What does Ice, Aroma and Blind Tests in Frozen Fatty Pulp conclude, in numbers, about “Designing Blind Tests of Frozen against Fresh: Replication, Position Bias and the Burden of Proof for "Tastes the Same"”?

Answer: Chapter 6 covers overdispersion in repeated tasting, the similarity test of at least 39 people needed to claim "you can't tell", and false-positive rates as high as 0.3302 under unbalanced serving orders.

Scope of the numbers: Its parameters are illustrative; all calculations are generated by its programs; the book is not medical or nutritional advice, claims no health benefit for any food, and its pricing formula is not personalised financial or business advice. Marked boxes separate published evidence, runnable Python calculations (21 programs), classical sources and the author's hypotheses.

中文版

Data table

The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python onlyprinted values that match the answer: 39

pd0similarity nsimilarity ndifference n
(true pd 0)(true pd 0.05)(true pd = pd0)
0.103251322325
0.208615487
0.30395840
0.40252823
Show the full program
# Chapter 6. p62: how many tasters it takes to claim "frozen tastes the
# same as fresh" with a triangle test run for similarity.
# H0: pd >= pd0 (a meaningful share can tell them apart).  Claim
# similarity when X <= c, with P(X <= c | Pc0) <= alpha, Pc0 = 1/3 +
# 2/3 pd0 (ISO 4120 similarity form).  Power when nobody can tell
# (pd = 0) or when pd = 0.05.  The difference test (p53 logic) is shown
# for the same pd0 as the alternative.  alpha = 0.05, power 0.80;
# smallest n with power >= 0.80 is reported.
import math

ALPHA, TARGET, NMAX = 0.05, 0.80, 2500


def pmf(n, p):
    q = 1.0 - p
    out = [q ** n]
    for k in range(n):
        out.append(out[-1] * (n - k) / (k + 1) * p / q)
    return out


def sim_power(n, pc0, pc_true):
    probs0 = pmf(n, pc0)
    acc, c = 0.0, -1
    while c + 1 <= n and acc + probs0[c + 1] <= ALPHA:
        acc += probs0[c + 1]
        c += 1
    if c < 0:
        return 0.0
    return math.fsum(pmf(n, pc_true)[:c + 1])


def diff_power(n, pc1):
    probs0 = pmf(n, 1.0 / 3.0)
    acc, x = 0.0, n + 1
    while x > 0 and acc + probs0[x - 1] <= ALPHA:
        acc += probs0[x - 1]
        x -= 1
    return math.fsum(pmf(n, pc1)[x:])


def first_n(fn, lo=6, step=1):
    n = lo
    while n <= NMAX and fn(n) < TARGET:
        n += step
    return n


print("pd0    similarity n   similarity n   difference n")
print("       (true pd 0)    (true pd 0.05)  (true pd = pd0)")
for pd0 in (0.10, 0.20, 0.30, 0.40):
    pc0 = 1.0 / 3.0 + 2.0 / 3.0 * pd0
    a = first_n(lambda n: sim_power(n, pc0, 1.0 / 3.0))
    b = first_n(lambda n: sim_power(n, pc0, 1.0 / 3.0 + 0.1 / 3.0))
    c = first_n(lambda n: diff_power(n, pc0))
    print("{:4.2f}   {:12d}   {:12d}   {:13d}".format(pd0, a, b, c))

Single source and citation

Fan, Y. (2026). Ostwald Ripening in Liquid-Nitrogen-Frozen High-Fat Fruit Pulp, Volatile Thiol Release Kinetics and Triangle-Test Sensory Statistics (Version 1.0.0), Chapter 6: Designing Blind Tests of Frozen against Fresh: Replication, Position Bias and the Burden of Proof for "Tastes the Same". 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-169 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026169,
  author    = {Fan, Yafu},
  title     = {Ostwald Ripening in Liquid-Nitrogen-Frozen High-Fat Fruit Pulp, Volatile Thiol Release Kinetics and Triangle-Test Sensory Statistics},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book