Assessment does not always happen once, in one form, under one set of conditions.
In practice, assessments are often adapted, repeated, scored by different people, or administered in more than one version. That means reliability cannot be reduced to a single internal-consistency estimate. Serious assessment work requires understanding how consistency is evaluated when conditions vary.
This session examines four highly relevant forms of reliability: parallel forms reliability, split-half reliability, inter-rater reliability, and test-retest reliability. The emphasis is practical. You will learn what each approach is for, when it becomes important, and how it supports real-world assessment quality.
Core reliability types used across real testing conditions
Parallel forms reliability
Understand how reliability is examined when more than one version of a test is used and why equivalence across forms matters.
Split-half reliability
Learn how internal score consistency can be examined by comparing halves of a test and what this reveals in practice.
Inter-rater reliability
Examine reliability where human judgement is involved, including interviews, ratings, scoring rubrics, and behavioural assessments.
Test-retest reliability
Understand how reliability across time is assessed and what score stability or instability may imply in applied settings.
A more complete view of reliability makes for better assessment judgement
When professionals understand how reliability behaves across forms, scorers, and repeated administrations, they become better equipped to evaluate quality, spot weaknesses, and interpret results responsibly.
That is what this session is built to strengthen: practical understanding, not abstract familiarity.
