Analyzing our automated data in Business Intelligence Tools is totally dependent upon the quality of that data. Sounds obvious? Just how bad government data can be is simply not appreciated. The truth is most of our data is MUCH WORSE THAN WE THOUGHT! All information analysis is inextricably yoked to data quality. Quality is the major obstacle at the front end-not user requirements. You must let the data tell you what uses it can be put to. After a BI Tool is built, it is the IT analysts that become the de facto business analysts because they know the data> '> s limitations best. The documentation of data errors during analysis yield an important by-product--the users can tighten up on what in the future gets initially captured. Why don't we know how bad our data is? What makes it surface finally? Are the error rates the same for all systems-- regulatory, administrative, technical, financial? What is THE crucial step we must control to have any hope of ultimate quality? There's hope, once the problem is recognized.
Presenter:
Ken Currie
Team Leader
SAS Business Intelligence
Department of Environmental Protection
Commonwealth of Pennsylvania
Ken Currie is Team Leader for SAS Business Intelligence in the Department of Environmental Protection and has spent over 30 years in data processing (as Manager of End User Computing and as a principal developer for DEPs foundational enterprise-wide permitting/compliance/enforcement system). He holds a Master of Public Administration from the Maxwell School of Citizenship and Public Affairs, Syracuse University.