Compare courses

ICTD International Centre for Training and Development

Best Practice of Statistics Method and Measurement Uncertainty for Laboratory

Available dates

This course has no confirmed dates in the future. Subscribe to be notified when it is offered.

About the course

This course aims to show how this can be achieved in a practical and cost effective manner and how it can be used to increase the overall confidence in the measurements made.

ISO17025 also requires laboratories to produce uncertainty budgets for each accredited test. The workshop aims to dispel some of the myths of uncertainties and show, practically, how relevant uncertainty budgets can be calculated.

Course Objectives

Upon successful completion of this course, the delegates will be able to:

  • Concepts involved in the calculation of measurement uncertainty
  • Calculating measurement uncertainty in a practical and pragmatic manner
  • Defining measurement processes
  • Identifying sources of measurement error
  • Selecting appropriate error distributions
  • Using different methods to evaluate measurement uncertainty
  • Measurement uncertainty by practice methods

Course Outline

Day 1

  • Instrument analysis data
  • Peak evaluation
  • Interpolated graph calibration using external/ internal standards
  • Standard addition method extrapolated graph
  • Errors in quantitative analysis
  • Random and systematic errors in titration analysis
  • Standard deviation of repeated measurements

Day 2

  • Distribution of errors
  • Confidence limit of the mean of replicate measurements
  • Measurement uncertainty
  • Errors in instrumental analysis regression and correlation
  • Use of regression lines for comparing analytical methods
  • Confidence limit for X-value
  • Outliers in regression

Day 3

  • Limit of detection
  • Significance tests for evaluation of experimental results
  • (T-test) comparison of a mean with a known value
  • (T-test) comparison of the means of two samples with S1»S2
  • (T-test) comparison of the means of two samples with S1 ¹ S2
  • Paired T-test and One-tailed and Two-tailed tests
  • (F-test) for the comparison of standard deviations
  • Anova-test analysis of several means and variances

Day 4

  • Anova-test analysis of several means and variances
  • Testing for normality of distribution
  • Outliers test
  • Non-parametric or distribution-free methods
  • Box and whisker plots
  • Comparison of a median with a known value (the sign test)
  • Confidence interval for non-parametric methods
  • Comparison of the medians of two methods (the sign test)
  • Comparison median of two un-depended samples (Wilcoxon Rank-Sum est)
  • Comparison spread of two sets of non-parametric results (Siegel-Tukey test)

Day 5

  • Rank Correlation for not quantified results (spearman method)
  • Non-parametric method on more than two samples (Friedman´s test)
  • Non-parametric regression methods (Theil´s test) quality control harts
  • Quality control charts
  • Shewhart and Cusum Chart
  • Experimental design and optimization methods
  • Factorial designs
  • Estimation of factors interaction by two-way Anova test
  • Optimization method and three factors design

Course Methodology

A variety of methodologies will be used during the course that includes:

  • (30%) Based on Case Studies
  • (30%) Techniques
  • (30%) Role Play
  • (10%) Concepts
  • Pre-test and Post-test
  • Variety of Learning Methods
  • Lectures
  • Case Studies and Self Questionaires
  • Group Work
  • Discussion
  • Presentation

Who should attend

This course is intended for all chemists, laboratory technicians, chemical engineers, scientists, laboratory managers, R&D, instrument engineers and all laboratory professionals.

Course reviews

Downloadable files