Unit 4. -- Approaches to Working with Dimensions Quiz

This quiz will test your understanding of the concepts discussed in unit 4.

1. Which of the following are true when dealing with slowly changing dimensions?

a. a benefit of adding an old field to store the old value of the attribute and placing the new value in the existing field does not require changing underlying code

b. a benefit of adding dimension records to an existing dimension to store new values is that it does not require changing underlying code

c. adding new records to an existing dimension to store new values is not typically a problem when working with large tables such as customer and product

d. usually the recommended approach is just to overwrite existing values

e. a & c

2. Junk dimensions are?

a. dimensions with no values
b. dimensions holding flags and attributes for which their meaning is obscure
c. artificial
d. all of the above
e. b & c

3. Which of the following are true regarding foreign keys and primary keys?

a. the SQL date is a good value for the time primary key

b. intelligent keys allow querying directly on fact tables and so are recommended for implementation

c. the time field should be separated from the date field in the time dimension

d. for primary keys, production keys should be avoided because they may be reused by the production system

e. c & d

4. Which of the following are true regarding additivity of measures?

a. dollars_sold is an example of a fully additive measure
b. room temperature is an example of a non-additive measure
c. percentages and rates are additive
d. measures of intensity are always fully additive
e. a & b

5. For rapidly changing large dimensions which of the following are true?

a. a separate dimension can be created to hold the rapidly changing attributes

b. numerical values in the newly created dimension should be represented in banded values

c. if a separate key is not added to the original dimension those separate dimension attributes can only be linked to the original dimension through the fact table

d. a, b & c

e. none of the above

	a. a benefit of adding an old field to store the old value of the attribute and placing the new value in the existing field does not require changing underlying code
	b. a benefit of adding dimension records to an existing dimension to store new values is that it does not require changing underlying code
	c. adding new records to an existing dimension to store new values is not typically a problem when working with large tables such as customer and product
	d. usually the recommended approach is just to overwrite existing values
	e. a & c

	a. the SQL date is a good value for the time primary key
	b. intelligent keys allow querying directly on fact tables and so are recommended for implementation
	c. the time field should be separated from the date field in the time dimension
	d. for primary keys, production keys should be avoided because they may be reused by the production system
	e. c & d

	a. a separate dimension can be created to hold the rapidly changing attributes
	b. numerical values in the newly created dimension should be represented in banded values
	c. if a separate key is not added to the original dimension those separate dimension attributes can only be linked to the original dimension through the fact table
	d. a, b & c
	e. none of the above