Course

Measuring error without error

It is imperative for all measurement instruments to be reliable (provide consistent measurements) and valid (measure what it is meant to measure). Lack of reliability and validity can lead to incorrect conclusions, not replicable research findings and the impossibility, for everyday users of the measurement instrument, to assess changes.

This course aims to equip researchers with the statistical knowledge necessary to plan and perform reliability, agreement and validity studies of measurement instruments providing quantitative or qualitative information.

At the end of the course, participants should be able to:

  • Make the distinction between the concepts of reliability, agreement and validity. These words are too often used interchangeably in the literature leading to confusion and inappropriate statistical analyses;
  • Plan a reliability, agreement or validity study given practical constraints (e.g. number of subjects), particularities of the measurement instrument and study aims;
  • Determine which statistical measure(s) is (are) the most appropriate depending on the aim of the study, the study design and the level of measurement of the outcome;
  • Interpret the results of the study, communicate the results to an audience with minimum statistical knowledge, and acknowledge the limitations of the statistical approach chosen.

SPSS and R will be used as statistical software.

Instructor

Sophie Vanbelle
Department of Methodology & Statistics (DEB1)
Phone: +31 43 388 22 80

Target group

PhD candidates from the Faculty of Health, Medicine and Life Sciences

Required prior knowledge

A prerequisite for participating in this course is that you have successfully completed the course Introduction to Statistics Part 1.

Content

  • Concepts reliability, agreement and validity
  • Common study designs
  • Reliability, agreement, validity on quantitative scales (graphical methods, unscaled and scaled coefficients)
  • Reliability and agreement on qualitative scales (graphical methods, unscaled and scaled coefficients)
  • Methods to determine the minimal number of observers/devices, number of replicates and number of subjects needed in a reliability/agreement/validity study
  • Guidelines for reporting reliability, agreement and validity studies

Meetings

4 sessions, of 2 hours each.

Course fees

PhD candidates (Promovendi) of FHML, MaCSBIO, M4I and MERLN: no fee

These courses are free of charge in case you are employed or registered as FHML PhD candidate. 
Master students: no fee (*)
Others: €400,00

(*) PhD candidates and other participants are given preference. If some spots are still available, then master's students can apply.

Course dates

0830-Reliability, agreement and validity – Sophie Vanbelle (MAX 25 personen)
 

Days Dates Time Location
Thursday 09-11-2023 8.30-10.30hrs. UNS50 - K4.403, Wijnand Wijnenzaal
Thursday 16-11-2023 8.30-10.30hrs. UNS50 - K4.403, Wijnand Wijnenzaal
Thursday 23-11-2023 8.30-10.30hrs. UNS50 - K4.403, Wijnand Wijnenzaal
Thursday 30-11-2023 8.30-10.30hrs. UNS50 - K4.403, Wijnand Wijnenzaal

Information

PhD secretary
Available Monday until Thursday: 9 am – 5 pm
Friday morning: 9 am – 1 pm

+31 43 387 28 44
Visitors address: Fac.Bur. FHML, P.Debeyelaan 15/ Dr. Van Kleeftoren, 2N2.004
aioonderwijs@maastrichtuniversity.nl