Module 4 | Case Study | Chapter 10

FIVE GUYS HAMBURGERS

Submitted by:
Team Stat Wizards:
Prachi Gehlot and Vidhi Desai

IN PARTIAL FULFILLMENT OF THE POST GRADUATE DEGREE

MS IN BUSINESS ANALYSTICS

Under the Supervision of

Dr. Leyla Ozsen
Professor, Decision Sciences

Submitted to:

Lam Family College of Business

Downtown Campus
160 Spear Street, Suite 1230
San Francisco, CA 94105

October 17, 2024

 Table of Contents

Executive Summary ........................................................................................................3

Background ...................................................................................................................3
Methodology ..................................................................................................................3
Part I: Regional Pricing Analysis ...................................................................................3
Overview of Methodology ........................................................................................3
Chosen Methodology ..............................................................................................4
Research Hypothesis ..............................................................................................4
Hypothesis Testing ..................................................................................................4
Part II: Shifts in Consumption Habits ...........................................................................7
Overview of Methodology ........................................................................................7
Chosen Methodology ..............................................................................................7
Research Hypothesis ..............................................................................................7
Hypothesis Testing ..................................................................................................8
Analysis and Findings ................................................................................................... 10
Conclusion .................................................................................................................. 11
Appendix 1: Minitab Output for Paired t-test .................................................................. 12

2
Executive Summary
The case study on Five Guys Hamburgers presents findings from two key analyses. First, it
investigates whether there is a significant difference in hamburger prices between San
Diego and Chicago, prompted by previous reports suggesting higher prices in San Diego.
The results reveal no statistically significant price difference between the two cities,
indicating that the observed price variations are likely due to random chance. Thus, if the
business aims to adjust prices for strategic reasons, it can do so without concern for
regional discrepancies. If the goal is to standardize prices across locations, no action is
required based on the current data.

The second analysis examines trends in beef consumption across the United States,
particularly over the past five years. This study finds a significant decline in beef
consumption, suggesting a shift in consumer preferences. Given that beef is a key
ingredient for the chain, the business may want to consider adapting its menu to better align
with evolving market demands, potentially opening up new growth opportunities.

Background
The fast-food industry is continually adapting to consumer preferences and market
conditions. The goal of this study is to examine two distinct aspects of the fast-food
industry:

1. Regional Pricing Analysis: The first part of the analysis compares hamburger
prices between San Diego and Chicago, using average prices from sampled
restaurants. Identifying regional price differences can help businesses refine their
pricing strategies.

2. Shifts in Consumption Habits: The second part of the study examines the
declining trend in beef consumption in the United States over recent years. This
trend is particularly important for businesses like Five Guys, which heavily features
beef-based products. By understanding these changes in consumer behavior,
companies can adjust their menus to better meet market demand and explore
future growth opportunities.

Methodology
Part I: Regional Pricing Analysis
Overview of Methodology
This part of the study aims to evaluate if there is a difference between the hamburger prices
between San Diego and Chicago, based on a publications’ indication that the price of a
hamburger in San Diego might be higher than in Chicago. Random samples were collected

3
from each of the cities. A sample of 15 restaurants in San Diego reveals an average
hamburger price of $8.60. Additionally, a random sample of 18 restaurants from Chicago
shows an average price of $8.41.

Chosen Methodology
Since it is given that prices of hamburgers in any given city are approximately normally
distributed, the difference in two means is to be examined and the population standard
deviation is known, the z-test is an appropriate method. To assess whether the difference
in prices is statistically significant, a z-score is calculated and compared with the critical z-
value determined at a 95% confidence interval.

(𝑥̅1 − 𝑥̅2 ) − (𝜇1 − 𝜇2 )

𝑧=
𝜎12 𝜎22
√
𝑛1 + 𝑛2

Research Hypothesis
The case study seeks to investigate whether there is a difference in the average price of a
hamburger between San Diego and Chicago at the Five Guys Hamburger chain, hence, a
two-tailed test will be conducted.

The null and alternate hypothesis are:

Null Hypothesis (𝑯𝟎 ): There is no difference in average price of hamburgers in the two cities
Alternate Hypothesis (𝑯𝒂): There is a significant difference in average price of hamburgers
in the two cities
𝐻0 : 𝜇1 − 𝜇2 = 0
𝐻𝑎 : 𝜇1 − 𝜇2 ≠ 0

Hypothesis Testing
• Step 1: Establish the hypothesis
In this analysis, data from San Diego represents population 1, while data from
Chicago represents population 2. To determine whether there is a difference in the
average hamburger pricing between the two cities, the case study employs a two-
tailed test. The hypotheses are defined as follows:
𝐻0 : 𝜇1 − 𝜇2 = 0
𝐻𝑎 : 𝜇1 − 𝜇2 ≠ 0

• Step 2: Determine the appropriate Test statistic:

As mentioned above, the conditions are satisfactory to proceed with a z-statistic to
determine the difference in mean prices of population 1 & population 2.

4
 (𝑥̅1 − 𝑥̅2 ) − (𝜇1 − 𝜇2 )
𝑧=
𝜎2 𝜎2
√ 1+ 2
𝑛1 𝑛2
Where,
𝜇1 = the mean of population 1
𝜇2 = the mean of population 2
𝑛1 = size of sample 1
𝑛2 = size of sample 2
𝜎12 = the variance of population 1
𝜎22 = the variance of population 2

• Step 3: Determine the Significance level:

The confidence level is set to 95%, therefore the Type I error rate, or alpha, will be
0.05.

α = 0.05

• Step 4: Establish the decision rule:

Since the test is a two tailed test, the rejection area of 0.05 is split between both the
ends of the curve, with each end measuring to 0.025. Thus, the critical value of z-
score and critical p-value can be calculated as mentioned here:

For 95% confidence, 𝑧0.025 = ± 1.96 using Excel: = abs(NORM.S.INV (0.025))

Critical p-value = α = 0.05

Decision Rules:
o The non-rejection region for the z-value is defined as -1.96 < z < 1.96. This means
that if the observed z-value is less than -1.96 (in the lower tail) or greater than
1.96 (in the upper tail), the null hypothesis will be rejected. Conversely, if the
observed z-value falls within this range, the decision will be to fail to reject the
null hypothesis, as it lies within the non-rejection region.
o Since this is a two-tailed test with a significance level of 0.05, 2.5% of the
distribution's area lies in each tail. This creates the rejection region at both ends
of the distribution, with 95% of the area making up the non-rejection region.
Therefore, if the observed p-value, which is calculated for the entire distribution,
is less than 0.05, the null hypothesis will be rejected, indicating that the
observed result is statistically significant.

• Step 5: Collect and summarize the data:

A sample of 15 restaurants in San Diego reveals an average hamburger price of
$8.60. Additionally, a random sample of 18 restaurants from Chicago shows an

5
 average price of $8.41. The population standard deviation(σ) is known and is $0.64.
The data can be summarized as below:

San Diego Chicago

(Sample 1) (Sample 2)
n1= 15 n2= 18
̅𝟏 = $8.60
𝒙 ̅𝟐 = $8.41
𝒙
𝝈𝟏 = $0.64 𝝈𝟐 = $0.64

• Step 6: Compute the Test Statistic:

Using the given data values and considering the difference in population means
(null hypothesis) as zero, z statistic van be calculated as follows:
(8⋅60−8⋅41)−(0)
z= 2 2
√0⋅64 +0.64
15 18

z = 0.85 (Rounded off to two decimal places)

Based on the calculated z-value, the corresponding p-value will be obtained to
determine whether to reject or fail to reject the null hypothesis:
p-value = 0.20 using Excel: =1- NORM.S.DIST(z,1)

• Step 7: Reach a Statistical Conclusion

Observed z = 0.85 Observed p-value = 0.20

Critical z value = ± 1.96 Critical p- value α = 0.05

Since the test statistic, z= 0.85 lies in the non-rejection region of the normal
distribution curve, and p-value of 0.20 is greater than 0.05, the conclusion is to fail
to reject the null hypothesis. Thus, it can be said with 95% confidence that there
is not enough evidence to infer that there is a significant difference in the average
price of hamburger between Five Guys restaurants in San Diego and Chicago.

6
 • Step 8: Make a business decision:
Although indications from Burger Week suggest a potential average price difference
for hamburgers, the current sample data does not provide sufficient evidence to
support the indication. There might be other factors affecting the pricing decision
such as local market conditions, operational costs and promotional offers, which
must be investigated to achieve significant results to gain a clearer understanding
of the average hamburger price differences between San Diego and Chicago,
thereby informing more effective pricing strategies.

Part II: Shifts in Consumption Habits

Overview of Methodology
As noted by BeefProducer.com, beef consumption in the United States has been declining
gradually, relatively consistently. To verify this trend, beef consumption was measured in a
randomly selected group of 18 American adults six years ago (indicated as Early year), and
five years later (indicated as late year), the consumption of another closely matched group
of 18 adults was measured to observe changes over time. The beef consumption was
measured in pounds in two different time periods.

Chosen Methodology
The analysis utilizes closely matching individuals across two time periods, indicating that
the samples are related. Each consumption amount reported by an American adult at an
earlier stage has a corresponding consumption data by an adult surveyed at later stage.

As the data are related, the population variance is unknown and the data is assumed to be
normally distributed, the paired t-test is appropriate methodology for the analysis. This t
test for dependent measures uses the sample difference, d, between individual matched
sample values as the basic measurement of analysis instead of individual sample values.
Analysis of the d values effectively converts the problem from a two-sample problem to a
single sample of differences, which is an adaptation of the single-sample means formula.
To assess whether there is a significant change in beef consumption over time, a t-score is
calculated and compared with the critical t-value determined at a 95% confidence interval.

Research Hypothesis
Because there is a need to determine if there is a significant change in the beef
consumption over time (Early year and Late year), a two-tailed test is being conducted.

The null and alternate hypotheses are:

Null Hypothesis (𝑯𝟎 ): There is no difference in beef consumption over time

Alternative Hypothesis (𝑯𝒂): There is a significant change in beef consumption over time

7
 𝐻0 : D = 0
𝐻𝑎 : D ≠ 0
Where, D = mean population difference

Hypothesis Testing
The desired effect is a significant change in beef consumption over time, which means
that the hypothesis test is two-tailed.

• Step 1: Establish the hypothesis

To determine whether there is a significant change in the beef consumption over
time, the case study employs a two-tailed test. The hypotheses are defined as
follows:
𝐻0 : D = 0
𝐻𝑎 : D ≠ 0

• Step 2: Determine the appropriate Test statistic:

As mentioned above, the conditions are satisfactory to proceed with a t-statistic.
Following is the t-statistic Formula to test the difference in two dependent
populations:
𝑑̅ − 𝐷
𝑡= 𝑠
𝑑
√𝑛
Degrees of freedom (df) = n-1
Where,
n = number of pairs of American adults (sample size)
𝑑̅= mean of the differences between paired samples
D = mean population difference
sd = standard deviation of the differences between paired samples

• Step 3: Determine the Significance level:

As specified in the problem, the confidence level is set to 95%, therefore the Type I
error rate, or alpha, will be 0.05.
α = 0.05

• Step 4: Establish the decision rule:

Since the test is a two tailed test, the rejection area of 0.05 is split between both the
ends of the curve, with each end measuring to 0.025. Thus, the critical value of z-
score and critical p-value can be calculated as mentioned here:

df = n – 1 = 18 − 1 = 17
For 95% confidence, t0.025,17 = ± 2.11 using Excel: =abs(T.INV(0.025,17))
Critical p-value = α = 0.05

8
 Decision Rule:
o The non-rejection region for the t-value is defined as -2.11 < t < 2.11. This means
that if the observed t-value is less than -2.11 (in the lower tail) or greater than
2.11 (in the upper tail), the null hypothesis will be rejected. Conversely, if the
observed t-value falls within this range, the decision will be to fail to reject the
null hypothesis, as it lies within the non-rejection region.
o Since this is a two-tailed test with a significance level of 0.05, 2.5% of the
distribution's area lies in each tail. This creates the rejection region at both ends
of the distribution, with 95% of the area making up the non-rejection region.
Therefore, if the observed p-value, which is calculated for the entire distribution,
is less than 0.05, the null hypothesis will be rejected, indicating that there is a
significant change in beef consumption over time.

• Step 5: Collect and summarize the data:

In this study, beef consumption was assessed in a randomly selected group of 18
American adults six years ago, referred to as the "Early Year," with an average
consumption of 59.58 pounds per individual. Five years later, a closely matched
group of 18 adults was surveyed to observe changes over time, yielding an average
consumption of 58.60 pounds in the "Late Year." The sample standard deviation for
both groups is 11.58 pounds. Additionally, the mean and standard deviation of the
differences between the paired samples were calculated, resulting in a mean
difference of 0.986 pounds and a standard deviation of 1.619 pounds. This data will
facilitate a comparative analysis of beef consumption trends over the specified time
periods. The given sample data can be summarized as follows:

Early Year Late Year

Difference
(Sample 1) (Sample 2)
n1= 18 n2= 18
̅𝟏 = 59.58
𝒙 ̅𝟐 = 58.60
𝒙 ̅ = 0.986
𝒅
𝒔𝟏 = 11.58 𝒔𝟐 = 11.58 𝒔𝒅 = 1.619

• Step 6: Compute the Test Statistic:

The observed value of the t-test statistic and the p-value are given in the case, which
are as mentioned bellow (Appendix 1):

Observed t = 2.59
Observed p-value = 0.019

• Step 7: Reach a Statistical Conclusion

Observed t = 2.59 Observed p-value = 0.019

Critical t value = ± 2.11 Critical p- value α = 0.05

9
 Since the test statistic, t= 2.59 lies in the rejection region of the normal distribution
curve in the right tail, and p-value of 0.019 is lesser than 0.05, the conclusion is to
reject the null hypothesis. Thus, it can be said with 95% confidence that there is a
significant change in beef consumption pattern over time.

• Step 8: Make a business decision:

The decline in beef consumption suggests a shift in dietary trends, which fast-food
chains should consider when developing their menus.

Analysis and Findings

The analysis of hamburger prices reveals that the null hypothesis, which posits no
significant difference in the average price of hamburgers between San Diego and Chicago
at Five Guys restaurants, was not rejected, indicating there is insufficient evidence to
recommend changes to the pricing strategy for the restaurants in San Diego and Chicago.
Therefore, the analysts conclude with 95% confidence that the average prices lie between
$8.60 and $8.41, indicating no significant price difference between the two cities. To
enhance the robustness of future analyses, several potential factors influencing pricing,
including local economic conditions, market competition, consumer preferences,
manufacturing and operating costs, and promotional strategies tailored to each location
must also be considered.

The results of the paired t-test reveal a significant decrease in beef consumption over the
six-year period. The mean reduction of 0.986 pounds, with a 95% confidence interval
ranging from 0.181 to 1.791 pounds, indicates that this decline is unlikely to be attributed
to random variation. This outcome aligns with broader trends reported by
BeefProducer.com, further supporting the observed reduction in beef consumption.

10
However, there are two key limitations to consider in this hypothesis testing:

1. Small Sample Size: Both the analyses are based on data from only 15 and 18 samples,
which is too small to confidently generalize the results to the entire population of
American stores or American adults.

2. Assumption of Normality: The analysis assumes the data follows a normal distribution,
which may not be accurate. This assumption can limit the generalizability of the findings if
the data does not meet this condition.

These limitations should be considered when interpreting the results, as they may impact
the strength and applicability of the conclusions drawn from the analysis.

Conclusion
Based on the analysis of the given sample and population parameters, the average prices
of hamburgers at Five Guys restaurants revealed no significant differences between the
locations in San Diego and Chicago. Although the analysts found no meaningful variance in
pricing, they recommend conducting a more in-depth investigation to explore potential
underlying factors that could contribute to price differences, if they exist. Such factors may
include varying economic conditions, local competition, consumer preferences, operating
costs, and promotional strategies aimed at boosting sales. Since the current findings
indicate no price differences, the analysts advise against revising the pricing policies for
Five Guys in these two cities if the business aims to maintain consistency. However, if the
company seeks to modify its pricing strategy, it may consider adjusting prices in either city
to achieve its business objectives.

In addition to pricing analysis, the study confirms a significant decline in beef consumption
among American adults over the past six years, which has critical implications for Five
Guys’ future business strategy. This trend suggests that the company may need to adjust its
menu offerings, potentially focusing on alternative protein options or non-beef products to
align with evolving consumer preferences and drive future growth. Maintaining a
commitment to high-quality, fresh ingredients will be essential for preserving customer
loyalty. Furthermore, Five Guys should consider reevaluating its marketing strategies to
promote both traditional offerings that appeal to loyal customers and new products that
resonate with emerging consumer trends. By adapting to these changes, Five Guys can
better position itself to remain competitive in the marketplace.

11
Appendix 1: Minitab Output for Paired t-test

12