Lesson 5 (Optional): Do you want to review basic method evaluation practices?

QC planning requires estimates of the imprecision (random error, standard deviation, coefficient of variation) and inaccuracy (systematic error, bias) of a method. These estimates are initially obtained from method evaluation studies. Resources and materials for method evaluation are provided in this lesson. Internet calcluation tools are provided in the data analysis tool kit. (Revised 11/99).
 


View Outline    View Syllabus   View Archives   View Reference List   View Glossary

Objectives:

Review the concepts of precision and accuracy. 
Review the meaning of random error, systematic error, and total error. 
Review the experimental design for a method validation study. 
Review the application of statistics to estimate analytical errors from 
experimental data. 
Review the decision making process for judging the acceptability of 
performance of an analytical method. 

Web materials:

MV - The Real World Applications. 
MV - The Inner, Hidden, Deeper Secret Meaning. 
MV - The Experimental Plan.   
MV - The Data Analysis Tool Kit.
Use and interpretation of common statistics in method-comparison studies. PDF files with permission of Clinical Chemistry journal. 
A method evaluation decision chart (MEDx chart) for judging method performance. PDF files with permission from Clinical Laboratory Science journal 

Things to do:

Study the materials. 
Review your laboratory's guidelines for method evaluation studies. 
Review a method evaluation article in the literature. 

Assessment: View the Answers

Explain what is meant by random error, systematic error, and total error. 
What experiments are used to estimate the different types of analytical errors? 
How do you perform a replication study? 
How do you perform a comparison of methods study? 
What statistics are used to summarize the results of a replication study? 
How do you calculate "bias"? 
What statistics are most useful for summarizing the results of a comparison of methods study? 
Given a bias of 6 mg/dl (r=0.99) and a CV of 3.0% for a glucose method, what's your judgment on the acceptability of the method? 

Return to the top 
See the Participant Q & A 


Questions, comments,  suggestions, and feedback are encouraged!
Use feedback form on this website Or e-mail westgard@westgard.com.

This website is best viewed by Internet Explorer or Netscape 2.0 or better.

Copyright © 1998. All rights reserved.
WesTgard® Quality Corporation, 7614 Gray Fox Trail, Madison, WI 53717
Call 608-833-4718 or e-mail westgard@westgard.com

A Message from JOW
QC Lessons | QC Applications | Questions | Multirule
CLIA Requirements | What's New? | Catalog | Demo Download
Home  | Glossary | ARCHIVES| Links | Feedback