer:
Test statistic = 0.4383
P-value = 0.6644

10) How would you interpret the result of the test in context?
Answers will vary.
Sample answer: There is not sufficient evidence to conclude that the average
plastics count for PepsiCo in 2019 is different from 90.

Copyright © 2021, The Charles A. Dana Center at The University of Texas at Austin
Introductory Statistics First Edition (2021)
Instructor Edition

Copyright © 2021, The Charles A. Dana Center at The University of Texas at Austin

Common questions

Powered by AI

In a one-sample t-test, a one-tailed test examines if the sample mean is significantly greater than or less than a hypothesized value, which implies directional inference. In contrast, a two-tailed test assesses if the sample mean is significantly different from a hypothesized value in either direction, implying non-directional inference. The choice affects the hypothesis structure and the critical regions for testing; the entire significance level is in one tail for a one-tailed test, while it is split between both tails for a two-tailed test, resulting in different critical values for rejection regions .

Technological tools like the DCMP Data Analysis Tools facilitate hypothesis testing by automating computations, reducing human errors, and providing user-friendly interfaces for complex statistical analyses. They streamline the calculation of test statistics and P-values, thereby saving time and improving accuracy. Additionally, these tools often include data visualization features, making it easier to interpret results, and facilitate reproducibility of analyses for educational and research purposes. Users can focus more on interpreting results and less on computational complexities .

Interpreting the P-value in the context of a one-sample t-test is crucial because it provides the basis for decision-making regarding the null hypothesis. A P-value indicates the probability of observing a test statistic as extreme as, or more extreme than, the one observed, under the assumption that the null hypothesis is true. A smaller P-value suggests stronger evidence against the null hypothesis. Contextual interpretation helps to communicate the significance of this statistical measure in relation to the specific research question being investigated, such as identifying if there is evidence that the mean number of plastic items differs from a claimed value .

The steps involved in conducting a one-sample hypothesis test for means using a t-test are as follows: 1) State the null and alternative hypotheses. 2) Collect sample data. 3) Check test assumptions, such as normality and sample size. 4) Summarize the data using a test statistic calculated from the formula t = (x̄ - µ)/(s/√n), where x̄ is the sample mean, µ is the hypothesized population mean, s is the sample standard deviation, and n is the sample size . 5) Calculate the P-value after determining the test statistic. 6) Make an interpretative conclusion in the context of the hypothesis test .

Evidence from a hypothesis test is not always definitive due to several factors: statistical hypothesis tests are probabilistic rather than deterministic. There is always a possibility of Type I (false positive) or Type II (false negative) errors, and findings are influenced by the chosen significance level, sample size, and assumptions about data. The results indicate the strength of evidence against the null hypothesis but do not prove or disprove the hypothesis. Statistical testing merely provides a framework for making decisions with a certain level of confidence .

A new sample affects the calculation of a test statistic because it leads to different sample metrics such as the sample mean (x̄), the sample standard deviation (s), and potentially the sample size (n). Changes in these values will alter the test statistic, which is computed as t = (x̄ - µ) / (s/√n). Therefore, each alteration in sample data leads to a recalculated test statistic .

The hypothesized population mean (µ) acts as a benchmark or reference value against which the sample mean (x̄) is compared. It is integral in the calculation of the test statistic, where the formula used is t = (x̄ - µ)/(s/√n). The difference between the sample mean and the hypothesized population mean, divided by the standard error of the mean, reveals how many standard errors the sample mean is away from the hypothesized mean. This comparison is essential for determining the likelihood of the null hypothesis being true .

Failing to reject the null hypothesis in a one-sample t-test implies that there is insufficient statistical evidence to conclude that the observed sample mean is significantly different from the hypothesized population mean. It does not prove the null hypothesis true; rather, it suggests that any observed difference could be due to sampling variability or random chance. This result may prompt further research or adjustments in study design, such as increasing the sample size for more power, or reassessing the assumptions made during the test .

To evaluate if the sample size is sufficient for conducting a one-sample t-test, one must consider factors such as the test assumptions regarding normality and the Central Limit Theorem, which states that the distribution of the sample mean approaches a normal distribution as the sample size increases. If the sample size is large enough (commonly n > 30 is used as a rule of thumb), the sampling distribution of the mean is approximately normal, justifying the use of a t-test. Additionally, ensuring that the sample data adequately represents the population and providing sufficient data points to detect a meaningful effect or difference are key evaluative measures .

The significance level, often denoted as alpha (α), represents the threshold for determining whether to reject the null hypothesis. It is the probability of committing a Type I error, which is rejecting a true null hypothesis. Commonly set at 0.05, this level dictates the cutoff point for the P-value; if the P-value is less than or equal to α, the null hypothesis is rejected, indicating statistically significant results. This impacts the conclusion as it helps determine if the observed effect is likely due to chance or if there is enough evidence to suggest a real effect or difference exists .

You might also like