Chapter 6 Review Questions

Describe the three-step process of creating an intersection entity.

1. Remove the M:N relationship line and insert a new entity between the two existing ones. 2. Add two 1:N relationships to the model. 3. Name the intersection entity.

What type of high-level business rule can be stated by an ERD? Give two examples.

A business rule is a constraint or guideline to follow during operation of the system. Examples of business rules are: an order belongs to just one customer; a customer cannot cancel an order that has been shipped; a backorder can be created for an out of stock product. Business rules are expressed on ERDs by the kinds of relationships that the entities share.

What is meant by a concatenated identifier?

A concatenated identifier is one in which a combination of attributes serves to uniquely identify an entity. For instance, an appointment entity may have multiple instances for a single date identifier, but the combination of date and time will uniquely identify each instance.

Define what is meant by a relationship in a data model. How should a relationship be named? What information about a relationship should be stored in the CASE repository?

A relationship describes the association between entities. Relationship names are labeled as active verbs. Information stored in a CASE repository regarding a relationship includes: • Verb Phrase • Parent Entity • Child Entity • Definition • Cardinality • Modality • Special Notes

Define what is meant by an attribute in a data model. How should an attribute be named? What information about an attribute should be stored in the CASE repository?

An attribute is a characteristic that describes an entity. Attribute names are nouns. Information stored in a CASE repository regarding an attribute includes: • Name • Definition • Alias • Sample Values • Acceptable Values • Format • Type • Special Notes

Define what is meant by an entity in a data model. How should an entity be named? What information about an entity should be stored in the CASE repository?

An entity is a person, place, thing, or event about which data is collected and stored. Entities names are nouns. Information stored in a CASE repository regarding an entity includes: • Name • Definition • Special Notes

What is the purpose of an intersection entity? How do you know one is needed in an ERD?

An intersection entity is created when we need to capture more information about the relationship between two entities. This often occurs when two entities have a many-to-many relationship. One instance of entity A may be related to many instances of entity B, and one instance of entity B can be related to many instances of entity A. The intersection entity is inserted between entities A and B, and is used to capture information about a specific instance of entity A related to a specific instance of entity B.

A team of developers is considering including 'warehouse' as an entity in their data model. The company for whom they are developing the system has just one warehouse location. Should warehouse be included? Why or why not?

Entities represent something for which there exist multiple instances or occurrences. If there is only one instance of a warehouse, then it would not be best represented by an entity. However, if multiple warehouses were planned in the future, then a warehouse entity should be included.

Describe the analysis that is applied to a data model in order to place it in first normal form (1NF).

First normal form requires that all repeating fields or groups of fields have been removed to separate tables.

Is an intersection entity dependent or independent? Explain your answer.

If the intersection entity has a logical identifier that can uniquely identify instances within, then it would be considered an independent entity. If, however, the intersection entity requires the identifiers from its parent entities to be uniquely identified, then it is a dependent entity.

What is an independent entity? What is a dependent entity? How are the two types of entities differentiated on the data model?

Independent entities are entities that can exist without the presence of another entity. The independent entity does not rely on any other entity in order to exist. On the other hand, dependent entities require the presence of another entity in order to exist. These entities rely on attributes from the parent entity to uniquely identify an instance, and therefore depend on another entity. For example, when an order is placed for a product, an entity that represents a specific product on a specific order is usually created. This entity would not exist without the order and product entities, and in fact gets its identifiers from those entities. So, this ordered_product entity is a dependent entity. Independent entities are represented by rectangles, while dependent entities are usually represented as rectangles with double-border lines.

What is metadata? Why is it important to system developers?

Metadata is information we want to collect and document regarding the components of the data model. Metadata helps us more fully understand the meaning and use of the data model components. Since there are typically several members of the project team, specifying metadata helps ensure that each team member has a consistent understanding of the data model components. Metadata is usually stored in the project repository; CASE tools have their own structures for the entry of metadata. Metadata is captured to help designers better understand the system that they are building and to help users better understand the system they will use. The metadata information can be used to integrate the different pieces of the analysis phase and can lead to a much better design.

What is the purpose of normalization?

Normalization is a process that optimizes relational data storage for storage efficiency. Normalization helps analysts identify entities that are not represented correctly in a logical data model, or entities that can be broken out from a file. The rules of normalization help assure that the data is stored as efficiently as possible.

What is the purpose of developing an identifier for an entity?

One of the aspects of the definition of an entity is the fact that there are multiple occurrences of the entity. If there are not multiple instances of something that is a potential entity, that something is not an entity in the system. Consequently, there must be a way of identifying each individual occurrence of an entity so that it can be picked out from amongst all the other instances of the entity. That is the purpose of having identifiers with unique values.

Describe the analysis that is applied to a data model in order to place it in second normal form (2NF).

Second normal form requires that the data model be in 1NF, and that all partial dependencies have been removed.

What is a CRUD matrix? How does it relate to process models and data models?

The CRUD (create, read, update, delete) matrix shows how data is used by the processes within the system. In the design phase, it helps analysts ensure that all of the data stores used by the processes have been created. This is a tangible way to link the processes from the process models and the data stores from the data model, ensuring that no data required by the processes has been omitted from the data model. It will also depict data from the data model that is not used by any processes, and should therefore be considered for elimination from the data model.

Describe how the data model and process model should be balanced against each other.

The key to balancing DFDs and ERDs is to recognize that all system data must be accounted for on each type of diagram. The ERD shows the system data 'at rest,' while the DFD shows the flow and use of data in the system. Generally, all of the data entities shown on the ERDs will correspond to data stores on the DFDs. That is one aspect of balancing. In addition, the attributes that are a part of the data model should be used somewhere in the flows and stores of the process models.

Provide three different options that are available when selecting an identifier for a student entity. What are the pros and cons of each choice?

The three options available for selecting an entity identifier: (1) Put together a combination of attributes to serve as the identifier (such as first_name and last_name). (2) Sometimes, a single attribute is available that can serve as the identifier (such as social security number, though there may be legal constraints on use of this data in this way). (3) A new attribute can be created to serve as the identifier (such as student_id). Creating a new attribute is normally done in the design phase. Any of these ways of selecting an identifier is acceptable. The important point is to ensure that the chosen identifier uniquely identifies each instance of the entity.

Describe to a businessperson the cardinality and modality of a relationship between two entities.

These two terms are used to refer to the 'numerical' relationship between two entities in a data model. The term, cardinality, refers to the maximum number of times an instance of one entity can be related to instances of the other entity. If the cardinality is one -to-one, then we can infer that one instance of the parent entity can be related to just one instance of the child entity. If the cardinality is more than one, then we know that one instance of the parent entity can be related to more than one instance of the child entity. The determination of cardinality is based upon whatever is appropriate for the business situation being described. The term, modality, refers to whether or not an instance of a child entity can exist without a related instance in the parent entity. Modality values are either null or not null. If the modality is null, then we can infer that no instances of the child entity are required for an instance of the parent entity. If the modality value is not null, then there must be one instance of the child entity for an instance of the parent entity. Just as in the case of cardinality, the determination of modality is based upon whatever is appropriate for the business situation being described.

Describe the analysis that is applied to a data model in order to place it in third normal form (3NF).

Third normal form requires that the data model be in 2NF, and that all transitive dependencies are removed (i.e., that no fields are dependent on other, non-primary key fields).

Explain the distinction between identifying and non-identifying relationships.

When relationships have an independent child entity, they are called non-identifying relationships, When relationships have a dependent child entity, they are called identifying relationships.