Banner

 

22 - Data Integrity
Skill Builders

An Internet bank with more than 10 million customers has asked for your advice on developing procedures for protecting the existence of its data. What would you recommend?

Some of the precautions it should consider are:


1. A consumer electronics company with a well-respected brand offers a wide range of products. For example, it offers nine world band radios and seven televisions. What data quality strategy would you recommend?

Recommendation because customers might be uncertain as to what features will meet their needs.

2. What data quality focus would you recommend for a regulated natural gas utility?

Confirmation because customers should be certain of their needs (e.g., turn on or off gas), and firm performance variation should be small.

3. A telephone company has problems in estimating how long it takes to install DSL in homes. Sometimes it takes less than an hour and other times much longer. Customers are given a scheduled appointment and many have to make special arrangements so they are home when the installation person arrives. What data might these customers expect from the telephone company and how would they judge data quality?

Performance deviation because customers should be certain of their needs, but the firm has probelms in consistent performance.

1. What level of granularity of sales data would you recommend for an online retailer?

Hourly granularity because business volue will typically vary during the day and also differnt types of customers might shop at different times.

2. What level of data quality completeness might you expect for a university’s student table?

You could expect the student table to have an identifier, full name details, and demographic data. Address, major(s), classes, and so forther will be stored in related tables.

A university has decided that it will e-mail students their results at the end of each semester. What procedures would you establish to ensure that only the designated student opened and viewed the e-mail?


A small nonprofit organization has asked for your help in selecting a relational database management system (RDBMS) for general-purpose management tasks. Because of budget limitations, it is very keen to adopt an open source RDBMS. Search the Web to find at least two open source RDBMSs, compare the two systems, and make a recommendation to the organization.

I suggest that you compare PostgreSQL and MySQL.

Complete the data integration lab exercise described on the book’s Web site (Lab exercises > Data integration).


1. Global Electronics has nine factories in the U.S. and Asia producing components for the computer industry and each has its own information systems. Although there is some specialization, production of any item can be moved to another factory, if required. What level of data integration should the company seek and what systems should be targeted for integration?

High. Integrate most systems especially production planning and inventory

2. European Radio, the owner of 15 FM radio stations throughout Europe, has just purchased an electronics retailing chain of 50 stores in Brazil. What level of data integration should the company seek and what systems should be targeted for integration?

Low. Little scope for integration across the two divisions. Could integrate some HR functions, but even this might be problematic because of all differences in employment laws. Accounting standards might also vary, which would limit the integration of accounting systems.

This page is part of the promotional and support material for Data Management (sixth edition) by Richard T. Watson
For questions and comments please contact the author

Date revised: 19-Oct-2016