Data Validation Saves Your Bacon

Data Quality, Management, and Review
Oral Presentation

Prepared by D. Shepperd1, P. Newbold2, J. Rossi2, M. D'Almeida2
1 - ddms, inc., 2014 Carol Drive, Wilmington, DE, 19808, United States
2 - ddms, inc., 186 Center St, Suite 290, Clinton, NJ, 08809, United States

Contact Information: [email protected]; 302-750-7775


We use analytical data to do many things. For instance, to estimate baseline conditions, make projections, model potential impacts of an activity or event, verify completion of a process or event, evaluate impact to human health and the environment. Regardless of the intended application all testing data are subject to error, and all data must be supportable and of “known and documented” quality. Knowing the magnitude of potential error is critical to predicting outcome. So how can you trust the quality of the data your models or decisions are made from, unless you have a close look at how they were produced? Webster’s Dictionary defines validation as “the action of checking or proving the validity of accuracy of something.” Validation of testing data involves an intensive review of all aspects of the investigation and measurement systems, and assessment of the usability (accuracy and precision) of the results, as well as whether or not those results and supporting data are legally defensible. It can take considerable time, resources, and money to produce the results you will be using. Assuring the quality of those results may only take a fraction of that time and cost, but it just might save your bacon!