Chapter 6 Questions
In representing a 1:N relationship in a relational database design, the key of the table representing the entity of the 'many' side is placed as foreign key in the table representing the entity on the 'one' side of the relationship.
False
It is easy to enforce the referential integrity actions for M-M relationships.
False
The design transformation for all IS-A relationships can be summarized by the phrase "place the key of the parent table in the child table."
False
To represent a 1:N relationship in a relational database design, an intersection table is created.
False
To represent a one-to-many relationship in a relational database design, the key of the child table is placed as a foreign key into the other table
False
To represent an IS-A relationship in a relational database design, the IS-A relationship must be converted into a HAS-S relationship.
False
When creating a relational database design from E-R diagrams, first create a relation for each relationship.
False
When creating a table in the relational database design from an entity in the extended E-R model, the attributes of the entity become the rows of the table.
False
When the key of one table is placed into a second table to represent a relationship, the key is called a relational key in the second table.
False
When the parent entity has a surrogate key, the enforcement actions are the same for both parent and child.
False
When transforming supertype/subtype entities into a relational database design, all of the attributes for the supertype table are placed into the subtype relations.
False
When transforming supertype/subtype entities into a relational database design, entity is created for the supertype only.
False
Whether or not an attribute is required is determined during the database modeling phase.
False
A referential integrity constraint policy that insures that foreign key values in a table are correctly maintained when there is a change to the primary key value in the parent table called cascading insertions.
False
An entity needs to be examined according to normalization criteria before creating a table from it in the relational database design.
False
An intersection table can have additional attributes besides the keys of its parent tables.
False
By default, the identifier of the entity becomes the foreign key of the corresponding table.
False
Cascading deletions are generally not used with relationships for weak child entities.
False
Data types are consistent across all DBMS products.
False
Discriminator attributes can be represented in relational designs.
False
In a 1:N relationship, the term parent refers to the table on the 'many' side of the relationship.
False
All identifying relationships are 1:N
True
An ID-dependent table can be used to represent multivalued attributes.
True
An association tables sometimes connect more than two entities.
True
An intersection table is always ID-dependent on both of its parent tables.
True
Cascading deletions are generally not used with relationships between strong entities.
True
Each entity in the extended E-R model is represented as a table in the relational database design.
True
If the parent is required, then a new child row must be created with a valid foreign key value.
True
In 1:N relationships, the table on the 'many' side is called the child.
True
In a 1:1 relationship, the foreign key is defined as an alternate key to make the DBMS enforce uniqueness.
True
In a 1:N relationship, the table on the 'one' side is called the parent.
True
In a relational database design, all relationships are expressed by creating a foreign key.
True
In many-to-many relationship in a relational database design, the primary keys of both tables are joined into a composite primary key in the intersection table.
True
In representing a 1:N relationship in a relational database design, the key of the table representing the parent entity is placed as a foreign key in the table representing the child entity.
True
Like all ID-dependent relationships, the parents of an association table are required.
True
One of the important properties of a column is whether or not it can have a NULL value.
True
One of the important properties of an attribute is whether or not it is required.
True
Recursive M:N relationships are represented with an intersection table that shows pairs of related rows from a single table.
True
Referential integrity constraints should disallow adding a new row to a child table when the foreign key does not match a primary key value in the parent table.
True
The first step in the database design process is to create tables and columns from entities and attributes.
True
Many-to-many relationships are represented by ________.
an intersection table which has 1:N relationships with the two tables
A surrogate key should be considered when the key contains a lengthy text field.
True
Which of the following is not true about representing subtypes in a relational database design?
All of the attributes of the supertype are added to the subtype relations.
What is an association relationship and how does it differ from an N:M relationship?
An association relationship is very similar to an N:M relationship except that the intersection table has attributes of its own. This means that in addition to the foreign key fields linking to the two strong entities, there is at least one additional field in what would otherwise be called the intersection table but is now an association table. For example, intersection table ENROLLMENT for STUDENT and CLASS showing student enrollment in each class would normally have two columns: StudentID and ClassID. However, we can turn this intersection table into an association table by adding the column Grade, which records each student's grade in each class.
A referential integrity constraint policy that guarantees that a row in a parent table always has a required entry in a child table ________. A) is known as a minimum cardinality enforcement action B) needs to be documented by the database development team
Both A and B are correct
For the M-O (parent mandatory, child optional) case, what action(s) should be taken to ensure minimum cardinality is maintained?
Both A and B are correct
In a supertype-subtype structure, discriminator attributes ________.
Both B and C are correct
Which of the following is not true about N:M recursive relationships?
Both B and C are correct.
What are MUST NOT and MUST COVER constraints?
Both MUST NOT and MUST COVER constraints occur when working with three tables in a ternary relationship. When a MUST NOT constraint occurs, it indicates that one binary relationship of two of the tables contains combinations of values that are not allowed to occur in the ternary relationship. When a MUST COVER constraint occurs, it indicates that one binary relationship of two of the tables contains combinations of values that must all appear in the ternary relationship.
Which of the following is not a step in the database design process?
Create constraints and triggers
A default value is the value the user enters into the row the first time the user enters data.
False
A foreign key is a key that does not belong in any table.
False
A null value is an attribute value that has been set to zero.
False
What are the four sets of minimum cardinalities that can be present in a 1:N binary relationship?
If we use the labels "Parent" and "Child" to represent, respectively, the one and many side of a 1:N relationship, we can then consider whether the parent and child must participate in the relationship. If one must participate, it is mandatory (M). If one does not have to participate it is optional (O). The four possible sets of minimum cardinalities are therefore: Parent optional and Child optional (O-O) Parent mandatory and Child optional (M-O) Parent optional and Child mandatory (O-M) Parent mandatory and Child mandatory (M-M)
Which of the following is not true of MUST constraint?
It can be expressed directly in a relational model
Which of the following is not true about a NULL attribute?
It must have a default value specified.
Explain the representation of a many-to-many strong entity relationship in a relational database design.
Many-to-many relationships cannot be directly represented in a relational database design. Therefore, many-to-many relationships are essentially broken into two one-to-many relationships by creating an intersection table that represents the relationship itself. The intersection table takes its key as a combination of the keys of the two original, or parent, entities. Each of the parent entities has a one-to-many relationship with the intersection table that is represented by placing the keys of the parents into the intersection table.
Explain the representation of a one-to-many strong entity relationship in a relational database design.
One-to-many relationships are represented by placing the primary key of the table on the one side of the relationship into the table on the many side of the relationship as a foreign key. The term parent refers to the table on the one side of a 1:N relationship, and the term child refers to the table on the many side of the 1:N relationship. Therefore, the rule for representing a one-to-many relationship can be summarized as "Place the key of the parent table in the child table as a foreign key."
How are one-to-one recursive relationships addressed using the relational model?
One-to-one recursive relationships are addressed just the same as one-to-one nonrecursive relationships. The only difference is that both of the related entity instances are in the same entity class. The key of either instance is placed in the other instance as a foreign key. In the case of a recursive relationship, this means that a new attribute is added to the entity class with a recursive relationship. For each instance, this new attribute will contain the value of the key attribute of the instance that is related.
Explain the pragmatic reason for using surrogate keys.
Primary keys are commonly included in indexes, and are used to identify records to be retrieved by users. The ideal primary key is short, numeric and fixed. When the primary key contains a lengthy text field, this creates a large amount of duplicated data that must be frequently manipulated. For these reasons, it is often practical to use a surrogate key that is generated by the system and is relatively small and easy to manipulate.
In many-to-many relationships between strong entities in a relational database design, which of the following is not true?
The intersection table is ID-dependent on one of the parents
A surrogate key should be considered when ________.
The key contains a lengthy text field
Using the VRG database as an example, briefly describe what a final database design consists of.
The obvious part of the database design is the ER drawing itself, for example as shown for VRG in Figure 6-39. However, this needs to be accompanied by a summary of the database design relationships as shown in Figure 6-40. This document summarizes the cardinalities of relationships. For each relationship, we need a set of actions to enforce the minimum cardinalities. These can be based on Figures 6-28(a) and 6-28(b), the examples for VRG are shown in Figures 6-41 and 6-42. Finally, we need a set of column characteristics for each table, as shown in Figure 6-43 for VRG.
Which of the following is not true about surrogate keys?
They are non-unique within a table.
A surrogate key is a unique, system-supplied identifier used as the primary key of a table.
True
A surrogate key is appropriate when the primary key of a table contains a lengthy text field.
True
Briefly describe the process of converting an extended E-R model into a relational database design.
To convert an extended E-R model into a relational database design, first tables and columns must be defined for each entity attribute. The attributes of the entity are represented as the columns of the table. Primary keys are then selected for each table, after which the relationships between the entities are represented through the creation of foreign keys. Next, constraints are specified. Finally the normalization is verified.
A data constraint is a limitation on data values.
True
A key of an intersection table is always the keys of both parents.
True
A referential integrity constraint policy that insures that all rows containing a particular foreign key value in a table are eliminated from the table when the row containing the corresponding primary key value in a parent table is eliminated from the database is called cascading deletes.
True
The ideal primary key is short, numeric, and fixed.
True
The last step in creating a table is to verify table normalization.
True
The values of a surrogate key have no meaning to the users.
True
To represent a 1:1 binary relationship in a relational database design, the key of one table is placed into the second table.
True
To represent an N:M relationship in a relational database design, a table is created to represent the relationship itself.
True
To represent an N:M relationship in a relational database design, an intersection table is created.
True
To represent an N:M relationship in a relational database design, in essence it is reduced to two 1:N relationship.s
True
When creating a table for an ID-dependent entity, both the key of the parent and the key of the entity itself must appear in the table.
True
When placing a foreign key for a 1:1 relationship, the key of either table can be used as the foreign key in the other table.
True
When the child entity is required (M) in a relationship, there needs to be at least one child row for each parent row at all times.
True
When the parent entity is required (M) in a relationship, ever row of the child table must have a valid, non-null value of the foreign key.
True
When transforming an ID-dependent E-R data model relationship into a relational database design where the child entity is designed to use a surrogate key, the relationship changes to a week but not ID-dependent relationship.
True
When transforming an extended E-R model into a relational database design, recursive relationships are treated fundamentally the same as other HAS-A relationships.
True
When transforming supertype/subtype entities into a relational database design, the key of the supertype table is placed into the subtype table typically as the key.
True
Which of the following is not true of recursive relationships?
When the relationship is 1:N, a new table must be defined to represent the relationship.
The binary constraint MUST COVER indicates that ________.
a binary relationship includes a set of value combinations that must all occur in a ternary relationship
The binary constraint MUST NOT indicates that ________.
a binary relationship includes value combinations that must not occur in a ternary relationship
Which of the following is not true for an ideal primary key?
a composite of several long attributes
A referential integrity constraint policy that insures that all rows containing a particular foreign key value in a table are eliminated from the table when the row containing the corresponding primary key value in a parent table is eliminated from the database is called ________.
cascading deletes
A referential integrity constraint policy that insures that foreign key values in a table are correctly maintained when there is a change to the primary key value in the parent table is called ________.
cascading updates
Each attribute of an entity becomes a(n) ________ of a table.
column
The first step in transforming an extended E-R model into a relational database design is to ________.
create a table for each entity
In a relational database design, all relationships are expressed by ________.
creating a foreign key
When transforming an ID-dependent E-R data model relationship into a relational database design and the parent entity has a surrogate primary key and the child entity has a data identifier, then the primary key of the child table should be ________.
either A or B will work, but method B is preferable
When transforming an ID-dependent E-R data model relationship into a relational database design, the referential integrity constraints should allow ________.
parent updates to cascade
The identifier of the entity becomes the ________ of the corresponding table.
primary key
In relational database design, ID-dependent entities are not used to ________.
represent relationships where the child identifier does not include the key of the parent
When representing a 1:1 relationship in a relational database design, ________.
the key of either table may be placed as a foreign key into the other
To represent a one-to-many relationship in a relational database design, ________.
the key of the parent is placed as a foreign key into the child
In many-to-many relationships in a relational database design, ________.
the keys of both tables are placed in a third table
When representing a one-to-many relationship in a relational database design, ________.
the parent is always on the one side of the "one-to-many" relationship
When transforming an E-R data model into a relational database design, the key of the parent entity should be placed as part of the primary key into the child entity ________.
when the child entity is ID-dependent