Because learning changes everything.

MIS
Data Analytics and Data
Ecosystems

© McGraw Hill LLC. All rights reserved. No reproduction or distribution without the prior written consent of McGraw Hill LLC.
 Overview of Data Analytics and
Data Ecosystems
Data analytics
The process of investigating raw data
of various types to:
• Uncover trends and correlations.
• To answer specifically crafted questions.
The process of data analytics
includes.
• Descriptive analytics.
• Predictive analytics.
• Prescriptive analytics.

© McGraw Hill LLC andreypopov/123RF 2

 Descriptive Analytics
Uncovers historical trends in data sets
Can be thought of as trying to answer the questions:
• “what happened?” or
• “what is occurring?”
Examples:
• Return on Investment (ROI).
• Summaries of past events such as sales, operational efficiency,
impact of sales and marketing.

© McGraw Hill LLC 3

 Predictive Analytics
Focuses on understanding, predicting, and planning for
future events and business outcomes.
Utilize probability analysis techniques as well as data
mining, statistical modeling, machine learning, and
deep learning to generate possible future outcomes
given certain conditions.
Can be used in a variety of business areas including:
• E-commerce.
• Cybersecurity.
• Information technology.
• Healthcare.
© McGraw Hill LLC 4
 Prescriptive Analytics
Used to help determine best courses of action.
Considered to be the most advanced form of data
analytics.
Seek to predict what, when, and why a given scenario
might occur.
Examples:
• Tracking fuel prices by airlines to determine possible increases
and decreases.
• Monitoring flu strains and activity to determine possible
outbreak areas.
• Analysis of cyberthreats and activity to identify possible threats
© McGraw Hill LLC 5
 Trends in Data Analytics
Smarter, scalable artificial intelligence (AI).
• Allows for better learning algorithms.
• Allows for shorter development times.
Composable data and analytics.
• Creates a more user-friendly experience.
• Makes finding critical insights easier.
• Enables collaboration and insights.
Data fabric as a foundation.
• A set of data services that spans on-premise and
Cloud environments.
Data analytics as a core business function.

© McGraw Hill LLC antoniodiaz/123RF 6

 Uses of Data Analytics in Business
• Improved and effective decision making.
• Improved customer service.
• Increased efficiency of production and operation
processes.
• Improved patient experiences and internal procedures in
healthcare.

© McGraw Hill LLC 7

 Data Ecosystems
• The information technology (IT) architecture and
infrastructure, software applications, programming
languages, and storage technologies used for the
collection, storage, analysis, and interpretation of
meaningful data.
• Based on the unique circumstances of each
organization, a specialized ecosystem is created for
every business entity.

© McGraw Hill LLC 8

 5 Steps of Data Project Life Cycle
• Sensing - Meaningful data that should be collected by the
organization.
• Collecting - Gathering of data from determined and
validated data sources.
• Wrangling - Conversion of raw data into a user-friendly
format.
• Analysis - Examination of data.
• Storage - Involves hardware, software, and procedures for
securing, maintaining, and accessing data.

© McGraw Hill LLC 9

 Skills and Characteristics of Data
Analysts
Analytical thinking and skills
Include characteristics and capabilities that allow for.
• Observation.
• Research.
• Interpretation of a problem.
Can be applied for the investigation of:
• Supply chain issues.
• Customer relationship management.
• Product and service improvements.
© McGraw Hill LLC 10
 Five Aspects of Analytical Skills
• Curiosity – involves investigation into learning more about a problem or
change of a desired state as well as knowing the right questions to ask to
uncover issues.
• Understanding context – involves understanding where data and information
fit into the plan and approach.
• Having a technical mindset – involves the ability to break down processes and
information into smaller digestible/analytical steps.
• Data design – involves the ability to conceptualize of how data and information
should be organized.
• Data strategy – involves the ability to analyze the people, processes,
hardware, and software used in data analysis.

© McGraw Hill LLC 11

 Responsibilities of a Data Analyst
Data Analyst
Uses the knowledge or processing
software, business strategy, and
analytical skill to deliver data and
reports that guide management as
they create well-informed decision
making
Duties include:
• Working on data analytics teams to extract
data from large data sets.
• Creating reports that outline key findings.
• Monitoring KPIs to identify success or
© McGraw Hill LLC ’

Syda Productions/Shutterstock 12
 Additional Responsibilities
• Collaboration with executives and
other stakeholders to uncover areas
for improvement.
• Data visualization to aid in the
interpretation of data.
• Structuring large data sets to ensure
data is accessible and usable.
• Creating reports and presentations
for management and executives
that outline key findings and
recommendations.
© McGraw Hill LLC metamorworks/Shutterstock 13
 Key Skills of Data Analysts
• Technical writing skills.
• Experience with computer code including
SQL, Python, and Oracle.
• Strong analytical and problem-solving skills.
• Experience with data visualization software
including Tableau and Power BI.
• Microsoft Excel and spreadsheet experience.
• Effective time management and the ability to
multitask and to meet deadlines.
• Oral communication and presentation
software skills.

© McGraw Hill LLC Dragon Images/Shutterstock 14

 Employment
Data analysts are employed
in:
• Finance.
• E-Commerce.
• Healthcare.
• Government.
• Science.

© McGraw Hill LLC Andriy Popov/123RF 15

 U.S. Bureau of Labor Statistics for
Data Analysts
Entry-level education bachelor’s degree in:
• Management information systems.
• Computer science.
• Mathematics or statistics.
• Economics or finance.
Median salary.
• Across different disciplines is $88,770.
Job outlook.
• 15% growth (much faster than average).

© McGraw Hill LLC 16

 How Data Analysts Define Success
for a Project
Formula to define success:

Data
+
Organizational and business
knowledge
=
problem solved

© McGraw Hill LLC NicoElNino/Shutterstock 17

 Solve the Problem
Data must be available.
Data must be accessible.
Data analysts must be
knowledgeable about:
• The organization.
• The problem/question they
want to solve.

© McGraw Hill LLC adam Sternin/Cavan Images/Image Source 18

 Questions Data Analysts Can Ask
• What is the overall outcome and results that are
needed?
• Who is the receiver of the information and analysis?
• What is the question that is being asked? Am I
answering the question?
• What are the time constraints for the project? How
quickly must a decision be made or information
produced?

© McGraw Hill LLC 19

 Data-Driven Decision Making
• What is data driven decision making?
• What are the steps involved in decision making?
• What are the tools used to make decisions?

© McGraw Hill LLC 20

 What is Data-Driven Decision
Making?
According to Tableau, data-driven decision making
(DDDM) is the following:
• The use of facts, metrics, and data to guide strategic business
decisions that align with organizational goals, objectives, and
initiatives.
Organizations are investing in the development of three
data-driven competencies:
• Data proficiency.
• Agility in analytics and data analysis.
• Building of a data-driven culture and community.

© McGraw Hill LLC 21

 Six Steps to Create Data-Driven
Decisions
1. Ask – Determine what questions you need to ask and establish a clear
definition of the problem
2. Prepare – Data should be collected, stored, and secured in preparation for
data processing
3. Process – Data cleansing and checking should occur to ensure high-quality
data is ready for analysis
4. Analysis – Data is analyzed to find patterns, trends, and relationships
within the data set
5. Share – Data is shared with the appropriate audience. This can include
visualization and presentations to key stakeholders
6. Act – Once data and results have been shared, decisions must be made.

© McGraw Hill LLC 22

 Three Key Data Analyst Tools 1

• Spreadsheets.
• Databases and query
languages.
• Data visualization tools.

© McGraw Hill LLC chormail/123RF 23

 Three Key Data Analyst Tools 2

Spreadsheets.
• The two most popular spreadsheet programs are Microsoft
Excel and Google Sheets.
• Cloud-based and desktop versions of Excel, Sheets, and
Numbers are widely used today.
• Allow for the collection, storage, organization, and
manipulation of data.
• Users can identify patterns in data as well as create data
visualization tools with charts and graphs.

© McGraw Hill LLC 24

 Three Key Data Analyst Tools 3

Databases and query languages.

• Allow for the collection, storage, organization, and manipulation of
data.
• Users can extract and query data which cannot be done using
spreadsheet software.
• Databases use structured query language (SQL) to ask questions (or
query) the data.
• SQL allows users to isolate specific information contained in a
database.
• SQL helps in the selection, creation, addition, and extraction of data
from a database.
• Popular SQL programs include MYSQL, Microsoft SQL, and Oracle
SQL.
© McGraw Hill LLC 25
 Three Key Data Analyst Tools 4

Data visualization tools.

• The graphical and structured representation of data.
• Visualization makes it easier to see the results and conclusions
ascertained from data analysis.
• Allow for the creation of charts, tables, graphs, and maps.
• Popular data visualization tools include Tableau, Microsoft
Power BI, and Looker.

© McGraw Hill LLC 26

 Spreadsheets and Databases 1

The type of job and problem to be solved leads to the

choice of tool to be used for data analysis:
• If complex and attention-grabbing graphics are needed, a data
visualization tool might be the best choice.
• If data needs to be organized, cleansed, and analyzed, then
spreadsheet or database software might be the best choice.
Spreadsheets and databases are compared and
contrasted in the following table:

© McGraw Hill LLC 27

 Spreadsheets and Databases 2

Spreadsheets Databases
Software applications that rely on Software applications that rely on
hardware to run and store information hardware to run and store information
Data is structured in rows and columns Data is structured using rules and
The following content is arranged
relationships like aby the DBMS
defined
table.
Information is organized in cells Data is organized in complex and
meaningful collections specified by the
database administrator. These are
referred to as fields.
Often requires manual data entry that is Data entry is often automated and
not constrained constrained by defined parameters
Frequently restricted to one user at a Allows for multiple users to be working at
time, although Cloud-based spreadsheet the same time
programs allow for multiple users

© McGraw Hill LLC 28

 Using SQL to Communicate with a
Database 1

Characters.
• Includes letters, numbers, and
symbols that compose a field.
Fields.
A set of data values made up of
characters.
Also referred to as data attributes.
Are commonly found in columns.
Examples in a student information
table:
• first name, last name, student ID number,
address, email, phone.
© McGraw Hill LLC Wright Studio/Shutterstock 29
 Using SQL to Communicate with a
Database 2

Records.
Collections of related fields.
Are often structured in rows of data.
Example in a student information table:
• The structured information on a specific student who is enrolled at the
college.

© McGraw Hill LLC 30

 Using SQL to Communicate with a
Database 3

Tables.
Groups of assigned rows and columns that contain related data.
Most databases contain multiple tables.
Examples are tables for:
• Student data.
• Faculty data.
• Financial aid data.
• Staff data.
• Course offerings.

© McGraw Hill LLC 31

 SQL (Structured Query Language)
• A popular programming language used to communicate
with databases.
• Pronounced sequel or SQL.
• Useful when working with large data sets.
• Is widely used in today’s business environment.
• Allows user to filter specific data, and to track correlated
pieces of data.

© McGraw Hill LLC 32

 The Basic Syntax of SQL
• Syntax refers to a set of rules
and guidelines that define a
specific computer language.
• Includes the structure of
words, symbols, numbers, and
punctuation used to extract
data and information.
• SQL database queries are
executed using precise syntax
of SQL statements.
© McGraw Hill LLC Paolo De Gasperis/Shutterstock 33
 SQL Query
A query in a database is a question or a request for
specific information contained in a database.
Example:
• “How many people purchased our product in May 2021?“
• “How many customers do we have in Idaho and Washington State?”

SQL statements are used to ask a question or

communicate a request of the database management
system.

© McGraw Hill LLC 34

 Components of SQL 1

• SELECT – Allows the user to choose the precise fields to

be returned.
• FROM – Allows the user to choose the tables where the
fields needed for the query are located.
• WHERE – Allows the user to filter for desired and specific
information.

© McGraw Hill LLC 35

 Components of SQL 2

• In most SQL databases you can write SQL queries using

lowercase letters.
• You may not have to worry about using the correct
spacing.
• It is good practice to use proper syntax rules relating to
case (upper or lower) and spacing in your queries.
• Punctuation rules must be followed exactly.

© McGraw Hill LLC 36

 Components of SQL 3

SELECT followed by the asterisk indicates that

(asterisk)

the user wants to extract all the columns of data

SELECT * FROM Customers FROM the Customers table. Use SELECT without
context sparingly, because in a large database a
huge volume of information may be returned.
SELECT identifies that the user wants to select
SELECT City FROM Customers
The following content isthe Customers table.
the City column FROM
arranged like identifies
SELECT a table. the field to extract data from; in
this case it is the first_name field. FROM indicates
the table in which the field is located, and WHERE
SELECT first_name helps to narrow the query so that the only
FROM customers with the first name of 'David' are
customer_information.customer_name returned. You can use = LIKE ‘da%’ to return
WHERE first_name = 'david' results from all first names that begin with the
letters “da”. The (%) is treated as a wildcard to
match one or more characters. In some databases
the asterisk is used as a wildcard.
(asterisk)

© McGraw Hill LLC 37

 Components of SQL: Extracting
Data From Multiple Fields
• The structure of a SQL query that extracts from
multiple files is:
SELECT student_id, SELECT the columns
first_name, last_name named ‘student_id’,
FROM ‘first_name’, and
The following content is arranged
student_information.student_
like a table. ‘last_name’ FROM the
data field named 'student_data'
WHERE first_name = 'david' in a table named
‘student_information’
WHERE only records from
© McGraw Hill LLC
the first_name column are 38
 Data Visualization
• The graphical and
structured representation
of data.
• Makes it easier to see the
results and conclusions
ascertained from data
analysis.

© McGraw Hill LLC ra2studio/Shutterstock 39

 Three Step Process of Data
Visualization
• Step 1 – Explore data
sets for patterns.
• Step 2 – Planning for
visuals.
• Step 3 – Create your
visuals.

© McGraw Hill LLC ra2studio/Shutterstock 40

 Design Elements That
Lead to Misleading Data
Data visualization helps
Visualizations
Advantages:
individuals to better understand
and interact with data. • Easy access to complex data

• According to a recent study, 90% of the • Interactivity

information transmitted to the human
brain is visual. • Visualization of patterns and relationships in
the data
Various charts and graphs are
used for data visualization Disadvantages:
including:
• Representation of biased or inaccurate
• Pie information
• Bar
• A correlation that does not accurately
• Line represent causation
• Scatterplot • Desired messaging or conveyance of
• Infographics information that is lost in translation
© McGraw Hill LLC 41
 Truncated Graphs
• In a bar, line, or scatterplot
graph, the y-axis is the
horizontal axis, and the x-
axis is the vertical axis.
• Truncating is to shorten or
cut off.
• In data visualization,
shortening (truncating) the
y-axis can make
differences in data seem
much larger than they are.

Access the text alternative for slide images.

© McGraw Hill LLC 42

 Non-Truncated Graphs
• When designing
visualizations, it is
essential to consider the
data ranges represented in
the visualization to ensure
that data is properly
represented.
• This visualization more
accurately displays the
difference in patients that
reported fewer heartburn
symptoms.
Access the text alternative for slide images.

© McGraw Hill LLC 43

 Exaggerated or Improper Scaling
• A pictogram is data that is
represented by a graphic or image
instead of a number.
• The bar or line chart structure is
often used to create pictographs
representing numerical data.
• When designing a pictograph, data
should be uniformly scaled.
• Failure to uniformly scale
pictograms in charts can lead to
misleading comparisons and
inaccurate reflection and
interpretation of data. Access the text alternative for slide images.
© McGraw Hill LLC 44
 Too Many Variables or Too Much
Information
Creating data visualizations that
include too many variables or
contain too much information
can distract the audience from
the message or information
attempting to be conveyed.

Text that is not legible,

unnecessary gridlines, and
chart elements can be
distracting and deter the
Access the text alternative for slide images.

effectiveness of the
© McGraw Hill LLC 45
 Cherry-Picked Data
• Cherry-picking data is the
selection and representation
of data in a visualization that
supports a desired
conclusion or result.
• Data visualizations that
include cherry-picked data
often ignore or omit data
that may contradict the
desired conclusion or result.
• These types of visualizations
Access the text alternative for slide images.

© McGraw Hill LLC may mislead and cause false 46

 Representing All Data (Not
Cherry-Picked)
• This visualization
represents data from the
entire survey and
proposes a different
scenario.
• All areas of customer
service, as measured in
the survey, may not be at
acceptable levels.
• Omitting these areas in
the first visualization
© McGraw Hill LLC
Access the text alternative for slide images.

47
 How to Structure Data
Visualizations to Ensure Data Is
Appropriately
Data visualizations are tools used
Represented
When designing data
to convey a variety of data and visualizations, there are several
information in a meaningful and best practices that can be applied
efficient way. to ensure effective development
and interpretation.
Data visualization:
• Define the purpose
• Is the graphical and structured
representation of data. • Design visualizations designed for
the target audience
• Makes it easier to see the results
and conclusions ascertained from • Include all data
data analysis. • Format data visualizations
• Use color (when appropriate)

© McGraw Hill LLC 48

 Most Popular Charts
and Graphs for Data
Visualization
The most popular charts and graphs for data
visualization are:
• Line charts
• Scatterplots
• Bar charts
• Pie charts

© McGraw Hill LLC 49

 Line Charts
Also referred to as line plots or line
graphs.
Used to compare values over long
and short periods.
Can efficiently display changes
over time in a single category of
data or groups of data categories.
In a line chart:
• The y-axis frequently represents a
continual progression that is often
represented by a measure of time.
• The x-axis demonstrates the values
of a variable across the progression.
Access the text alternative for slide images.

© McGraw Hill LLC Source: https://www.spc.noaa.gov/wcm/adj.html 50

 Scatterplots
Also referred to as a scatter
diagram or X‑Y graph.
Are designed to diagram pairs of
numerical data with one variable
placed on each axis.
• If there is a correlation between the
variables, the data points will fall on a
line or curve.

• The stronger the correlation between

the variables, the tighter the
placement of the data points will be
along the line.
Best used when there are pairs of
numerical data or when tryingAccess
to the text alternative for slide images.
determine if there is a correlation
© McGraw Hill LLC Source: https://www.spc.noaa.gov/wcm/adj.html 51
 Bar Charts
Used to display slices of information.
Designed to plot numerical values in
the form of bars.
Often used to demonstrate the
distribution of data points or to
display values across different
subgroups of data.
Allows for:
• Comparing one group of data to
another.
• Comparing which variables occur most
frequently or have the highest rating.

Access the text alternative for slide images.

© McGraw Hill LLC Source: https://www.weather.gov/hazstat/ 52

 Pie Charts
Used to represent part-to-whole
relationships in a data set.
• Each slice of a pie chart denotes the
quantity/frequency of one variable or
data point.
• All slices added together equal a
whole.
• All slices of a pie chart, when added
together, should equal 100%.
If you see the terms “percent of”
or “part of” it is a good indication
that a pie chart should be used to
visualize the data.
Access the text alternative for slide images.

© McGraw Hill LLC Source: https://www.noaa.gov/organization/acquisition-grants/acquisition 53

 Tableau
Analytics software that offers business intelligence and data
visualization across many areas as well as for organizational
processes.
Allows users to explore and analyze data in seconds using a drag
and drop interface as well as natural language processing which
allows for the presentation of questions.
Can connect external data from:
• Spreadsheets.
• Databases.
• Big data.
• Data warehouses.
• Cloud data.
© McGraw Hill LLC 54
 Tableau Licensing Options
Three role-based licensing options provided by Tableau:
• Creators – can build analytical content including data design,
cleansing, curation of data sources, and the creation of
visualizations and dashboards.
• Explorers – can access and analyze data published by Creators
and can create their own dashboard.
• Viewers – can view and interact with visualizations and
dashboards.

© McGraw Hill LLC 55

 Microsoft Power BI
Analytics software that allows for the processing, manipulation,
and visualization of data.
Users can quickly connect to data, prep it, model and visualize it
using a variety of built-in tools and then securely share insights
both internally and externally.
Users can connect to thousands of data sources located on-
premise or in the Cloud including:
• Microsoft Excel, Salesforce, Google Analytics, a variety of social networks,
and Internet of Things (IoT) devices.
Users can create data visualizations of the data as well as
dashboards that give a 360-degree view of the organization.

© McGraw Hill LLC 56

 Three Tiers of Microsoft Power BI
• Power BI desktop (free to
download).
• Power BI Pro.
• Power BI Premium.

© McGraw Hill LLC Sergei Bachlakov/Shutterstock 57

 Selecting and Using the Right
Data: Data Anonymization
• Aims to protect private or sensitive information by
eliminating or encrypting this type of information.
• "personally identifiable information" (PII) is often used
in data anonymization.
• Organizations have a legal and ethical responsibility to
protect their data and the data of stakeholders and
customers.

© McGraw Hill LLC 58

 Personally Identifiable
Information (PII)
• Name.
• Social security numbers.
• Medical records.
• Email addresses.
• Account numbers.
• Phone numbers.
• IP addresses.

© McGraw Hill LLC maxkabakov/Getty Images 59

 Open Data 1

Data openness (or open data) refers to free access,

distribution, and usage of data.
To be considered as open, data must meet the following
criteria:
• The public must have access and datasets must be available for
use.
• Datasets must have access rights that allow them to be reused
and redistributed.
• Datasets must be universally available so that anyone can use
the data.

© McGraw Hill LLC 60

 Open Data 2

Open databases and datasets allow for more wide use

of data which can lead to many benefits.
For example, data collected during the Covid-19
pandemic was openly shared and used by governments,
health institutions, and municipalities across the globe.
• Allowed for scientific collaboration, research advances, and
eventually vaccine development.

© McGraw Hill LLC 61

 Third Party Data
• Data collected by an organization that often does not
have a relationship with the organization collecting the
data or the data being collected.
• Example: A third party might be tasked with collecting
information on visitors to a social media site of an
organization.
• In order to maintain privacy, it is important that third-
party data and PII are anonymized to ensure personal
identifiable information is not made public or misused.

© McGraw Hill LLC 62

 Selecting the Right Data: Key
Considerations 1

How will the data be collected?

• Will you use internal data that has been collected by the organization
or will you use data from other sources?
Where will you get your data?
• First-party data - data that has been collected internally by the
organization.
• Second-party data - data that has been collected directly by another
entity and then sold for use.
• Third-party data - data that is sold by an entity that did not actually
collect the data.

© McGraw Hill LLC 63

 Selecting the Right Data: Key
Considerations 2

What types of data do you need to solve a specific

problem?
• It is important to choose and use data that will help to solve the
business or organizational problem under investigation.
How much data should be collected?
• Will a random sample size from existing internal data be sufficient?
• Do you need a large sample size in order to achieve statistical
significance?
How much time do you have?
• Time constraints should be identified and data collection should be
closely monitored to ensure deadlines are met.
© McGraw Hill LLC 64
 Selecting the Right Data:
Key Considerations 3

Access the text alternative for slide images.

© McGraw Hill LLC 65

 Data Formats in Datasets 1

• There are a variety of data formats included in

datasets.

© McGraw Hill LLC 66

 Classification of Definition of Data Format Examples of Data Format
Data Format
Primary Data Newly collected data by a researcher, also referred • Data from interviews.
to as first‑hand data
• Data from surveys and questionnaires.

Secondary Data Data that has already been collected by other • Data purchased from second-party and third-
researchers party vendors.
• Demographic data from census.gov.

Internal Data Data from an organization’s own systems • Sales data.

• Data from human resources.
• Inventory data.

Data External Data Data from outside the organization

The following
•
•
Customer credit reports.
Wage information from the Bureau of Labor
Statistics.

Formats Continuous Data content is

Measurable and counted data that can have
almost any numeric value
• Temperature data.

arranged like a • Grade distribution in MIS courses.

in Discrete Data Measurable and counted data that has a limited

number of values
table.
•
•
Number of people who visit a storefront daily.
Number of items sold in a month.

Datasets 2
Qualitative Data Subjective measures of qualities and • How using a product makes you feel.
characteristics
• Favorite brands.

Quantitative Data Specific measures of numerical facts and values • Population of the United States.
• Number of customers.

Nominal Data Type of qualitative data that is not categorized • New product listing, existing product listing,
de-listed products.

Ordinal Data Type of qualitative data that is categorized • Ranked choice of product (1st, 2nd, 3rd).

Structured Data Data that is organized according to a defined • Product inventory.

format (rows, columns)
• Expenses.
• Accounting information.

Unstructured Data Data that is not organized according to any • Most email inboxes.
structure
• Instagram posts.

© McGraw Hill LLC 67

 Data Transformation 1

• The process of converting the format of data from one

form to another.
• Often data must be transformed to make it easier to
analyze.
• Often data needs to be transformed so that it increases
usability.
• Data can be transformed by changing format,
structure, or value.

© McGraw Hill LLC 68

 Data Transformation 2

The transformation of data often involves:

• Copying or replicating data.
• The deletion of database fields or records to ensure data
integrity.
• The standardization of names and variables across data sets.
• Joining of tables or datasets.
• Converting a file to a different format such as saving an Excel
spreadsheet to comma separated values (CSV).

© McGraw Hill LLC 69

 Key Reasons for Data
Transformation
• Data must be organized in a way that makes it easier to
use.
• Data must be compatible so that it can be used between
different systems and computer applications.
• Data must be designed so that it can move from one
system to another.
• Often data must be merged to enhance decision making
ability.
• Data should be structured to allow for enhancement and
comparison.
© McGraw Hill LLC 70
 End of Main Content

Because learning changes everything. ®

www.mheducation.com

© McGraw Hill LLC. All rights reserved. No reproduction or distribution without the prior written consent of McGraw Hill LLC.