Overview
SIST EN ISO 5495:2007/A1:2016 (CEN) is an amendment to the paired comparison test methods used in sensory analysis. This document updates the European adoption of ISO 5495 and clarifies procedures for analysis and interpretation of paired-comparison results. The amendment provides practical examples showing how to compute proportions, standard errors and confidence limits for one-sided and two-sided tests, and it highlights conventions used in similarity testing.
Keywords: paired comparison, sensory analysis, confidence limits, proportion of distinguishers, similarity test, EN ISO 5495.
Key Topics
- Scope and status: Amendment A1 modifies EN ISO 5495:2007 and was approved by CEN in 2016.
- Analysis & interpretation: The amendment replaces Annex B.5.2 and provides worked examples (for instance n = 30, x = 21; n = 44, x = 32; n = 120, x = 67) illustrating how to calculate the sample proportion, standard error and one- or two-sided confidence limits.
- One-sided and two-sided tests: Examples show how to interpret one-sided lower confidence limits (e.g., concluding a minimum difference) and two-sided intervals (estimating a range for the proportion of distinguishers).
- Similarity testing convention: For two-sided paired similarity tests, x is chosen as the maximum of the two choice counts regardless of which sample was chosen most frequently; only the upper limit may be required for comparing to a pre-established critical proportion.
Applications
This standard is directly applicable to food product developers, sensory laboratories, quality assurance teams and researchers who need robust, standardized methods to:
- Detect and quantify sensory differences between product formulations or processing variants.
- Assess whether two products are sufficiently similar for interchangeability or acceptability decisions.
- Estimate the proportion of a target population able to distinguish samples, with stated confidence.
Practical value:
- Enables consistent reporting of paired-comparison outcomes across laboratories and jurisdictions.
- Supports decision thresholds (e.g., critical p for similarity) with formal confidence limits.
- Provides worked examples that guide analysts in computing and interpreting results.
Related Standards
- ISO 5495:2005 - Sensory analysis - Methodology - Paired comparison test (original international text).
- EN ISO 5495:2007 - European adoption prior to Amendment A1.
- ISO/TC 34 (Food products), Subcommittee SC 12 (Sensory analysis) - technical committee responsible for the standard.
For implementation, follow the procedures and example calculations provided in the amendment when designing tests, computing confidence limits and making similarity or difference decisions in sensory studies.