Data Modelling Data modellingis the process of analyzing the data required by the processes of an organization to support it both operationally and strategically. As an example, a business cannot determine what sales tax it needs to remit to the government if it has not gathered this information at the point of sale of its products or services. Therefore, that piece of data must be collected, stored, and able to be reported on when required. Data models are often the first, high level step in designing a relational data-base; in this process the required data is defined and any integration with other systems is determined. Typical products of data modelling include a data diction-ary, which documents the origin, format, and meaning of the data and other more detailed models such as an entity relationship diagram and a data ﬂow diagram.Entity-Relationship Diagram and Logical Data Model The entity-relationship dia-gram (ERD)and the logical data model are the two most commonly used models for designing the organization of a relational database. The ERD indicates the enti-ties and relationships for the data that the IS will store (see Tech Guide D for a more detailed discussion of ERDs). The logical data model then translates the ERD into a diagram of the tables in the database. Figure 6.10 shows a partial ERD and logical data model for an electronic voting system.FIGURE 6.9 Organizations often use SQL queries, similar to the one shown here, to obtain specific information from their databases, such as profit calculations. This example shows that of the two items BackPacks R’ Us supply, the DayTripper provides more profit per item. Another query can be done to combine this profit information with sales information.E1C06.indd 218E1C06.indd 21810/05/10 4:43 PM10/05/10 4:43 PM
219CHAPTER6Data Flow Diagram As the name implies, a data ﬂow diagram (DFD)is a tradition-al IS model that depicts how data move or ﬂow through a system: (1) the external entities (boxes) that send input or receive output from the system; (2) processes (boxes with rounded corners) that show activities that move or transform data; (3) data stores (open-ended boxes) that usually correspond to tables in the data model; and (4) data ﬂows (arrows) that connect the components.Figure 6.11 shows a partial DFD for an electronic voting system. In this figure, one external entity, the voter, interacts with the system. Three data stores (tables from the logical data model in Figure 6.10) indicate the need to store various data Voter(a)(b)VoterIDVoterFirstNameVoterLastNameVoterRidingVoterAddressVotesGets votes fromVotes forM:1M:1VoterIDDateTimePollLocationFK_VoterIDFK_CandidateIDCandidateVoterCandidateCandidateIDCandidateFirstNameCandidateLastNameCandidateOfficeCandidatePartyCandidateTotalVotesFIGURE 6.10 The two most commonly used models for designing the organization of a relational database are the (a) ERD and (b) logical data model, used here to represent an e-voting system.