Deck 2: Data Management

Question 1

A common adage used in data management and utilization circles about poor data quality is

A) "a bad workman always blames his tools."
B) "a cat has nine lives."
C) "garbage in, garbage out."
D) "adversity and loss make a man wise."

Answer 1

"garbage in, garbage out."

Question 2

Achieving value through big data requires a thorough understanding of the goals and objectives of a business.

Answer 2

True

Question 3

Which of the following characteristics of big data increases complexity and reduces confidence in the data?

A) variety
B) volume
C) veracity
D) velocity

Answer 3

veracity

Question 4

Most traditional Extract, Transform, and Load (ETL) tools can process only relational datasets for unstructured data, semi-structured, and machinery sensor data, but newer systems are much more flexible.

Question 5

Describe and explain the four processes of data transformation.

Question 6

In a dataset, values that are at a considerable distance from any of the other data clusters are treated as outliers.

Question 7

Data inconsistencies can lead to uncoordinated strategic decision making and embarrassing dilemmas for businesses.

Question 8

Which of following characteristics of big data means the data can be converted into meaningful, quality information that can be used to achieve tangible business benefits?

A) velocity
B) variety
C) value
D) volume

Question 9

A major challenge of today's data management is that

A) the inbound data goes on to increase exponentially.
B) data is not held in vaults for viewing by business department heads.
C) converting structured data into a usable form is a complex process.
D) data tends to trickle in, making simultaneous processing difficult.

Question 10

A ________ of data can contain as much information as half of the contents of all U.S. academic research libraries.

A) petabyte
B) terabyte
C) megabyte
D) gigabyte

Question 11

Companies that adopt ________ typically achieve up to 6 percent higher productivity and output than their peer companies.

A) multi-channel interactions
B) static information and fixed rules
C) social media sales campaigns
D) data-driven decision making

Question 12

A company combined monthly sales data into a single group to calculate the total sales by a quarter or a year. This is an example of the ________ step of data transformation.

A) feature construction
B) aggregation
C) dummy coding
D) normalization

Question 13

Compare and contrast relational and non-relational databases.

Question 14

In data aggregation, the ________ function helps in obtaining the total values within a dataset.

A) SUM ()
B) MAX ()
C) MIN ()
D) AVG ()

Question 15

Streaming data, the continuous transfer of data from numerous sources in different formats, cannot be included to capture customer data.

Question 16

In data transformation, ________ involves subtracting a variable from the mean and then dividing it by the standard deviation.

A) dummy coding
B) normalization
C) new column construction
D) overfitting

Question 17

Asking the question "Is the data correct, reliable, and precisely measured?" will help in determining the data quality of ________.

A) accuracy
B) format
C) consistency
D) timeliness

Question 18

Unlike relational databases, non-relational databases

A) make drilling down to very specific types of data easy.
B) distribute all data into structured tables.
C) are used by companies dealing with fixed, well-defined data.
D) provide greater flexibility for storing ever-changing data.

Question 19

Structured querying language is a language developed by IBM and used to access and update data stored in a database.

Question 20

The ________ characteristic of big data enables real-time business responses and strategies.

A) volume
B) variety
C) velocity
D) value

Question 21

Hadoop, a data warehouse built using HIVE, provides SQL-like query to access data stored in different file systems and databases that are used by HIVE.

Question 22

The veracity characteristic of big data provides businesses a holistic understanding of their customers and market situations.

Question 23

In data aggregation, employing the COUNT () function within a dataset returns the

A) average value within the dataset.
B) sum of values within the dataset.
C) number of rows corresponding to a certain feature.
D) maximum value within the dataset.

Question 24

In the context of data quality, ________ is the extent to which information is adequately presented or delivered for efficient and effective understanding.

A) accuracy
B) timeliness
C) consistency
D) format

Question 25

Explain how the Extract, Transform, and Load (ETL) integration process uses Hadoop to capture, store, process, secure, and then analyze complex data.

Question 26

Identify the second step the MapReduce platform performs to solve problems of big data computation.

A) Provide SQL-like query to access data stored in different file systems and databases.
B) Map the data by dividing it into manageable subsets.
C) Distribute data to a group of networked computers for storing and processing.
D) Combine the answers from the computer nodes into a single answer for the original problem handled.

Question 27

In the context of database management systems, primary keys and foreign keys are important in relational databases because they help

A) database users collate data from different tables.
B) manage the volume and speed of incoming data.
C) customers search for products online effectively.
D) databases consistently maintain the same properties.

Question 28

Imagine a company wants to send a follow-up email to its best customers immediately after a purchase. However, email addresses were not collected from the customers at the time of purchase. This is an example of missing the data quality of ________.

A) accuracy
B) timeliness
C) consistency
D) variety

Question 29

A gigabyte of data can contain ________.

A) a single character of text
B) all the X-rays in a large hospital
C) a small novel
D) Beethoven's 5th Symphony

Question 30

In the context of data preparation, a power calculation helps determine

A) if any data observations have a considerable distance to other data clusters.
B) whether an existing, valuable customer is likely to leave the business.
C) the what, when, and who of the analysis.
D) what outcomes will be estimated from a sample with a sufficient level of precision.

Question 31

In data transformation, ________ involves creating a dichotomous value from a categorical value.

A) feature construction
B) overfitting
C) dummy coding
D) data compression