Lakhasly

Sensory testing often requires iterative problem clarification and objective definition before acceptable test design emerges. These tests are costly and may yield uninterpretable results. Choosing a sensory test depends on the project and test objectives. Four main types exist: difference tests (detecting sensory differences), attribute difference tests (comparing specific attributes), affective tests (assessing preference and acceptability), and descriptive tests (rating attributes). Discrimination tests, including triangle, duo-trio, two-out-of-five, paired comparisons, pair-wise ranking, and difference from control, identify product differences. Descriptive tests, like quantitative descriptive analysis, flavor profile, and texture profile methods, characterize product attributes, requiring fewer, well-trained panelists and careful language selection. Affective tests gauge consumer responses to products or attributes using consumer panelists, measuring acceptability and preference. Qualitative affective tests gather direct consumer feedback. Attribute difference and similarity tests include paired comparison, pair-wise ranking, multiple paired comparison, simple ranking, and rating methods. Pair-wise ranking ranks samples by attribute intensity. Measuring sensory responses uses methods of increasing complexity: classification (grouping items nominally), grading (expert-based scaling), ranking (ordinal scaling of intensity), and scaling (numerical or verbal intensity assessment using line scales or magnitude estimation). Scale methods assess liking on a numerical scale.

 In sensory testing, a given problem frequently requires decide a practical tests

 The initial conception of the problem may require clarification.
 Test objectives to be defined and redefined several times before an acceptable design
emerges.
 Sensory tests are expensive, and they often give results that cannot be interpreted.
Guidelines for Choice
of Technique
 Define the Project Objective
 Define the Test Objective
Four tables are available for this purpose:
 Difference tests: Does a sensory difference exist between samples?
 Attribute difference tests: How does attribute X differ between samples?
 Affective tests: Which sample is preferred? How acceptable is sample X?
 Descriptive tests: Rate each of the attributes listed in the score sheet.
Guidelines for Choice of Technique
Choosing sensory tests based on objectives

Guidelines for Choice of
Technique
 Discrimination test
(e.g. Do two or more products differ from each other?)
 Descriptive analysis
(e.g. What are the characteristics of two or more products?).
 Discrimination tests for distinguishing differences in
products
 Triangle test
 Two-out-of-five
 Duo-trio,
 Difference from control
among other difference and similarity tests.
Discrimination Test

1. Triangle test. This test is a simple test to distinguish differences in
   products.
    In this test, three samples are presented simultaneously. Two of the
   samples are the same, representing a single lot and the third
   represents another lot and might be different.
    Companies may wish to test products made in different processing
   facilities (or a cost-reduced product versus original formulation) to see if
   any differences are realized.


2. Duo-trio tests.
    This test one of the identical samples is identified as the
   “control,” and the panelist must choose the unidentified
   sample that is different from the control.
    The control can remain constant in the test, or can be
   alternated between the two samples.


3. Two-out-of-five test.
    Like the other discriminative tests, the two-out of-five test is
   used to identify differences in samples. Panelists are given
   five samples and asked to identify the two that are
   different from the other three.
    five samples are used, the panelists have less opportunity to
   guess the correct samples by chance.


4. Paired comparisons for attributes.
    Two samples are presented, and the panelist chooses one of
   the samples on the basis of the specified characteristic or
   attribute.
    An example of a question, Which of the samples is sweeter?
   Panelists would usually taste several pairs of samples.


5. Pair-wise ranking test for attributes.
    The ranking of several samples that are grouped into
   pairs to test differences of a single attribute
    Example of sweetness: Samples are presented in randomized
   pairs (using every possible combination on each panelist).


6. Difference from control tests.
    This test combines some aspects of difference testing
   with some scaling.
    Subjects are asked to rate the size of difference between
   each sample and a control on a scale.
    Some test samples can be the same as the control.
    description of the characteristics of products.
    Such tests typically require fewer panelists and more training.
    Training panelists is vital to descriptive analysis along with
   determining scales.
    The use of language is important to quality attribute analysis.
   Descriptive Tests
   Quantitative Descriptive Analysis. This is a descriptive
   analysis technique that uses a panel of 10–12 highly trained
   participants to characterize and compare products.
    Flavor Profile Method. Trained panelists can be used to
   determine the intensity of attributes.
    Texture profile method.
   Paired preference test.
    Samples are offered simultaneously and asked which is
   preferred.
    Participants are asked to choose one of the samples over
   another.
    Multiple samples should be randomized and given unique
   three-digit codes
   Quantitative Affective Tests
    Affective tests: assess the personal response by potential consumers to a
   product, a product idea, or specific product attributes.
    Consumers are used for these tests.
    Affective tests of acceptability and preference measure subjective attitudes
   about a product.
   Affective Test
    Qualitative affective consumer tests aim to collect information from
   consumers regarding their acceptability of products or product
   concepts.
    Qualitative affective tests are used to gather data directly from
   consumers.
   Qualitative Test
    Attribute difference and similarity tests include
    Paired comparison
    Pair-wise ranking
    Multiple paired comparison
    Simple ranking
    Rating of several samples.
    Pairwise ranking test: Used to rank three to six samples
   according to intensity of one
    attribute; paired ranking is simple to perform and the
   statistical analysis is uncomplicated.
   Pair-wise ranking test
   Measuring Response
    The most frequently used methods of measuring sensory
   response to a sample are, in order of increasing complexity
    Classification
    Grading
    Ranking
    Scaling
   Measuring response
    Classification: The items evaluated are sorted into groups
   that differ in a nominal manner
   for example, marbles sorted by color
   Measuring response
   Grading: Time-honored methods used in commerce, which depend on expert
   graders who learn their craft from other graders
    Grading is a method of evaluation, used frequently in commerce, that depends
   on expert “graders” who learn the scale used from other graders.
    Scales usually have four or five steps such as “choice,” “extra,” “regular,” and
   “reject.”
    Examples of items subjected to sensory grading are coffee, tea, spices, butter,
   fish, and meat

Measuring response
 Ranking: The samples (usually from three to seven) are arranged in order of
intensity or degree of some specified attribute; the scale used is ordinal.
 For example, four samples of yogurt are to be ranked for degree of sensory
acidity, or five samples of breakfast cereal may be ranked for preference.
 Ranking tests have wide application, but with sample sets above three, they do
not discriminate as well as tests based on the use of scales.
 Scaling: The subjects judge the sample by reference to a scale of
numbers (often from zero to ten) that they have been trained to use.
 line scales usually yield interval data; and magnitude estimation,
although designed to yield ratio data, in practice seems to produce mixed
interval/ratio data
 techniques involve the use of numbers or words to express the intensity
of a perceived attribute (sweetness, hardness, smoothness) or a reaction
to such attribute (e.g., too soft, just right, too hard).
Measuring response
Measuring response
Scale method for measuring the level of liking
for foods. Samples are presented in succession,
and the subject is asked to indicate how much he
likes or dislikes each one on a scale.