作者: JIAN BI
DOI: 10.1111/J.1745-459X.2011.00331.X
关键词: Statistics 、 Binomial distribution 、 Similarity (network science) 、 Monte Carlo method 、 Mathematics 、 Two-alternative forced choice 、 Sample size determination 、 Limit (mathematics) 、 Measure (mathematics) 、 Continuity correction 、 Food science 、 Sensory Systems
摘要: This paper discusses similarity tests using forced-choice methods in terms of Thurstonian discriminal distance, d′, a measure pure sensory difference or similarity. Critical values are derived for the 2-alternative forced choice (AFC), duo-trio, 3-AFC and triangle d′. Power analysis is conducted four methods. Three approaches used to estimate sample sizes tests. The based, respectively, on exact binomial distribution, normal approximation with continuity correction Monte Carlo simulation. S-Plus R codes were developed provided calculation estimation critical values, testing powers tests. PRACTICAL APPLICATIONS Similarity has wide application perspective consumer research. It should be when rather than interest. discussed can infer if difference, between, e.g., prototype control, smaller limit, i.e., specified allowed difference.