Scribd
Upload
282 views

FST568 - Difference Tests

This document provides an overview of common sensory evaluation methods used to analyze differences and attributes of food samples, including difference tests, attribute tests, and ranking tests. Difference tests like triangle, duo-trio, and two-out-of-five tests are used to determine if an overall difference exists between samples without identifying specific attributes. Attribute tests like directional paired comparison and multiple comparison tests identify how attributes like sweetness or bitterness differ between two or more samples. Ranking tests require panelists to rank multiple samples based on attribute intensity. The document explains the objectives, principles, procedures, and analysis of these key sensory evaluation methods.

Uploaded by

Syahirah Fazial
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
282 views

FST568 - Difference Tests

This document provides an overview of common sensory evaluation methods used to analyze differences and attributes of food samples, including difference tests, attribute tests, and ranking tests. Difference tests like triangle, duo-trio, and two-out-of-five tests are used to determine if an overall difference exists between samples without identifying specific attributes. Attribute tests like directional paired comparison and multiple comparison tests identify how attributes like sweetness or bitterness differ between two or more samples. Ranking tests require panelists to rank multiple samples based on attribute intensity. The document explains the objectives, principles, procedures, and analysis of these key sensory evaluation methods.

Uploaded by

Syahirah Fazial
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 32

Sensory

Evaluation
Methodology
Getting started
What do I want to know/measure?
Which sensory test should I use?
Which subjects should I select?
How do I interpret the data?
Methods
DIFFERENCE TESTS
Introduction
Used to determine whether a difference(s) exists between samples, do
not allow personal likes/dislikes to influence judgement
◦ Formulate product with different ingredient expensive vs cheap
vanilla flavor
◦ Change in processing/packaging/storage condition
◦ Reformulate new product or improved version
◦ Product matching
◦ Quality control of finished product
◦ Investigate presence of off-flavour and taints over shelf-life
◦ Selection, training and monitoring panelist performance or sensitivity
towards stimulus
◦ Preliminary testing prior to QDA test

Difference must be subtle


Introduction
Test for similarity if you wish to:
– replace an ingredient and maintain consistency in the product

Test for difference if you wish to:


– compare your product with that of a competitor
– improve a formulation
Introduction
Divide into two groups
Overall difference test – Are there sensory differences between the samples?
◦ Triangle test
◦ Duo-trio test
◦ Two-out-of-Five test
◦ Simple difference test
◦ A-not A test

Attribute/descriptor difference test – How does attribute X differ between


samples?
◦ Directional paired comparison test (2-AFC)
◦ Multiple comparison test
Steps in conducting a
discrimination test
Triangle test
Objective - to determine whether
a sensory differences exists
between two products
Determine whether an overall
difference exists, no specific
attributes can be identified as
having been affected
Effective in situations
◦ Change in product ingredients,
processing, packaging or storage
◦ Select and monitor panellists for
ability to discriminate given
differences
Principle of triangle test
Panelists receive three coded
samples. Two of the samples are
the same (ie. duplicate samples)
and one is different (odd sample).
The panelist is asked to taste each
product from left to right and
identify the odd sample. Count the
number of correct replies and
refer to statistical table for
interpretation
Subjects and procedure of
triangle test
Involve 20-40 subjects, similarity testing also can be conducted but
need more panellists
Offer sample simultaneously
Prepare equal number of the six combinations and present at random
◦ ABB, BAA, AAB, BBA, ABA, BAB

The chance probability associated with this test is only 1/3


Advantages of test:
◦ Researcher receives information on size difference between samples
◦ Researcher also receives information on product acceptance (although
results need to be confirmed by a preference test)
Sensory ballot form/
Score sheet
Master Sheet
Duo-trio test
Objective and applications similar to triangle test
Minimum 16 panellists but discrimination improved >30 panellists
Two forms
◦ Constant reference mode – RA AB/RA BA or RB AB/RB BA
◦ Balanced reference mode – RA AB/RA BA/RB BA/RB AB

Principle - present identified reference sample, followed by two coded sample


one of which matches the reference sample. Subject indicates which coded
sample matches the reference
The chance probability associated with this test is only 1/2
Advantages of test:
◦ Simple and easily understood
◦ Reference sample is presented to avoid confusion
Sensory ballot form/
Score sheet
Master sheet: Constant ref
Master sheet: Balance ref
Two-out-of-five test
Common for visual, auditory and tactile application rather than flavour
testing
Chances of correctly guessing are 1 in 10
Similar to triangle test objective but small number of subjects is
available to involve
Principle – present five coded sample. Instruct panellist that two
samples belong to one type and three to another
Involved 30 panellists
Possibility of sensory fatigue, works well with tactile or visual test
Master sheet & Score sheet
Simple difference test
Similar to triangle & duo-trio tests but not often used
Wants to determine whether 2 samples differ without specifying the
attribute(s).
Test is one-tailed test, since the experimenter knows the correct answer
to the questions asked of each of the panelist
Normally used when product has lingering effect, short supply of
product, presentation of simultaneously three products not possible
Master sheet & Score sheet
“A”- “Not A” test
Is a sequential paired difference test or simple difference test

Panelist receives & evaluates the 1st (A) sample; that sample is then
removed

Panelist is asked to indicate whether the 2 samples are perceived to be


the same or different

Since panelists do not have the samples available simultaneously, they


must mentally compare the 2 samples & decide whether they are
similar or different

Panelists must be trained to understand the task described by the score


sheet
Score sheet & Master sheet
Directional paired comparison
test
Used when experimenter wants to determine whether two samples
differ in a specified characteristic such as sweetness, yellowness and
crispiness.
aka Two-Alternative Forced Choice
Sample present simultaneously
Simplest method and most used sensory test
Example of test objective
- to decide which of the two flavour systems tastes more like fresh-
squeezed lemon
- To confirm which coffee sample is more bitter than the other
Master sheet & Score sheet
Multiple paired comparisons
test
Aka pairwise ranking test or n-AFC
More than two samples to be evaluated
Each sample must be compare with other samples for single attribute
Normally three to six samples
Number of pairs is determined by the formula ½ n (n-1)
If four samples of corn syrups need to be evaluated, therefore six
possible pairs should be presented in balanced random order to the
panellists
Ranking test
Panellist receive three or more coded samples and has to rank them for
the intensity of specific characteristics
Less time-consuming and testing several samples at once
Useful when samples are to be pre-sorted or screen for later analysis
No indication of size of differences between samples
Result of one rank cannot be compare the results of another set of
ranks
Master sheet & Score sheet

You might also like