Serving Duplicates in a Single Session Can Selectively Improve Sensitivity of Duplicated Intensity Ranking Tests

Area of Science:

• Sensory Science
• Food Science
• Experimental Design

Background:

• Duplicating samples in ranking tests can increase statistical power and reduce panelist numbers.
• Two protocols exist for serving duplicates: one serving session (1SS) or two serving sessions (2SS).
• The impact of these serving protocols on difference detection in duplicated ranking tests is not well understood.

Purpose of the Study:

• To compare the effectiveness of 1SS versus 2SS protocols for duplicated sensory ranking tests.
• To investigate how sample similarity and sensory attribute influence protocol performance.
• To determine optimal serving strategies for duplicated ranking tests to maximize statistical power.

Main Methods:

• Seventy-five panelists performed duplicated ranking tests on orange juice samples (k=3) with varying similarity.
• Tests included sweetness and yellow color intensity, assessed separately under both 1SS and 2SS protocols.
• Mack-Skillings (M-S) statistics and rank sum differences were used to evaluate the magnitude of differences.

Main Results:

• For highly similar samples, the 2SS protocol improved yellow color intensity differentiation.
• For sweetness, the 1SS protocol yielded higher M-S statistics and rank sum differences, indicating better discrimination.
• With highly dissimilar samples, both 1SS and 2SS protocols showed comparable performance.

Conclusions:

• Serving duplicates in a single session (1SS) can enhance the power of duplicated ranking tests.
• The benefit of 1SS depends on task difficulty and the specific sensory attribute being evaluated.
• Consider 1SS for duplicated ranking tests to improve statistical power, especially when samples are similar or the attribute is challenging.