George C. Philip

Teaching Database Modeling and Design: Areas of Confusion and Helpful Hints

Introduction Database design has been an integral part of MIS curricula for decades. This field is wellestablished, with an abundant supply of research papers and textbooks, many of which have undergone several revisions. Yet, lack of clarity and inaccuracies in the presentation of some basic database modeling and design concepts persist in textbooks. This, in turn, creates confusion and difficulty for students, and even for faculty. This paper identifies such problems in several areas and provides helpful hints for faculty and students. Specifically, the paper discusses the practical role of normalization in database design, addresses the confusion in representing attributes with repeating values, discusses how to remove inconsistencies in defining relations and first normal form, simplifies the process of identifying candidate keys to normalize relations, clarifies the conditions under which insertion and deletion anomalies may occur, and sheds light on the confusion in defining weak entities. The discussions here are based on an examination of seventeen database textbooks (Connolly & Begg, 2005; Date, 2004; Elmasri & Navathe, 2007; Frost & Van Slyke, 2006; Gillenson, 2005; Hoffer, Prescott, & McFadden, 2007; Kifer, Bernstein, & Lewis, 2006; Kroenke, 2006; Mannino, 2007; Post, 2005; Pratt & Adamski, 2008; Riccardi, 2003; Rob & Coronel, 2006; Rob & Semaan, 2004; Ullman & Widom, 2008; Umanath & Scamell, 2007; Watson, 2005). Practical Significance of Normalization Because several areas discussed here are related to normalization, we first examine the practical role of normalization and normal forms in designing databases. Database design may use a variety of inputs, including existing/proposed forms, reports, queries and transactions, data stored in spread sheets and legacy systems, and entities/attributes identified from knowledge of the business. Thus, for example, to develop a database for a new inventory control system to replace an existing spreadsheet-based system, the designer might use existing spreadsheets, reports, and additional data items represented by entities, such as vendors and product lines, that are identified from user requirements and knowledge of the business. The Top-down Approach In order to translate the data from various sources to a well-designed database, the top-down approach popularly used in relational database design uses three basic steps: 1) Create a conceptual schema that is represented by entity-relationship models, 2) Create a logical schema by mapping the conceptual schema to relation schemas using mapping rules, and 3) Apply normalization rules to test the goodness of the design and to improve the design, if necessary. In general, if the attributes, entities, and their relationships are identified correctly, the proper application of mapping rules should result in normalized relations. However, because E-R modeling is an intuitive process, errors could occur in identifying entities and their relationships and in applying mapping rules, resulting in un-normalized relations. Consider a segment of an order form for a mail-order nursery, shown in Table 1. Experience with student assignments indicates that inexperienced designers are likely to make mistakes in identifying entities and their relationships. Two common mistakes are: 1) Correctly recognize Customer as an entity, but incorrectly group the rest of the information as a single Order entity with a one-to-many relationship between Customer and Order. Applying the mapping rules to this conceptual schema would result in an un-normalized Order schema with data redundancy: Order (Order#, Date, Cust#, Item#, Description, Qty, Unit Price). 2) Correctly identify all three entities--Customer, Order, and Item--but incorrectly identify the relationship between Order and Item as one-to-many, resulting in an un-normalized Item schema: Item(Item#, Order#, Description, and Unit Price). …

Software design guidelines for event-driven programming

Abstract This paper deals with software design principles and guidelines to improve the reusability and maintainability of event-driven (E-D) programs. The paper examines how well the principles of structured software design from the procedural programming field can be applied to the event-driven environment. Taking into account the unique characteristics of event-driven programming (E-DP), additional guidelines that are specific to this field are proposed. The guidelines presented here deal with modularizing event procedures, graphical representation of E-D programs, sharing data between event/general procedures, using user-defined objects, and developing cohesive procedures and forms.

The Effects of Involvement on E-Satisfaction Models

E-tailing has changed the fundamental structure of marketing channels and consumer behavior. At the same time, consumer satisfaction is becoming even more critical to the survival and growth of e-tailers as they face both traditional and nontraditional competition. The two main objectives of this study are to find: (a) whether the disconfirmation model or the perceived performance model is better to explain the satisfaction of e-shopping for the high involvement decision group and the low involvement decision group, and (b) the relationship between the level of e-satisfaction and the probability of shopping at the same e-tailer for these two groups. The study found that for both low involvement and high involvement cases, the performance model clearly was superior to the disconfirmation model. The results also supported the relationships between e-shopping satisfaction and purchase intention in both levels of involvement.

Operations Research in Medium-Sized Companies and the Microcomputer Revolution: Educational Implications

Abstract This research focuses on the current practices of operations research/management science (OR/MS) techniques in medium-sized companies and identifies their impact on the education of OR/MS. Data were collected from 479 organizations on characteristics of the firms that use or do not use OR/MS, techniques used and where they are used, types of computer systems used, problems of implementation, and reasons for not using OR/MS. Our survey indicated that OR/MS courses should provide training in microcomputer-based OR/MS tools and in accessing databases on larger computers; familiarize students with embedding OR in other types of applications and interfacing with them; include real world projects or cases; place more emphasis on applications in the service sector; and provide students with a strong background in the application of statistics, project management, simulation, linear programming, and “quick and dirty” techniques.

The Task of Problem Formulation

Problem formulation is an important organizational task because it facilitates effective problem solving. However, it is not well understood. Past studies have examined problem formulation by examining the behavior of individuals performing problem formulation tasks. This approach has proven difficult due to differences in individual knowledge of the task domain. This study examines problem formulation by focusing on the characteristics of the task rather than on an individual performing the task. A rational model of the task of problem formulation is developed and examined for its implications on individual problem formulation behavior. An empirical investigation of individuals performing problem formulation shows limited support for the proposed model.

An Investigation of Student Expectation, Perceived Performance and Satisfaction of E-textbooks

This paper examines the use of e-textbooks in a college level introductory information systems course using an empirical study that gave students the option to buy electronic or print versions of the same textbook. The study measured and analyzed student expectations prior to purchase, perceived performance and satisfaction after use, intention to use e-textbooks in the future and the determinants of satisfaction using factor analysis and regression modeling, with a consumer behavior focus.

Guidelines on improving the maintainability and consulation of rule-based expert systems

Abstract This paper deals with the issues of maintenance and consultation of rule-based expert systems. General approaches to knowledge-base design to improve the maintainability of expert systems are reviewed. The paper focuses on guidelines on developing rules to improve the maintainability and consultation. Existing guidelines are presented, some guidelines expanded, and new ones proposed. The concept of cohesive rule is introduced by applying the principle of functional cohesion to development of rules. The paper also examines the problem of redundant questions during consultation and presents guidelines to minimize the problem.

An Examination of the Characteristics Impacting Collaborative Tool Efficacy: The Uncanny Valley of Collaborative Tools.

As collaboration among teams that are distributed in time and space is becoming increasingly important, there is a need to understand the efficacy of tools available to support that collaboration. This study employs a combination of the Technology Acceptance Model (TAM) and the Task-Technology Fit (TTF) model to compare four different technologies used to support the task of collaboratively creating and editing a report. The characteristics of the four technologies operationalize different collaborative processes and quality of the editor enabling the examination of how well each addresses previously identified challenges to electronic collaboration. The study compared the use of MS Word combined with emailing the document among collaborators, Twiki, Google Docs, and Office Live by university students to collaboratively write a research paper. The study measured Task-Technology Fit, Perceived Ease of Use, and Perceived Usefulness of the technologies. The findings suggest that certain technological factors associated with the collaborative tool can impact effective use of the tool. A powerful interface with which the user has some experience is necessary, and support for distributed collaboration can be the distinguishing factor in making a tool an effective collaborative writing and editing technology. Our study suggests that Word/email and Google Docs outperform Twiki and Office Live due to tool experience and superior Task-Technology Fit that may be due to the sophistication of the writing and editing tool, support for collaboration, and the clarity of the collaboration process. We speculate that the surprisingly poor showing for Office Live is due to mixing a familiar editing interface with a collaborative model that is different from students’ mental model of collaboration, leading to a phenomenon similar to the ‘Uncanny Valley,’ which is the experience of many people when interacting with humanoid robots and animations that have very good, but not perfect, realism.

Normalization of Relations with Nulls in Candidate Keys

This paper discusses normalization of relations when the candidate keys of a relation have missing information represented by nulls. The paper shows that when the missing information is of the type “not applicable†or “does not exist,†problems and confusion can arise in normalizing relations. Candidate keys with missing information commonly are found in relations that represent information on two entities with a one-to-one relationship between them. The current definition of Boyce-Codd Normal Form (BCNF) is ineffective in identifying poor designs in such relations that may have insertion/deletion anomalies. It is shown that the above problem can be corrected by incorporating the concept of entity integrity rule into the definition of BCNF. This paper also shows that incorporating the entity integrity rule into the definition of either a relation or a candidate key does not provide a satisfactory solution to the problem.