FULL NAME: Yeji Kim

BSAN310
FALL 2022
Exam 1

INSTRUCTIONS
Write your name at the top of the page.

This exam has three (3) essay questions. The total points for the exam is 100 points.

Before beginning to write your answer, it is useful to step back and consider what your
main points will be and to establish an outline of your thoughts.

Ensure you answer the questions. With essay exams, students frequently will not answer
the question; they will veer off from the subject. Keep referring back to what the question
is asking and ensure your essay is focused on it. If there are multiple parts to the question,
ensure you thoroughly answer all parts of the question.

This is an open book/open notes exam.

Once you submit your exam in Canvas, please ensure it is actually submitted.

Section Exam 1 Open From: Wed Oct 5th –Fri Oct 7th (11:59pm)

EXAM INTENT: The intent of this exam is for you to demonstrate proficient command
of the subject matter in the course thus far. Your answers should show depth in your
ability to apply course content and concepts to practical examples. Moreover, your
answers should demonstrate a level of critical thinking demonstrating a strategic
viewpoint of business analytics and data topics/concepts leading you to logical
conclusions.

The Exam Begins on the Next Page


Page 1 of 3
1. After reading of failures within other businesses due to poor data quality, your boss is
concerned that data quality may be an issue for your organization.

Data quality is important for operating business. In a nutshell, data quality refers
to a data set's capacity to fulfill whatever purpose a corporation intends to utilize it for.
No matter how important data quality is, there are some ways that business
might get bad data. First, since people are fallible, even a small data collection with
manually entered data from humans is likely to include errors. Typographical mistakes,
data input in the incorrect field, missing entries, and other data entry errors are
essentially examples. Also, business commonly encounters the issue of not having all of
the data for each item while assembling a data set, which in short, the lack of complete
information.
Acknowledging data quality issues as unavoidable is the best way to approach
them. Even the best-run data operation cannot prevent the kinds of data problems
mentioned above. Thankfully, using data integration and data quality tools may help
business reduce the number of mistakes that are introduced during operations like data
conversions, then identify and automatically resolve any issues with data quality that do
occur.

2. Explain to your new CEO (Chief Enterprise Officer) the many ways business analytics
can benefit organizations. You are trying to convince him/her of the importance of
business analytics. Within your essay, ensure you discuss two (2) examples of
businesses that benefitted from business analytics.

Today, every organization generates a sizable amount of data in a particular way.

Business analytics are currently using statistical technology and methodologies to
evaluate historical data. Organization can use this to learn fresh information that may
aid in future strategic decisions.
Business analytics is important, because it can turn raw data into more useful inputs
so that decision-makers may use it. Also, organization must keep ahead of their
competitors and have access to the most up-to-date toolkits to help them make
decisions that will increase efficiency and revenue.
One major example of utilizing business is Tesla. Neural net software is used in each
Tesla vehicle to handle radar, sonar, and visual data. Additionally, they employ "fleet
learning," which allows automobiles to exchange data as they pass one another, to
enhance their overall analytics procedures. In essence, a fleet of automated cars
exchanging practical knowledge to keep drivers safe and to make tesla’s decision-driven
making procedure accurately.
The UK-based retail store Tesco is another intriguing business that uses data
analytics. Tesco introduced a new rewards program called Tesco Clubcard in 1995, long
before online retailers started gathering our data with the press of a mouse. The concept
was straightforward: each time a consumer made a purchase, they would produce their
Clubcard and receive points that they could then use in-store. Tesco received a record of
the transaction together with the name and zip code of the buyer in return. Tesco
discovered through data analytics that a small number of devoted clients made up much
of their revenue. They also determined how far customers would travel to see their

Page 2 of 3
establishments. Tesco developed customized discounts and offers based on this
information to influence consumer behavior, rewarding large spenders, and enticing
casual shoppers to interact more with the business.

3. As a student retention specialist at the University of Kansas, Chancellor Girod has

tasked you with analyzing various data that is stored at KU. He wants you to provide an
analysis of this data to provide predictions on student drop-out rates in the future.
You tell Chancellor Girod the project will require time for data wrangling /data
preparation. Write an essay to the Chancellor explaining why data wrangling/data
preparation will most likely take a lot of time to accomplish. Within your essay, explain
why data wrangling/data preparation is a necessary step within business analytics.

Dear Chancellor,

To analyze data and thus predicting on student future drop-out rates, data wrangling
is essential. To acquire, choose, and modify data to respond to an analytical inquiry is
the process of "data wrangling."
Analyzing and predicting can often be impacted by incomplete and erroneous data.
Data wrangling concentrates on removing unnecessary raw data to improve business
flow. Therefore, data wrangling is becoming more important as data grows more
unstructured, diversified, and dispersed. It expedites data processing and aids in
gaining insights more quickly. I may access high-quality data for analysis and other
downstream procedures with the help of data wrangling. Therefore, I think this project
will require time for data wrangling.
Also, according to an ancient proverb, "90% of data science is data wrangling”. Data is
more than just text and numbers, which is why it can be so challenging. To accomplish
data wrangling, I must deal with missing entries, mixed data types and ambiguous
values. First, completely a real-world data set, not every column or row will be filled in,
yet I still need to work with the data. Next, data entry users may include comments with
their data, which the I must separate and remove to work with the real data values.
Lastly, I cannot determine from the available data whether a number of 4/13 represents
a fraction, a month, and a day, or merely a string.
With these reasons, I require times for me to spend on data wrangling. This step will
allow me to analyze and get the final insights for you to predict on student drop-out
rates in the future.


Page 3 of 3