GIFSQ Tools · Module 12 — Sensory · Tool 226 of 500
Analyse a paired comparison test: the exact one-sided binomial p-value for a stated preference against the 1-in-2 chance split, and the minimum count for significance at α = 0.05.
Try sample values — fills the form with the worked example below.
–
p-value
–
Minimum for A (α = 0.05)
–
Preferring A (%)
Equation: p = Σ C(n,k)(1/2)^k(1/2)^(n−k) for k = preferring A … n; critical value = smallest count with p ≤ 0.05.
Method: ISO 5495:2005 paired comparison test; exact one-sided binomial test with p = 1/2.
Assumptions: Each assessor chooses one of the two samples; no ties are allowed; responses are independent.
Limitations: One-sided test for a stated direction of preference; use a two-sided reading when no direction was stated in advance.
Rounding: p-value to 4 decimal places; percentages to 2 decimal places; intermediates unrounded.
This is a calculation aid, not a laboratory report — confirm against your accredited method before citing a value.
Method: ISO 5495:2005 paired comparison test; exact one-sided binomial test with p = 1/2.
Assumptions: Each assessor chooses one of the two samples; no ties are allowed; responses are independent.
Limitations: One-sided test for a stated direction of preference; use a two-sided reading when no direction was stated in advance.
Rounding: p-value to 4 decimal places; percentages to 2 decimal places; intermediates unrounded.
This is a calculation aid, not a laboratory report — confirm against your accredited method before citing a value.
Worked example: 30 assessors, 21 preferring A: p = 0.0214, minimum for A 20, 70.00% preferring A — significant at α = 0.05.
Enter values above to see results.
Tool 226 of 500 — GIFSQ
